ISO SWS-LWS observations of the prototypical reflection nebula NGC 7023

合集下载

The effect of soil water content, soil temperature, soil pH-value and the root mass on

Plant and Soil(2005)268:21–33©Springer2005 DOI10.1007/s11104-005-0175-5The effect of soil water content,soil temperature,soil pH-value and the root mass on soil CO2efﬂux–A modiﬁed modelSascha Reth1,3,Markus Reichstein2&Eva Falge11Department of Plant Ecology,University of Bayreuth,D-95440Bayreuth,Germany.2Potsdam Institute for Climate Impact Research Natural Systems Department,D-14473Potsdam,Germany.3Corresponding author∗Received10February2004.Accepted in revised form21April2004Key words:ﬁne-root mass,non-linear regression model,pH respiration,soil moisture,soil temperatureAbstractTo quantify the effects of soil temperature(T soil),and relative soil water content(RSWC)on soil respiration we measured CO2soil efﬂux with a closed dynamic chamber in situ in theﬁeld and from soil cores in a controlled climate chamber experiment.Additionally we analysed the effect of soil acidity andﬁne root mass in theﬁeld.The analysis was performed on three meadow,two bare fallow and one forest sites.The inﬂuence of soil temperature on CO2emissions was highly signiﬁcant with all land-use types,except for oneﬁeld campaign with continuous rain. Where soil temperature had a signiﬁcant inﬂuence,the percentage of variance explained by soil temperature varied from site to site from13–46%in theﬁeld and35–66%in the climate chamber.Changes of soil moisture inﬂuenced only the CO2efﬂux on meadow soils inﬁeld and climate chamber(14–34%explained variance),whereas on the bare soil and the forest soil there was no visible effect.The spatial variation of soil CO2emission in theﬁeld correlated signiﬁcantly with the soil pH andﬁne root mass,explaining up to24%and31%of the variability.A non-linear regression model was developed to describe soil CO2efﬂux as a function of soil temperature,soil moisture,pH-value and root mass.With the model we could explain60%of the variability in soil CO2emission of all individualﬁeld chamber measurements.Through the model analysis we highlight the temporal inﬂuence of rain events.The model overestimated the observedﬂuxes during and within four hours of the last rain event. Conversely,after more than72h without rain the model underestimated theﬂuxes.Between four and72h after rainfall,the regression model of soil CO2emission explained up to91%of the variance.IntroductionSoil CO2ﬂuxes are the second major component of the global carbon cycle(Reich and Schlesinger1992), and play an important role in climate change.Very often is it hypothesized that soils provide a positive feedback to climate warming due to the exponential response of soil CO2efﬂux to temperature(e.g.Cox et al.,2000;Kirschbaum,1995).However,the gas ex-change between the soil and the atmosphere depends on numerous complex and non-linear relationships, like physiological,biochemical,chemical,ecological and meteorological conditions(Jarvis,1995;Schimel et al.,1994).Soil respiration represents the biolog-ical activity of the entire soil biota,including soil ∗FAX No:+490921552061microbes(e.g.bacteria,fungi,algae,protozoa),plant roots and macroorganisms(e.g.earthworms,nema-todes,insects).The rates of soil CO2efﬂux vary by ecosystem(Reich and Schlesinger,1992)and are the major component of whole-ecosystem respiration,that in turn explains much of the continental gradient of the net carbon balance(Schulze et al.,1999;Valen-tini et al.,2000).So show Kelliher et al.(1999)and Law et al.(1999b)for forested ecosystems,that soil respiration amounts to76–77%of the annual GPP, whereas agricultural crops during fallow periods act as a carbon emitter(Soegaard,1999;Soegaard,et al. 2003).Despite these general trends emissions of CO2 are highly spatially variable within one site(Law et al., 2001;Longdoz et al.,2000;Simek et al.,2004).22A positive correlation between soil temperature and soil CO2efﬂux is well described by several re-views(Kätterer et al.,1998;Lloyd and Taylor,1994; Reich and Schlesinger,1992;Singh and Gupta,1977). Also,soil moisture affects the soil CO2efﬂux(Bun-nell et al.,1977;Orchard and Cook,1983;Reichstein et al.,2002;Simek et al.,2004;Subke et al.,2003). Furthermore,soil CO2efﬂux is inﬂuenced by other factors like,substrate amount(Zak et al.,2000),the pH-value of the soil(Hall et al.,1997)as well as the activity of the vegetation(Reichstein et al.,2003b) since root respiration(Janssens et al.,1998;Kutsch et al.,2001;Law et al.,1999a)and heterotrophic res-piration(Goulden et al.,1996;Hollinger et al.,1998) comprise total soil CO2efﬂux,and plants continu-ously excrete exudates into the soil.Several studies showed signiﬁcant effects of soil pH values on soil respiration(Andersson and Nilsson,2001;Hall et al., 1997;Sitaula et al.,1995)since,in particular,mi-crobial activity increases with rising pH values(Ellis et al.,1998).Furtheron,temporal effects like litter fall,decom-position dynamics and the amount and the timing of rainfall(Ball et al.,1999;Jackson et al.,1998)inﬂu-ence soil respiration.The effect of rainfall was often larger than expected from the relationship between soil moisture and CO2efﬂux(Davidson et al.,1998; Russell and V oroney,1998).A series of models try to explain the relation-ship between the factors governing soil CO2ef-ﬂux.Most studies use different principles to describe temperature effects, e.g.linear regression analysis (Witkamp,1966),Q10(Maljanen et al.,2002;Reich and Schlesinger,1992)or power relationship(Kucera and Kirkham,1971),as well as relationships based on the Arrhenius form(Howard and Howard,1979). However,all existing models cannot explain the total variation of the CO2soil efﬂux.Numerous empirical models were developed for crop(Boegh et al.,1999) and meadow soils(Bremer and Ham,2002)or bare soils(Gupta et al.,1981;Reth et al.,2004).These models are not useful for forest soils.In contrast forest models(Baldocchi and Wilson,2001;Janssens et al., 2001;Nakano et al.,2004;Rasse et al.,2001)are often of low use at bare soil or meadows.The aim of this study is to analyse the inﬂuence of soil temperature,moisture,pH value and root mass on soil CO2efﬂux through a combination ofﬁeld and lab-oratory experiments.In a second step these effects are assembled into an empirical model that should work on meadow and cropland as well as in forest and bare fallow soil.Finally,we explore a robust regression method to identify temporal effects on soil CO2efﬂux in theﬁeld that are not represented by the model. MethodsSite descriptionThe measurements used in this study were carried out in the course of special observation periods of the VERTIKO(Vertical transport under complex natural conditions)project,which is part of the AFO2000 (German Atmospheric Research,2000)programme. The target area of the VERTIKO project comprises the region between the Erzgebirge in the South and the Oder-Spree lake district in the North(100km WE and 300km NS).It includes a variety of natural small-scale variability from land use to orographic effects typical for Germany.During three special observation periods (SOPs)measurements were performed at anchor sta-tions located in the target area.For an overview of the parameters observed during theﬁeld campaigns that are expected to inﬂuence soil CO2efﬂux see Table1.The measurements of SOP1were carried out in September and October2001at the Anchor Station Melpitz of the Institute for Tropospheric Research,lo-cated near Melpitz,Saxony(51◦31 N,12◦55 E,86m a.s.l.).The area is aﬂat managed meadow(MW)of approximately20ha surrounded by farmland(MA, see e.g.Spindler et al.,2001).The annual mean air temperature is8.7◦C and the annual precipitation 539mm.The dominant species were Lolium perenne, Taraxacum ofﬁcinale and Leontodon autumnalis.The leaf area index(LAI)was2.0m2m−2.The SOP2experiment took place in June and July2002at the Falkenberg Boundary-Layer mea-surement site of the German Meteorological Service, the Lindenberg observatory,Brandenburg(52◦10 N, 14◦07 E,73m a.s.l.).The landscape in this region was formed by inland glaciers of the last ice age, with a slightly undulating orography and a hetero-geneous land use structure(see e.g.Beyrich et al., 2002).The Falkenberg site itself isﬂat and consists of about18ha of managed meadow(LW)with short grass.An area of approximately3ha of the meadow was ploughed during the experiment(LA).The annual mean air temperature is8.6◦C and the annual precip-itation560mm.Main species were Lolium perenne, Bromus hordeaceus,Festuca rubra,Leontodon autum-nalis,Taraxacum ofﬁcinale,Trifolium pratense and23 Table1.Observed range of the parameters T soil(soil temperature),RSWC(relative soil watercontent),pH and RRM(ﬁne root biomass,d.w.based)expected to inﬂuence soil CO2emissionduring theﬁeld campaigns(MW=Melpitz Meadow,MA=Melpitz Agricultural Fallow,LW=Lindenberg Meadow,LA=Lindenberg Agricultural Fallow,TW=Tharandt Meadow,TF=Tharandt Forest)T soil T soil RSWC RSWC pH pH RRm RRMmin(◦C)max(◦C)min(%)max(%)min max min(%)max(%)MW10.316.6398 5.87.10.0359MA10.721.53999 6.97.400LW14.325.81738 4.5 6.90.4126LA14.518.91617 5.0 5.900TW11.520.95897 5.0 5.526TF9.018.65681 3.3 3.80.3636Trifolium repens.Meadow LAI showed a spatial gra-dient during theﬁeld campaign,with maximum LAI of4.7,and minimum LAI of1.3m2m−2.The SOP3measurements were performed in May and June2003at the Anchor Station Tharandter Wald of the Technical University Dresden near Tharandt, Saxony(50◦58 N,13◦34 E,375m a.s.l.).The slightly undulating experimental area is located inside a closed forest of approximately6000ha.The annual mean air temperature is7.6◦C and the annual precipita-tion820mm.The forest(TF)is dominated by114 years old,approximately28m high Picea abies(L.) KARST trees.The projected leaf area index(LAI) was6.9m2m−2.The meadow area(TW)of the anchor station is1.5ha and dominated by Rumex ob-tusifolium(L.),Holcus lanatus(L.),Cirsium arvense (L.)and Carex spp.The leaf area of the meadow was 2.6m2m−2at the beginning(May,22)of theﬂux measurements and increased to6.1m2m−2at the end of the campaign(June,13).Theseﬁeld measurements were complemented by climate chamber experiments to extend the range of soil temperatures and soil moisture observed during the SOPs.Soil efﬂux and soil analysis in theﬁeldSoil CO2efﬂux was measured with a non-steady-state ﬂow-through chamber system,andﬂuxes were deter-mined from the concentration increase in closed cham-bers.The system consists of cylindrical steel chambers (height80mm and diameter197mm)with plexi glass lids attached during the measurement.No fan was used in the system.Overheating could be avoided because single measurements were completed within12min,and soil temperatures inside and outside the chamber differed less than0.2◦C.The chambers were inserted 2cm into the soil and all plant material was removed from the chambers’interiors.Theﬁrstﬂux measure-ments started approximately12h after plant cutting to avoid effects on soil CO2efﬂux by collar inser-tion or plant cutting.Collars remained in place during all subsequent measurements at the site.Ten cham-bers were installed as spatial replicates at each land use type,except at the bare soil of Lindenberg with onlyﬁve chambers.Measurements took place from the early morning to late in the evening.The chambers were moved to another site or land use afterﬁnishing 5to14measurements at the same point.The sys-tem allowed to measureﬁve chambers alternately with magnetic valves controlling theﬂow of the different chambers.For the concentration measurements the air was pumped in a closed loop from the chamber to the analyser(Photoacoustic Multi-gas Monitor,INNOV A 1312)and back to the chamber.Through a20m long tube with an inside diameter of3mm,the air was sucked for approximately30s with a speed of4m s−1. The concentrations of CO2(for control)and water were determined from the air stream.The CO2efﬂux was determined from the slope of the concentration increase within a chamber using four concentrations measured at238s intervals.The system was tested against other measurement systems(non-steady-state ﬂow-through chambers,non-steady-state non-ﬂow-through chambers,non-steady-state non-ﬂow-through chambers and a calibration system)in a calibration experiment(Pumpanen et al.,2004).In this experi-ment he system employed showed an underestimation of approximately4%for soils comparable to those in this study,and at maximum11%for dryﬁne sand.24Parallel to the soilﬂux measurements environ-mental parameters were observed quasi-continuously. Soil temperature(Thermistor,Siemens M841)was recorded at2cm depth inside each chamber and out-side the chambers every5min.V olumetric soil water content(SWC,m3water per m3total soil volume, Theta Probe,ML2)was measured half hourly in the upper10cm of the soil at each stand.Relative soil wa-ter content(RSWC)is deﬁned as SWC divided byﬁeld capacity,allowing for a better comparison of soils with different textures.Reichstein et al.(2002)found very similar RSWC1/2parameters for a sandy and clayey soil,Nevertheless one would not expect exactly the same values of RSWC1/2in all soils,and our assump-tion that reduces the number of model parameters introduces(albeit little)model error.Analysis of root mass and pH of soil samples of each chamber were performed afterﬁnishingﬂux measurements.Soil cores with a diameter of5cm and a depth of10cm were taken in theﬁeld,and soil and roots were separated manually.The remaining soil was sieved to remove stones.The root biomass was dried three days at105◦C.The dry mass of the roots was expressed per unit dry mass of the oven-dry soil.The pH-value was determined from a fresh soil slurry using a glass electrode(Scheffer and Schachtschabel,2002). Incubation time of20g soil was1h in50g distilled water.Soil efﬂux in the climate chamberFor the climate chamber experimentsﬁve replicate fresh soil cores(diameter=31cm,height=25cm) were collected under minimal disturbance from the meadow of Melpitz,Lindenberg and Tharandt as well as from the fallow of Lindenberg and the forest of Tharandt.CO2efﬂux of the soil samples was recorded over112days.Soil temperature was manipulated by changing the air temperature inside the climate chamber.Starting at20◦C the soil temperature was decreased every two days by2◦C.After reaching4◦C the soil temperature was increased every two days by 2◦C up to38◦C,then decreased again and so forth. Soil water content was altered by irrigation and drying cycles.At beginning and end of eachﬂux measure-ment(for description of the system see above),we weighed the soil cores for gravimetric determination of soil water content.In the climate chamber it was not possible to analyse the soil without destroying the soil cores.Therefore the soils were not analysed for the parameters pH and RRM.Soil CO2modelThe non-linear regression model of soil CO2efﬂux was adapted from Reichstein et al.(2003b,2002)that includes a function for soil temperature(T soil,Equa-tion3)–following an exponential response,a function of relative soil water content(RSWC,Equation4) and a function of vegetation activity.We modiﬁed the model by Reichstein(2003b)for better incorporation of the actually measured data in two ways:(1)The soil CO2emission rate under standard conditions,was made dependent on root mass per soil mass as a proxy for vegetation activity(instead of leaf area index in Reichstein,et al.2003b);(2)We included the inﬂu-ence of soil chemistry through including the soil pH as additional predictor.Mathematically,the model is described by the following equations:R soil=R ref∗F(T soil)∗g(RSW C)∗h(pH),(1) where R soil is the soil CO2efﬂux.The emission under standard conditions(R ref),at T ref and non-limiting soil water content,is described by:R ref=H resp+a resp,(2) where h resp represents heterotrophic respiration and a resp autotrophic respiration.The heterotrophic respi-ration is aﬁtted parameter and a resp is a linear function of the root mass per dry soil mass(RRM)and a parameter(rf):a resp=R RM∗r f.(3) The exponential increase of the CO2emission with soil temperature is described by:f(T soil)=expE0∗1T ref−T0−1T soil−T0,(4) where E0is a free parameter analogue to the activation energy in the standard Arrhenius model,T ref is the reference soil temperature and T0the lower temper-ature limit for R soil.T ref was set to15◦C and T0at −46.02◦C(Lloyd and Taylor,1994).The response of changes on relative soil water content is described by: g(RSW C)=RSW C−RSW C0(RSW C1/2−RSW C0)+(RSW C−RSW C0).(5)RSWC1/2represents the RSWC at half-maximum soil CO2efﬂux and RSWC0is the residual soil water con-tent,below which the efﬂux ceases.RSWC1/2and RSWC0are free parameters.Finally,the response of the CO2emission with soil pH-value follows an optimum curve:25 Table2.Coefﬁcients of variation for the univariate analysis of soil CO2emission inﬂuencing parametersduring theﬁeld(FM)and the climate chamber measurements(CCM).T soil(soil temperature),RSWC(relative soil water content),pH and RRM(ﬁne root biomass,d.w.based),n represents the number ofCO2ﬂux observations not affected by rain,and used for the regression model parameterisation.For siteabbreviations see Table1FM CCMT soil RSWC RRM pH n T soil RSWCMW0+0+0+0+00.45∗0.1MA0+0+ND0+00.35∗0.09LW0.25∗0.34∗0.12∗0.19∗1410.52∗0LA0.46∗0ND0.24∗30ND NDTW0.31∗0.18∗0.3∗0.19∗500.65∗0.14∗TF0.13∗0.040.31∗0.23∗740.66∗0P<0.01,ND=not determined.+not used for model parameterisation.h(pH)=exp−pH−phOptphSens2,(6)where phOpt is a free parameter and represents the optimal pH value.The parameter phSens describes the sensitivity of soil CO2efﬂux to deviation from this optimal value.Parameter estimationFor the data analysis with the non-linear regression model we used a robust regression technique that is able to objectively identify outliers,or more pre-cisely data points,that are inconsistent with the model assumptions.We used the non-linear least trimmed squares(LTS)regression(Reichstein et al.,2003a; Stromberg,1997),that seeks to minimize the sum of squared residuals as ordinary non-linear regression, but with exclusion of the largest x%of residuals,that are assumed to be due to contaminated data or due to data inconsistent with the model.Formally the objec-tive function that has to be minimised is the trimmed sum of squared errors(TSSE):T SSE=i≤N·(1−0.001−t)r2i,(7)where r i is the i-th smallest residual,N is the to-tal number of data points,and(0.01t)is the fraction of residuals to be excluded.The procedure was per-formed with trimming percentages of10,20,30% and subsequently analysed which data was classiﬁed as‘contaminated’by the procedure.ResultsWe examined the effect of soil temperature changes on CO2efﬂux at the four soil types of theﬁeld mea-surements and ofﬁve soil types in the climate chamber experiment.Due to the continuous rain during SOP1, the results of theﬁeld measurements in Melpitz were not used in the temperature,and all further analyses. An exponential increase with increasing soil tempera-ture was observed at all soils(Figure1),both during theﬁeld measurements and in the climate chamber experiment(Table2).Meadow soils,except MW in bothﬁeld and cli-mate chamber measurements,and LW in the climate chamber measurements,responded to changes of rel-ative soil water content.There was no statistically signiﬁcant effect at the fallow and the forest soil,both in theﬁeld and in the climate chamber measurements (Table2).Variation of the pH-value of the soils and between the single measurement points at each site showed a positive correlation with the CO2efﬂux(Figure2). During simultaneous measurements with similar soil temperature and soil water content,the chambers with higher soil pH exhibited higher CO2ﬂuxes,except in Melpitz(Table2).Also the presence ofﬁne roots signiﬁcantly af-fected the observed soil CO2efﬂux.At all meadow and forest stands,again except for Melpitz,the relative root mass was correlated positively(Figure3)with the CO2ﬂux rates(Table2).While the univariate relationships between soil CO2efﬂux and environmental factors were gener-ally weak,the above soil CO2efﬂux model already26Figure1.Soil CO2efﬂux as an exponential function of soil temperature(T soil)for the meadow soil of Tharandt(TW)in theﬁeld(closed dots, r2=0.31),and in the climate chamber(open dots,r2=0.65).The lines are regression lines,calculated from the used data using equation4.Figure2.Soil CO2efﬂux as function of soil pH-value at the fallow in Lindenberg(LA).Dots represent the mean CO2ﬂuxes(n=5)with error bars(r2=0.24).The line is the regression line,calculated from the used data using equation6.27 Figure3.Soil CO2efﬂux as function of relative root mass(RMM)at the meadow soil of Tharandt(TW).Dots represent the mean CO2ﬂuxes (n=6)with standard deviation(r2=0.30).The line is the regression line,calculated from the used data using equation3.explained60%of the variability of soil CO2efﬂux (Figure4a).With the robust regression approach we analysed inconsistence of the model(Figure4b–d). 30%of the data could be identiﬁed as inconsistent with the model and could be related to disturbance by precipitation events(Figure4e).Thereby we could identify3periods:During and up to4h after a rain event(period1)the model overestimated the measured CO2ﬂuxes.In contrast, after a dry period of more than72h(period2)the model underestimated theﬂuxes.In the time period in between,that is4to72h after the last rain event (period3),the model reﬂected the measured emissions well,and explained91%of their variability(Table3). Interestingly,the amount of rain did not affect the per-formance of the model.The robust regression method rejected87%of the data points falling into period1or 3,supporting the rationale to identify and exclude data that are inconsistent with the model.DiscussionIn this paper we conﬁrmed well known correlations of soil CO2efﬂux and abiotic factors,although some-times the ranges of driving forces in theﬁeld were too small to detect previously reported effects,e.g.on Q10.The strong correlation of temperature and soil CO2 emission was quantiﬁed for many soils under different conditions(see e.g.Epron et al.,1999;Kätterer et al., 1998;Lloyd and Taylor,1994;Reich and Schlesinger, 1992).In our climate chamber measurements,all soils showed an exponential increase of soil CO2efﬂux with increasing soil temperature(P<0.01).In the ﬁeld the soil CO2efﬂux was more variable,indicating increasing inﬂuence of other parameters.A similar distinction was observed comparing soil CO2exchanges at changing soil moisture.Only at the meadow stands in Tharandt and in theﬁeld measure-ments of Lindenberg soil CO2efﬂux showed signiﬁ-cant(P<0.01)response to soil moisture changes.At the other stands,and partly during the climate chamber experiments,the relative soil water content span al-lowed only for small limiting effects due to soil water. Also,Reichstein et al.(2003b)observed a broad range28parison of modeled and measured soil CO2efﬂux:(a)without a trimmed fraction(r2=0.60),(b)with a trimmed fraction of 10%(r2=0.80),(c)with a trimmed fraction of20%(r2=0.86),(d)with a trimmed fraction of30%(r2=0.91)and(e)evaluating the temporal effect of last rain event.Lines are1:1lines.29 Table3.Parameters of the nonlinear regression model for all stands with a trimmed fraction of30%(n=295).240values were not effected by rain or drought,and indicated as consistentwith the model(LW:n=97,LA:n=30,TW:n=44,TF:n=69).Root mean squareerror(RMSE)of the model results was0.89µmol m−2s−1.To take into account the4%underestimation of the system(Pumpanen et al.,2004),the parameters h resp and rf have to bemultiplied with the correction factor0.96Parameter Units Value Standard errorh respµmol m−2s−19.11 2.12E0K247.7816.84RSWC0%92RSWC1/2%170.9phOpt9.35 1.23phSens−4.870.69rfµmol m−2s−1g d.w.19.91 5.23Soil(g d.w.root)−1of near optimum soil water content where changes in soil moisture have little or no effect and correspond to our observations.At the bare soil of SOP2at Linden-berg the soil water content was nearly constant while the measurements were performed,so there was no effect of soil moisture changes on CO2efﬂux at this time.Even when taking soil temperature and water con-tent into account,the spatial variation of soil CO2 efﬂux at one site can be large.Buchmann(2000)re-ported spatial variations among soil collars,which were larger than the diurnal variability of soil CO2 emission rates measured with the same collars during a day.This corresponds well with ourﬁeld measure-ments,in particular at the forest stand,where soil temperature changes were very small,but spatial vari-ability was high.Thus,multivariate interaction of various other factors has to be accounted for as in the model presented here.The link of respiration to vegetation productivity established by Reichstein et al.(2003b)with poten-tially confounding factors at the continental scale,was here conﬁrmed at small scale for soil CO2efﬂux.An-derson(1992)and Janssens et al.(1998)showed,that root respiration may account for half of the soil efﬂux. In general,this agreed with our observations for the forest and meadow sites.In addition,samples with higher root mass per soil showed higher CO2emission (P<0.01)at comparable meteorological conditions. Thisﬁnding held within a site,but not among differ-ent sites,where other factors determined the overall magnitude of soil CO2efﬂux.An inﬂuence of spatial heterogeneity of soil pH on soil CO2emission was conﬁrmed at all stands (P<0.01),except Melpitz.Several studies described a similar positive correlation of pH-value and soil CO2 efﬂux(Andersson and Nilsson,2001;Ellis et al.,1998; Hall et al.,1997;Sitaula et al.,1995).Baath(1996) and Högberg et al.(2003)demonstrated the direct pos-itive effect on soil respiration with pH tolerance of the bacterial community.A biological activity of soil microorganisms is permitted between a soil pH of a minimum of3and a maximum of7to8(Scheffer and Schachtschabel,2002).Between these values(see Ta-ble1),and otherwise constant conditions we observed a nearly linear increase of soil CO2emission.In the model however,we described the response to soil pH with an optimum curve to account for potential decline in soil CO2emission above a pH of9.Similar pattern were shown in Wittmann et al.(2004)with an opti-mum curve for the dependence of hydrolytic enzyme activities in a forest soil.The nonlinear regression model gave good results for all investigated sites.Up to60%of the data vari-ance could be explained by soil temperature,relative soil water content,soil pH and relative root mass. Evaluating the time span between measurement and last occurring rain,the modeled soil CO2efﬂuxes overestimated the measuredﬂuxes in the case of rain or maximum4h after the last rain.The main cause for this could be the reduction of the soil air-ﬁlled pore space resulting in reduced gaseous diffusivities. The negative effect of waterﬁlled pores on soil CO2 emission is often discussed in the literature(see e.g. Ball et al.,1999;Lee et al.,2002).30After three or more days without rain,the model underestimated the observedﬂuxes.An explanation for this might be a shift of the main respiratory ac-tivity to deeper soil layers,with soil moisture and temperatures more favourable to respiration than those recorded by the sensors in the top soil layer.Another explanation could be thatﬁne roots dying in the upper soil,and new root development in deeper soil layers led to an increase in CO2release.For the time be-tween4and72h after a rain event,the model worked well,explaining91%of the soil efﬂux variation with changes in soil temperature,soil water content,root mass and soil pH.Potential limitation of the model could be that the temperature,soil water content,and pH responses of respiration arising from roots and soil heterotrophs might differ.Root respiration depends on current pho-tosynthetic products as substrate,and is therefore mainly controlled by light availability during the last2 days(Fitter et al.,1998).Heteorotrophs use older pho-tosynthetic products(e.g.litter,turnover ofﬁne roots), but also use root exudates(Grayston et al.,1997),as rhizosphere micro-organisms rapidly acquire the iso-topic signature of the current photosynthate(Pendall et al.,2003),therefore being partly coupled to light availability too.In our case,we had to simplify these effects,as it was not possible to separate the responses of these two componentﬂuxes from data measured with our technique.We included only the relative amounts of autotrophic and heterotrophic respiration in the model,and applied identical temperature,soil water content,and pH functions.We tried to overcome the limiting effects of rel-atively short measurement campaigns at the various sites with soil cores taken to the climate chamber for wider ranges of temperature and moisture.However, this setup still did not allow for proper assessment of threshold events or sudden shifts in key variables de-termining soil respiration or soil CO2efﬂux.Yet in the ﬁeld,Jensen et al.(1996),Lee et al.(2002)and Rey et al.(2002)observed a steep increase of CO2efﬂux with theﬁrst rain after drought,indicating dynamic effects on soil CO2efﬂux.However,with our method we could identify periods in ourﬁeld data set that were not consistent with our static model by the robust re-gression approach.As we removed aboveground plant material before the measurements of soil CO2efﬂux, and determined only root biomass,we were not able to include the dynamic effects of root activity,or current photosynthates on root and heterotrophic respiration.Due to these limitations our model might be re-stricted from its formulation and parameterisation to ﬁner time scales,yet we believe that the model can be used for long-term predictions(up to a year), when coupled to a prognostic model for soil moisture, temperature,ﬁne root biomass and pH.The model equations per se do not allow for feedbacks,dynamic responses or nonlinear(sensu strictu)events,but could enhance current generation carbon cycle mod-els,which mainly concentrate on temperature effects (e.g.Cox et al.,2000),with additional factors as soil moisture,ﬁne root biomass and pH,to help address complex ecological relationships to identify feedbacks between soil respiration and climate change.We have shown that the robust regression approach is very useful as an objective means of ecological data analysis,when carefully interpreted.Through this approach we obtained parameters that are valid for normal conditions and that describe the data very well, while at the same time highlighting model problems under non-normal,transient conditions,namely dur-ing or shortly after rain events or after longer periods (>72h)of dry conditions.With a standard regression approach on the contrary,one would have got average, effective parameters that are affected by the conditions under which the model is not valid,and thus are‘ﬁtted’parameters in the bad sense of the word.The robust regression approach helps to avoid including periods in the parameterisation that are beyond the scope of the model,e.g.transient changes in diffusion path-ways or location and status of biological activity and lead to unwanted errors even in the range where the model could be valid.Moreover,we determined4to 72h as the time scale for our investigated systems, where a model based on steady-state conditions is suit-able when accounting for changes in soil temperature, moisture,pH andﬁne root biomass.ConclusionIn this study we developed a model that allows esti-mation of soil CO2efﬂux on bare soils,meadow soils as well as forest soils.The study conﬁrmed soil tem-perature and soil water content as the most important factors inﬂuencing soil CO2emission.In addition soil pH and relative root mass are found as important fac-tors to describe spatial variation of soil CO2emission due to vegetation productivity and microbial activity spans.。

Observation of the beam-size effect at HERA

Abstract
E-email: piotrzkowski@desy.de.
1 Introduction
The suppression of bremsstrahlung in beam-beam collisions due to nite transverse beam sizes, called the beam-size e ect in this paper, was rst observed in 1982 at the e+e collider VEPP-4 in Novosibirsk 1]. Soon after the discovery an explanation of this e ect was proposed 2]. In the following years several papers were published where di erent methods of calculation were used and also some predictions for future colliders were given 3]. Recently, an extensive review of these results 4] as well as some considerations in the context of the HERA collider experiments 5, 6] were published. In this article we report the measurement of the beam sizes e ect at the HERA electron-proton collider. The beam-size e ect is analogous to the phenomenon of the suppression of the electronnucleus (eN ) bremsstrahlung due to the screening of the nucleus' charge by the atomic electrons (see, for example 7]) when momentum transfers are small and correspond to distances larger than the atom size. The connection between these two e ects can be better seen when one considers the bremsstrahlung cross-sections in impact parameter space rather than in momentum transfer space. The typical impact parameter, , depends on the momenta of the colliding particles and the energy of the radiated photon - if it is larger than the atom size for eN bremsstrahlung we deal with the screening of atomic nucleus or, if it is greater than the transverse size of the interaction region (being a convolution of both beam shapes) we deal with the beam-size e ect. In the latter case, however, quantum interference has to be also taken into account 4]. Both e ects result in a decrease of the bremsstrahlung cross-sections compared to the situation of in nite atom and beam sizes.

recent findings on the phytoremediation of soils contaminated with environmentally toxic heavy metal

ReviewRecent ﬁndings on the phytoremediation of soils contaminated with environmentally toxic heavy metals and metalloids such as zinc,cadmium,lead,and arsenicI.Alkorta 1,J.Hernandez-Allica 2,J.M.Becerril 3,I.Amezaga 3,I.Albizu 2&C.Garbisu 2,*1Unidad de Biofı´s ica,Centro Mixto UPV/EHU,Apdo.644,E-48080Bilbao,Spain;2NEIKER,BasqueInstitute of Agricultural Research and Development,Berreaga 1,E-48160Derio,Spain;3Department of Plant Biology and Ecology,University of the Basque Country,Apdo.644,E-48080Bilbao,Spain (*author for correspondence:phone:+34-94-403-43-00;fax:+34-94-403-43-10;e-mail:cgarbisu@)Key words:metalloids,metals,phytochelatins,phytoextraction,phytoremediation,transgenic plants AbstractDue to their immutable nature,metals are a group of pollutants of much concern.As a result of human activities such as mining and smelting of metalliferous ores,electroplating,gas exhaust,energy and fuel production,fertilizer and pesticide application,etc.,metal pollution has become one of the most serious environmental problems today.Phytoremediation,an emerging cost-eﬀective,non-intrusive,and aesthet-ically pleasing technology,that uses the remarkable ability of plants to concentrate elements and com-pounds from the environment and to metabolize various molecules in their tissues,appears very promising for the removal of pollutants from the environment.Within this ﬁeld of phytoremediation,the utilization of plants to transport and concentrate metals from the soil into the harvestable parts of roots and above-ground shoots,i.e.,phytoextraction,may be,at present,approaching commercialization.Improvement of the capacity of plants to tolerate and accumulate metals by genetic engineering should open up new possibilities for phytoremediation.The lack of understanding pertaining to metal uptake and translocation mechanisms,enhancement amendments,and external eﬀects of phytoremediation is hindering its full scale application.Due to its great potential as a viable alternative to traditional contaminated land remediation methods,phytoremediation is currently an exciting area of active research.1.Environmental metal pollution and phytoremediationSoil pollution has recently been attracting consid-erable public attention since the magnitude of the problem in our soils calls for immediate action (Garbisu &Alkorta 2003).As a result of human activities such as mining and smelting of metallif-erous,electroplating,gas exhaust,energy and fuel production,fertilizer and pesticide application,etc.,metal pollution has become one of the most serious environmental problems today.Due to their immutable nature,metals are a group of pollutants of much concern.In fact,although several metals are essential for biological systems and must be present within a certain concentration range (Garbisu &Alkorta 2003),at high concen-trations,metals can act in a deleterious manner by blocking essential functional groups,displacing other metal ions,or modifying the active confor-mation of biological molecules (Collins &Stotzky 1989).Metal toxicity for living organisms involves oxidative and/or genotoxic mechanisms (Briat &Lebrun 1999).Based on their chemical and physical proper-ties,three diﬀerent molecular mechanisms of heavy metal toxicity can be distinguished:(i)pro-duction of reactive species by autooxidation and Fenton reaction (Fe,Cu),(ii)blocking of essential functional groups in biomolecules (Cd,Hg),and (iii)displacement of essential metal ions from biomolecules (Schutzendubel &Polle 2002).Reviews in Environmental Science and Bio/Technology 3:71–90,2004.Ó2004Kluwer Academic Publishers.Printed in the Netherlands.71Metal-contaminated soils are notoriously hard to remediate.Current technologies resort to soil excavation and either landﬁlling or soil washing followed by physical or chemical separation of the contaminants.Although highly variable and dependent on the contaminants of concern,soil properties,site conditions,and so on,the usually enormous costs associated with the removal of metals from soils by means of traditional physi-cochemical methods explain why most companies tend to ignore the problem.Due to the fact that very often large areas are aﬀected by heavy metal contamination,a removal is certainly diﬃcult. Therefore,some methods are developed to keep the metals in the soil but reduce the risks related to this presence(e.g.,by decreasing bioavailability by in situ immobilisation processes)(Diels et al. 2002).One way to facilitate such immobilisation is by altering the physicochemical properties of the metal-soil complex by introducing a multipurpose anion,such as phosphate,that enhances metal adsorption via anion-induced negative charge and metal precipitation(Bolan et al.2003a).Heavy metals cannot be destroyed biologically (no‘‘degradation’’,change in the nuclear struc-ture of the element,occurs)but are only trans-formed from one oxidation state or organic complex to another(Garbisu&Alkorta2001). Although microorganisms that use metals as ter-minal electron acceptors or reduce them as part of a detoxiﬁcation mechanism can be used for metal remediation(Garbisu&Alkorta1997), when considering the remediation of metal-pol-luted soil,metal-accumulating plants oﬀer numerous advantages over microbial processes since plants can actually extract metals from the polluted soils,theoretically rendering them clean (metal-free)(Garbisu&Alkorta2001;Garbisu et al.2002).Phytoremediation,the use of plants to extract, sequester,and/or detoxify pollutants,has been reported to be an eﬀective,non-intrusive,inex-pensive,aesthetically pleasing,socially accepted technology to remediate polluted soils(Alkorta& Garbisu2001;Weber et al.2001;Garbisu et al. 2002).Phytoremediation is widely viewed as the ecologically responsible alternative to the envi-ronmentally destructive physical remediation methods currently practiced(Meagher2000).The US phytoremediation market is expected to ex-pand more than ten-fold between1998and2005, to over$214million(Evans2002).In the last few years,some excellent reviews have been published focusing on diﬀerent aspects of phytoremediation(Salt et al.1995a,1998;Chaney et al.1997;Raskin et al.1997;Chaudhry et al. 1998;Wenzel et al.1999;Meagher2000;Navari-Izzo&Quartacci2001;Lasat2002;McGrath et al. 2002;McGrath&Zhao2003;McIntyre2003;Singh et al.2003).In any case,and in contrast to its many positive aspects,phytoremediation does have cer-tain disadvantages and limitations(Table1).Table1.Advantages and limitations of the phytoremediation technology Advantages LimitationsApplicable to a wide variety of inorganic and organic contaminants.Limited by depth(roots)and solubility and availability of the contaminant.Reduces the amount of waste going to landﬁlls.Although faster than natural attenuation,it requires long timeperiods(several years).Does not require expensive equipment or highlyspecialized personnel.Restricted to sites with low contaminant concentration.It can be applied in situ.Reduces soil disturbance and the spread of contaminants.Plant biomass from phytoextraction requires proper disposal as hazardous waste.Early estimates of the costs indicate that phytoremediation is cheaper than conventional remediation methods.Climate and season dependent.It can also lose its eﬀectiveness when damage occurs to the vegetation from disease or pests.Easy to implement and maintain.Plants are a cheap and renewable resource,easily available.Introduction of inappropriate or invasive plant species should be avoided(non-native species may aﬀect biodiversity).Environmentally friendly,aesthetically pleasing,socially accepted,low-tech alternative.Contaminants may be transferred to another medium,the environment,and/or the food chain.Less noisy than other remediation methods.Actually, trees may reduce noise from industrial activities.Amendments and cultivation practices may have negative consequences on contaminant mobility.72Within theﬁeld of phytoremediation,diﬀerent categories have been deﬁned such as,among others, phytoextraction,phytoﬁltration(rhizoﬁltration, blastoﬁltration),phytostabilization,phytovolatil-ization,phytodegradation(phytotransformation), plant-assisted bioremediation(plant-assisted deg-radation,plant-aided in situ biodegradation,phyt-ostimulation,enhanced rhizosphere degradation, rhizodegradation),etc.(Table2).Plants for phytoextraction,i.e.,metal removal from soil,should have the following characteris-tics:(i)tolerant to high levels of the metal,(ii) accumulate reasonably high levels of the metal, (iii)rapid growth rate,(iv)produce reasonably high biomass in theﬁeld,and(v)profuse root system(Garbisu et al.2002).The idea of using plants to remediate metal polluted soils came from the discovery of‘‘hyper-accumulators’’(Table3),deﬁned as plants,often endemic to naturally mineralized soils,that accu-mulate high concentrations of metals in their fo-liage(Baker&Brooks1989;Raskin et al.1997; Brooks1998).In fact,plants growing on metal-liferous soils can be grouped into three categories according to Baker(1981):(i)excluders,where metal concentrations in the shoot are maintained, up to a critical value,at a low level across a wide range of soil concentration;(ii)accumulators, where metals are concentrated in above-ground plant parts from low to high soil concentrations; and(iii)indicators,where internal concentration reﬂects external levels(McGrath et al.2002).The criterion for deﬁning Ni hyperaccumulation is 1000l g Ni g)1on a dry leaf basis(Brooks et al. 1977),whereas for Zn and Mn the threshold is 10,000l g g)1and for Cd100l g Cd g)1.Finally, the criterion for Co,Cu,Pb and Se hyperaccu-mulation is also1,000l g g)1in shoot dry matter (Brooks1998;Baker et al.2000;McGrath et al. 2002).In general terms,metal concentrations in hyperaccumulators are about100–1000-fold high-er than those found in normal plants growing on soils with background metal concentrations,and about10–100-fold higher than most other plants growing on metal-comtaminated soils(McGrath et al.2002).Hyperaccumulators are also charac-terized by a shoot-to-root metal concentration ratio of>1(i.e.,hyperaccumulator plants show a highly eﬃcient transport of metals from roots to shoots),whereas non-hyperaccumulators usually have higher metal concentrations in roots than in shoots(Baker1981;Gabbrielli et al.1990;HomerTable2.Categories of phytoremediationTerm DeﬁnitionPhytoextraction The use of plants to remove pollutants(mostly,metals)from soils.Phytoﬁltration The use of plants roots(rhizoﬁltration)or seedlings(blastoﬁltration)to absorbor adsorb pollutants(mostly,metals)from water.Phytostabilization The use of plants to reduce the bioavailability of pollutants in the environment. Phytovolatilization The use of plants to volatilize pollutants.Phytodegradation The use of plants to degrade organic pollutantsPhytotransformationPhytostimulation The use of plant roots in conjunction with their rhizospheric microorganisms toremediate soils contaminated with organics.Enhanced rhizosphere degradationRhizodegradationPlant-assisted bioremediationPlant-asssisted degradationPlant-aided in situ biodegradationTable3.Examples of hyperaccumulatorsMetal SpeciesZinc(Zn)T.caerulescensCadmium(Cd)T.caerulescensNickel(Ni)Berkheya coddiiSelenium(Se)Astragalus racemosaThallium(Tl)Iberis intermediaCopper(Cu)Ipomoea alpinaCobalt(Co)Haumaniastrum robertiiArsenic(As)P.vittata73et al.1991;Baker et al.1994a,b;Brown et al. 1995;Kra mer et al.1996;Shen et al.1997;Zhao et al.2000;McGrath et al.2002).Hyperaccumulation of heavy metal ions is in-deed a striking phenomenon exhibited by<0.2% of angiosperms(Baker&Whiting2002).Most recently,a plethora of papers are being published in an attempt to dissect the mechanisms of metal uptake,transport and accumulation,both at the physiological and molecular level(Baker& Whiting2002).The‘‘model’’hyperaccumulator Thlaspi caerulescens has been much screened in the search for new and more extreme ecotypes (McGrath et al.2001;Lombi et al.2002).Our understanding of the internal processes that confer the hyperaccumulation phenotype is advancing in leaps and bounds,and the mechanisms of trans-port,tolerance and sequestration in some species, at least in the genera Thlaspi and Alyssum,are partially elucidated(Lasat2002).Trees have also been considered for phyto-remediation of heavy metal-contaminated land, with willow and poplar being promising candi-dates,among others,in this respect(Pulford& Watson2003).According to some authors,trees potentially are the lowest-cost plant type to use for phytoremediation(Stomp et al.1994).Many trees can grow on land of marginal quality,have mas-sive root systems,and their above-ground biomass can be harvested with subsequent resprouting without disturbance of the site(Stomp et al.1994).Following the harvest of metal-enriched plants, the weight and volume of the contaminated material can be further reduced by ashing or composting(Garbisu&Alkorta2001;Garbisu et al.2002).Metal-enriched plants can be disposed of as hazardous material or,if economically fea-sible,used for metal recovery(Salt et al.1998). Recently,some studies have reported on the utili-zation of pyrolysis to separate heavy metals from hyperaccumulators(Koppolu&Clements2003).Although plants acquire essential minerals such as Fe,Cu,Ni,Zn and Se from the soil,for reasons that are not yet clear,they also have the ability to acquire and detoxify non-essential elements such as As,Cd,Cr and Pb(Salt et al.2002).Certain themes in the physiology and biochemistry of trace element accumulation by plants appear common (Salt et al.2002).Most phytoremediation studies have consid-ered metal extraction eﬃciency in relation to metal concentration of bulk soil samples or metal con-centration of the soil solution,but little is known about the eﬀect of various metal-bearing solids on metal extraction by hyperaccumulators.In fact,it has been shown that it is essential to consider the nature of the metal-bearing solids to better predict the eﬃciency of plant extraction(Dahmani-Muller et al.2001).Besides,it is also important to con-sider that metal bioavailability changes between the bulk soil and the rhizosphere,the latter being a microbiosphere which has quite diﬀerent chemical, physical and biological properties from bulk soils (Wang et al.2002b).In this respect,recently,it has been reported that root growth is a more sensitive endpoint of metal availability than chlorophyll assays(Morgan et al.2002).In order to improve phytoremediation of heavy metal polluted sites, the speciation and bioavailability of the metals in the soil,the role of plant-associated soil microor-ganisms and fungi in phytoremediation,and that of plants have to be elucidated(Kamnev&van der Lelie2000).Phytoremediation has been used in mined soil restoration,since these soils are sources of air and water pollution,by means of phytostabilization and phytoextraction techniques to stabilize toxic mine spoils and remove toxic metals from the spoils,respectively(Wong2003).Some higher plant species have developed heavy metal tolerance strategies which enable them to survive and reproduce in highly-metal contam-inated soils.Dahmani-Muller et al.(2000)inves-tigated metal uptake and accumulation strategies of two absolute metallophyte species and one pseudometallophyte.In the former two species, real hyperaccumulation in the leaves as well as metal immobilisation in roots and/or a detoxiﬁ-cation mechanism by leaf fall were found as possible strategies to deal with the high metal concentrations.By contrast,the strategy of the pseudometallophyte,i.e.,Agrostis tenuis,pre-sented a signiﬁcant metal immobilisation by the roots.Most plants have mycorrhizal fungi associated with them,providing their hosts with an increased capacity to absorb water and nutrients from the soil.The formation and function of mycorrhizal relationships are aﬀected by anthropogenic stressors including metals(Entry et al.2002). Arbuscular mycorrhizal fungi are of interest for their reported roles in alleviation of diverse74soil-associated plant stressors,including those in-duced by metals,so it has been claimed that theevaluation of the eﬃcacy of plant-mycorrhizal associations to remediate metal-polluted soils de-serves increased attention (Entry et al.2002).In addition,phytoextraction practices, e.g.,the choice of plant species and soil amendments,may have a great inﬂuence on the quantity and species composition of glomalean propagules as well as on arbuscular mycorrhizal fungi functioning during long-term metal-remediation treatments (Paw-lowska et al.2000).A unique testing system,the target-neighbour method,has been described to allow evaluation of how planting density inﬂuences metal uptake,so that the information needed to manipulate plant density for optimization of metal removal could be obtained (Shann 1995).Finally,most recently,phytoremediation has been combined with electrokinetic remediation,applying a constant voltage of 30V across the soil,concluding that the combination of both tech-niques represents a very promising approach to the decontamination of metal polluted soils (O 0Connor et al.2003).2.New ﬁndings on the phytoextraction of some of the most relevant environmentally toxic heavy metals (zinc,cadmium,lead)and metalloids (arsenic)This section is divided into four diﬀerent sub-headings.The ﬁrst three correspond to three of the most environmentally relevant heavy metals,i.e.,Zn,Cd,and Pb.The fourth sub-heading deals with As,a well-known toxic metalloid.It is important to emphasize here that very often information regarding one metal appears under a diﬀerent,apparently wrong,sub-heading.Since many of the reviewed publications deal with more than one metal at the same time,it has been preferred to present them as part of the same research,despite the fact that section structure could not be main-tained as desired.2.1.ZincZinc and Cd are ubiquitous pollutants that tend to occur together at many contaminated sites.While Zn is often phytotoxic,Cd rarely inhibits plant growth.In T.caerulescens ,an integrated molecular and physiological investigation of the fundamental mechanisms of heavy metal accumulation was conducted (Pence et al.2000).A metal transporter cDNA,znt1(expressed at very high levels in roots and shoots of this plant),was cloned from T.caerulescens through functional complementa-tion in yeast and was shown to mediate high-aﬃnity Zn uptake as well as low aﬃnity Cd uptake.Alteration in the regulation of znt1gene expression by plant Zn status results in the over-expression of this transporter and in increased Zn inﬂux in roots,even when intracellular Zn levels are high.Thus,speciﬁc alterations in Zn-respon-sive elements (e.g.,transcriptional activators)possibly play an important role in Zn hyperaccu-mulation in T.caerulescens (Pence et al.2000).In this respect,Lasat et al.(1998)found that the en-hanced root-to-shoot Zn transport in T.caerules-cens was,at least partly,achieved through an altered Zn compartmentation in the root sym-plasm,which reduces Zn sequestration in root vacuoles.A further step at elucidating the mech-anisms underlying Zn hyperaccumulation was gi-ven thanks to the cloning of metal transporter genes encoding putative vacuolar ion transportproteins in T.caerulescens (Assunc ao et al.2001).Cd NiZnPbAsCuCoTlRoot uptakeTranslocationAccumulationHarvestingDisposal/Recovery Figure 1.Phytoextraction of metals.75In fact,ZTP1(a transporter belonging to the cat-ion-eﬄux family)has been found to be highly ex-pressed in T.caerulescens,predominantly in leaves (Assunc a o et al.2001).Long et al.(2002)have recently described a large biomass Zn hyperaccumulating plant,i.e., Sedum alfredii.Similarly,Wang et al.(2003)have reported on the discovery of two new plants with potential for phytoremediation of Zn-polluted soils,i.e.,Polygonum hydropiper and Rumex ace-tosa.In the same study,the authors indicate that the consumption of rice grown in paddy soils contaminated with Cd,Cr or Zn may pose a serious risk to human health,because from22to 24%of the total metal content in the rice biomass was concentrated in the rice grain.Interestingly, Platanus acerifolia growing on heavily contami-nated soil accumulated only very low levels of heavy metals,and this mechanism for excluding metal uptake may have value in crop improvement (Wang et al.2003).Holcus lanatus L.genotypes tolerant to Zn toxicity seem to grow better than Zn-sensitive genotypes,even in Zn-deﬁcient soil,because of their greater capacity for taking up Zn from Zn-deﬁcient soil,revealing the coexistence of traits for tolerance to Zn toxicity and Zn deﬁciency in a single plant genotype(Rengel2000).Metal responses in the metallophyte Arabidop-sis halleri,a close relative to the model plant A.thaliana,that is Cd hypertolerant and Zn hy-peraccumulating,have been studied,and metal-regulated genes isolated and molecularly analyzed as interesting candidate genes for phytoremedia-tion(Bert et al.2000;Dahmani-Muller et al.2000; Clemens2001;Macnair2002).Unlike Thlaspi, A.halleri seems to be largely allogamous,seeds profusely over a longer growing season and can also spread vegetatively by stolons(Baker& Whiting2002).Macnair(2002)demonstrated that the heritability of Zn accumulation was between 25and50%,the highest yet recorded for any hy-peraccumulator,probably because of the out-breeding nature of A.halleri.Intriguingly,the genetic variability of Zn accumulation in A.halleri was manifested more when grown in low-Zn media than in high(Baker&Whiting2002).Zinc-tolerant callus lines of Brassica spp.have been developed,and this might help in the selection and characterization of heavy metal tolerance in plants for breeding programmes(Rout et al.1999).In a paper focusing on soil solution Zn and pH dynamics during phytoextraction using T.cae-rulescens J.&C.Presl,data indicate that the po-tential of this hyperaccumulator to remove Zn from contaminated soil may not be related either to acidiﬁcation of the rhizosphere(McGrath et al. 1997;Luo et al.2000)or to exudation of speciﬁc metal-mobilizing compounds(Zhao et al.2001). Intriguingly enough,some studies have provided evidence suggesting that roots of T.caerulescens are able to sense and actively forage in the Zn rich patches in soil(Schwartz et al.1999;Whiting et al. 2000).A modiﬁed glass bead compartment cultiva-tion system for studies on nutrient and trace metal uptake by arbuscular mycorrhiza using two host plant species,maize(Zea mays L.)and red clover(Trifolium pratense L.),and two arbuscu-lar mycorrhizal fungi,Glomus mossae and G. versiforme,found a striking,very high aﬃnity of the fungal mycelium for Zn,suggesting the po-tential use of arbuscular mycorrhiza in the phy-toremediation of Zn-polluted soils(Chen et al. 2001).2.2.CadmiumCadmium is one of the more mobile heavy metals in the soil-plant system,easily taken up by plants and with no essential function known to date (Lehoczky et al.2000).This element can accumu-late in plants without causing toxicity symptoms (Lehoczky et al.1998).In soybean plants,results reveal that the con-tent of Cd in diﬀerent parts of the plants was roots>>stems>seeds,indicating that the accu-mulation of Cd by roots is much larger than that of any other part of the soybean plant,and might cause deleterious eﬀects to root systems(i.e.,de-creased nodulation,changes in the ultrastructure of root nodule,etc.)(Chen et al.2003).It has been suggested that vetiver grass could be used to remediate Cd-polluted soil,since it accu-mulated218g Cd ha)1at a soil Cd concentration of0.33mg Cd kg)1(Chen et al.2000).Although sequestration of Cd by rhizosphere microorgan-isms may have an important inﬂuence on plant Cd uptake,further research is still required to estab-lish whether the accumulation of Cd by rhizobacteria inhibits,or accelerates,Cd uptake by the host plant(Robinson et al.2001).76Recent works have aimed to identify the role of antioxidative metabolism in heavy metal tolerance in T.caerulescens(Boominathan&Doran2003a, b).Hairy roots were used to test the eﬀects of high Cd environments,demonstrating that metal-in-duced oxidative stress occurs in hyperaccumulator tissues even though growth is unaﬀected by the presence of heavy metals.Superior antioxidant defenses,particularly catalase activity,may play an important role in the hyperaccumulator phe-notype of T.caerulescens.Phragmites australis plants were exposed to a high concentration of Cd,ﬁnding out that most of this element accumulated in roots,followed by leaves(Iannelli et al.2002).In roots from Cd-treated plants,both the high amount of glutathione and the parallel increase of glutathione-S-transferase activity seemed to be associated with an induction of the detoxiﬁcation processes in response to the high Cd concentration.Superoxide dismutase,ascorbate peroxidase,glutathione reductase and catalase activities as well as reduced and oxidized glutathione contents in all samples of leaves,roots and stolons were increased in the presence of Cd.Despite the fact that Cd has a redox characteristic not compatible with the Fenton-type chemistry that produces active oxygen species,Cd tolerance in Phragmites plants might be associated to the eﬃciency of these mechanisms.The eﬀect of Fe status on the uptake of Cd and Zn by two ecotypes of T.caerulescens,i.e.,Gan-ges and Prayon(the former being far superior in Cd uptake)was studied(Lombi et al.2002). Moreover,the T.caerulescens zip(Zn-regulated transporter/Fe-regulated transporter-like protein) genes,TcZNT1-G and TcIRT1-G,were cloned from Ganges and their expression under Fe-suf-ﬁcient and-deﬁcient conditions analyzed.Cad-mium uptake was signiﬁcantly enhanced by Fe deﬁciency in Ganges,while Zn uptake was not inﬂuenced by the Fe status of the plants in either of the ecotypes.These results are in agreement with the gene expression study,since the abun-dance of ZNT1-G mRNA was always similar, independently of the Fe status or ecotype,while that of TcIRT1-G mRNA was greatly increased only in Ganges root tissue under Fe-deﬁcient conditions,suggesting a possible relationship with an up-regulation in the expression of genes encoding Fe uptake,possibly TcIRT1-G(Lombi et al.2002).Recently,Song et al.(2003)reported on the utility of the yeast protein YCF1,a protein which detoxiﬁes Cd by transporting it into vacuoles,for the remediation of Cd and Pb contamination,ﬁnding out that transgenic Arabidopsis thaliana plants overexpressing YCF1showed enhanced tolerance and accumulated greater amounts of Cd and Pb.Interestingly enough,Lahner et al.(2003) analyzed several essential and nonessential ele-ments in shoots of6000mutagenized A.thaliana plants,demonstrating the utility of genomic scale proﬁling of nutrients and trace elements as a functional genomics tools.A better understanding of how plants handle mineral elements has the potential of yielding new phytoremediation capa-bilities(Rea,2003).While Cd detoxiﬁcation is certainly a complex phenomenon,probably under polygenic control, Cd real tolerance found in mine plants seems to be a simpler phenomenon,possibly involving only monogenic/oligogenic control(Sanita di Toppi& Gabbrielli1999).These authors concluded that adaptive tolerance is supported by constitutive detoxiﬁcation mechanisms,which in turn rely on constitutive homeostatic processes.Symmetric and asymmetric somatic hybridiza-tions have been used to introduce toxic metal-resistant traits from T.caerulescens into Brassica juncea(Dushenkov et al.2002).B.juncea hypo-cotyl protoplasts were fused with T.caerulescens mesophyll protoplasts,and all putative hybrids had morphological characteristics of B.juncea. Hybrid plants,produced by asymmetric somatic hybridization between the two species,demon-strated high metal accumulation potential,toler-ance to toxic metals,and good biomass production.In any case,B.juncea is by itself a good candidate for eﬃcient phytoextraction of heavy metals-such as Cd-from polluted soils (Schneider et al.1999).Two years after the toxic spill caused by the failure of a tailing pond dam at the Aznalcollar pyrite mine(SW Spain)in1998,none of the trace elements measured-As,Cd,Cu,Pb,Tl-reached levels either phytotoxic or toxic for humans or animals in seeds and the above-ground part of spill-aﬀected sunﬂower plants(Madejon et al. 2003).However,the potential for phytoextraction of these plants is very low,though they might be useful for soil conservation.The production of oil (usable for industrial purposes,whose production77。

Improving the accuracy of static GPS positioning with a new stochastic modelling procedure

INTRODUCTION GPS carrier phase measurements are extensively used for all high precision static and kinematic positioning applications. The least-squares estimation technique is usually employed in the data processing step, and basically requires the definition of two models: (a) the functional model, and (b) the stochastic model. The functional model describes the mathematical relationship between the GPS observations and the unknown parameters, while the stochastic model describes the statistical characteristics of the GPS observations (see, eg., Leick, 1995; Rizos 1997; and other texts). The stochastic model is therefore dependent on the selection of the functional model. A double-differencing technique is commonly used for constructing the functional model as it can eliminate many of the troublesome GPS biases, such as the atmospheric biases, the receiver and satellite clock biases, and so on. However, some unmodelled biases still remain in the GPS observables, even after such data differencing. Many researchers have emphasised the importance of the stochastic model, especially for high accuracy applications, for example, Barnes et al. (1998), Cross et al. (1994), Han (1997), Teunissen (1997), Wang (1998), Wang et. al. (2001) for both the static and kinematic positioning applications. In principle it is possible to further improve the accuracy and reliability of GPS results through an enhancement of the stochastic model. Previous studies have shown that GPS measurements have a heteroscedastic, space- and timecorrelated error structure (eg., Wang 1998; Wang et al., 1998a). The challenge is to find a way to realistically incorporate such information into the stochastic model. This paper deals only with the static positioning case. Several stochastic modelling techniques have recently been proposed to accommodate the heteroscedastic behaviour of GPS observations. Some are based on the signal-to-noise (SNR) ratio model (eg., Barnes et al., 1998; Brunner et al., 1999; Hartinger & Brunner, 1998; Lau & Mok, 1999; Talbot, 1988), others use a satellite elevation dependent approach (eg., Euler & Goad, 1991; Gerdan, 1995; Han, 1997; Jin, 1996; Rizos etPHY Chalermchon Satirapod is currently a Ph.D. student at the School of Geomatic Engineering, The University of New South Wales (UNSW), supported by a scholarship from the Chulalongkorn University. He graduated with a Bachelor of Engineering (Surveying) and Master of Engineering (Surveying) from Chulalongkorn University, Thailand, in 1994 and 1997 respectively. He joined the Department of Survey Engineering at Chulalongkorn University as a lecturer in late 1994. In early 1998 he joined UNSW's Satellite Navigation and Positioning (SNAP) group as a Ph.D. student. His research is focussed on automated and quality assured GPS surveying for a range of applications. ABSTRACT For high precision static GPS positioning applications, carrier phase measurements have to be processed. It is well known that there are two important aspects to the optimal processing of GPS measurements: the definition of the functional model, and the associated stochastic model. These two models must be correctly defined in order to achieve high reliability in the positioning results. The functional model is nowadays sufficiently known, however the definition of the stochastic model still remains a challenging research topic. Previous studies have shown that the GPS measurements have a heteroscedastic, space- and time-correlated error structure. Therefore, a realistic stochastic modelling procedure should take all of these error features into account. In this paper, a new stochastic modelling procedure is introduced. This procedure also takes into account the temporal correlations in the GPS measurements. To demonstrate its performance, both simulated and real data sets for short to medium length baselines have been analysed. The results indicate that the accuracy of GPS results can be improved to the millimetre level.

Soil Dynamics and Earthquake Engineering

Pile response in submerged lateral spreads:Common pitfallsof numerical and physical modeling techniquesYannis K.Chaloulos n,George D.Bouckovalas,Dimitris K.KaramitrosNational Technical University of Athens,School of Civil Engineering,9Heroon Polytechniou str.,15780Zografou,Greecea r t i c l e i n f oArticle history:Received11March2013Received in revised form31August2013Accepted16September2013Available online25October2013Keywords:Single pilesLateral spreadingLiquefaction3D numerical analysisLaminar boxa b s t r a c tA three dimensional dynamic numerical methodology is developed and used to back-analyze experi-mental data on the seismic response of single piles in laterally spreading slopes.The aim of the paper isnot to seek successful a-priori(Type A)predictions,but to explore the potential of currently availablenumerical techniques,and also to get feedback on modeling issues and assumptions which are not yetresolved in the international literature.It is illustrated that accurate simulation of the physical pile–soilinteraction mechanisms is not a routine task,as it requires the incorporation of advanced numericalfeatures,such as an effective stress constitutive soil model that can capture cyclic response and shear-induced dilation,interface elements to simulate theﬂow of liqueﬁed ground around the pile and propercalibration of soil permeability to model excess pore pressure dissipation during shaking.In addition,the“conventional tied node”formulation,commonly used to simulate lateral boundary conditions duringshaking,has to be modiﬁed in order to take into account the effects of the hydrostatic pore pressuresurplus that is created at the down slope freeﬁeld boundary of submerged slopes.A comparative analysiswith the two different lateral boundary formulations reveals that“conventional tied nodes”,which alsoreﬂect the kinematic conditions imposed by laminar box containers in centrifuge and shaking tableexperiments,may underestimate seismic demands along the upper part of the pile foundation.&2013Elsevier Ltd.All rights reserved.1.Problem outlinePile response in laterally spreading ground has raised considerableinterest among the engineering community in recent years,as moreand more cases of foundation-related damage are reported followingeach strong earthquake(e.g.Chile,2010;Christchurch,2011).So farthis problem has been studied mostly through centrifuge or shakingtable experiments(e.g.[1–5]),which provided insight into themechanisms that govern the interaction between the pile and theliqueﬁed ground,and also led to a number of empirical p–y relationsfor the liqueﬁed soil.Still,there are unresolved differences betweenthe different views and empirical methodologies which deservefurther investigation.For instance,all current design methodologies claim that soilpressures under laterally spreading conditions are drastically reducedcompared to the non-liqueﬁed case.The associated decrease is solelyrelated to the relative density of the soil and is estimated eitherthrough empirical reduction multipliers(e.g.[6])or by applyingempirical relationships for the residual strength of the liqueﬁed soil(e.g.[7]).However,in the same studies,it is recognized that pileresponse may also be affected by other soil and pile parameters,suchas the permeability of the soil or the stiffness and the type of the pile,but their effect could not be quantiﬁed.More recently,Gonzalez et al.[8]further observed that,for typical soil and pile characteristics,largenegative excess pore pressures may develop near the head of thepile,due to shear-induced dilation duringﬂow of the liqueﬁed soilaround the pile,thus increasing and not decreasing soil pressures.To appreciate the impact of the later observation,Fig.1com-pares the variation with normalized depth over pile diameter(z/D)of the ultimate soil pressures(p ult,liq)obtained from the test ofGonzalez et al.with the corresponding recommendations of thewidely used methodologies by Brandenberg et al.[6],Cubrinovskiand Ishihara[7]and Tokimatsu and Suzuki[9](shaded area).It isthus realized that dilation phenomena may increase drastically theseismic demands at the upper part of the pile relative to whatcurrent methodologies predict.The preceding short discussion reveals that pile response inlaterally spreading ground deﬁnitely deserves further study,espe-cially with respect to possible dilation phenomena appearing nearthe ground surface.Nevertheless,pursuing this task only by experi-mental means may prove cumbersome,given the large number ofthe associated parameters and the subsequent large number of teststhat should be performed.On the other hand,recent advancesin numerical methods may provide the means for a reasonablyContents lists available at ScienceDirectjournal homepage:/locate/soildynSoil Dynamics and Earthquake Engineering0267-7261/$-see front matter&2013Elsevier Ltd.All rights reserved./10.1016/j.soildyn.2013.09.009n Corresponding author.Tel.:þ302107723815;fax:þ302107723428.E-mail address:ioannischaloulos@(Y.K.Chaloulos).Soil Dynamics and Earthquake Engineering55(2013)275–287accurate investigation.Still,building such a robust numerical meth-odology is not a routine task,as it requires thorough understanding and advanced modeling of complex physical mechanisms,such as:The highly nonlinear and dilative response of the lique ﬁed soil. The interface interaction between the pile and the soil.The effects of dynamic loading and excess pore pressure build-up on soil permeability coef ﬁcient.The coupling between dynamic loading and pore ﬂuid ﬂow.The lateral boundary conditions which replicate the free ﬁeld seismic response of submerged in ﬁnite slopes.In this context,the present paper undertakes the challenge to simulate numerically the aforementioned centrifuge test of Gonzalez et al.[8].It is noted in advance that the aim is not to seek successful a-priori (Type A,according to Lambe [10])predic-tions,but to back analyze the experiment and get feedback on the suf ﬁciency of currently available numerical techniques,as well as on modeling issues and assumptions which are not yet resolved inthe international literature.Special attention is also placed on the development of a novel methodology for the accurate simulation of kinematic conditions of submerged in ﬁnite slopes.Application of the new concept reveals that the kinematic conditions imposed by tilted laminar boxes,widely used in the experimental simula-tion of such problems,cannot accurately capture boundary effects and may lead to un-conservative predictions of pile response.2.Numerical methodology 2.1.Model descriptionThe computations presented herein were performed with the Finite Difference Code FLAC3D (v.4.0).A major advantage of this code is that it allows coupling between pore water ﬂow and dynamic loading,while it makes use of an explicit integration algorithm,which is more ef ﬁcient for highly non-linear dynamic problems,such as earthquake-induced soil liquefaction.The mesh built to simulate the problem at hand is shown in Fig.2:a slender pile with given diameter D and bending stiffness EI ,is ﬁxed at the base of a submerged uniform sand layer with given Relative Density D r and permeability k .Soil stratigraphy is inclined by an angle θ,simulated through an added horizontal gravitational acceleration component,while the groundwater is maintained horizontal,at an average height of 1m above the ground surface,to ensure saturation of the sand.Following a number of parametric sensitivity analyses,the length and the width of the model are set to 74D and 17D respectively.The width of the zones is fairly small adjacent to the pile (e.g.0.1m)and increases gradually with radial distance at the free ﬁeld lateral boundaries (e.g.up to 5D).“Modi ﬁed Tied Node ”boundaries,thoroughly presented in the following section,were employed in order to simulate the free ﬁeld conditions away from the pile.Taking into account that the vertical plane through the axis of the pile is a plane of symmetry,only half of the physical problem needs to be simulated.Note that the numerical predic-tions presented in the following correspond to the reference pile,soil and excitation characteristics listed in Fig.2,except from the predictions in Section 4where the numerical analysis is used to replicate the centrifuge test of Gonzalez et al.[8].2.2.Simulation of sand responseSimilar to other advanced numerical codes,FLAC 3D allows the implementation of user-de ﬁned constitutive soil models through a C þþplug-in programming option.Hence,to cope with the anticipated highly nonlinear response of the lique ﬁed sand,the methodology adopts the NTUA Sand constitutive model [11,12],Fig.1.Existing methodologies and recent experimental data on the evaluation of ultimate soil pressures from laterally spreading soils (D r ¼40%,D ¼0.6m,EI ¼9000kNm 2).Fig.2.3D Finite Difference mesh and values of basic soil,pile and excitation parameters for the reference analysis.Y.K.Chaloulos et al./Soil Dynamics and Earthquake Engineering 55(2013)275–287276implemented to FLAC3D by Karamitros [13].NTUA Sand incorpo-rates the Critical State Theory of Soil Mechanics,so that the effects of initial state (relative density and mean effective stress)are simulated in terms of the uniquely de ﬁned state parameter ψ[14].In parallel,the hysteretic Ramberg –Osgood formulation is adopted for small strain increments,allowing for the accurate simulation of the non-linear hysteretic response of sands (decrease of shear modulus and increase of hysteretic damping with increasing cyclic shear strain amplitude)over a wide range of strain levels.Finally,shake-down effects and liquefaction softening during cyclic load-ing are properly simulated by quantifying the fabric evolution effects on plastic strain increments.To appreciate the necessity of advanced constitutive soil modeling for the problem at hand,Fig.3a and b illustrate typical stress paths obtained with the proposed methodology for two zones at depth z ¼1.25m:one in the free ﬁeld and the other right next to the pile.The black and the gray lines correspond to the N ¼2and to the N ¼10loading cycle,while the black dashed line depicts the Critical State Line (CSL).The free ﬁeld response,which resembles closely simple shear conditions,is shown in terms of shear (τvh )and vertical effective stress (s ′v ),while the response next to the pile,which is closer to triaxial conditions,is shown in terms of the effective con ﬁning (p ′)and deviatoric stress (q ).Observe that in the free ﬁeld (Fig.3a),the stress path is similar to what you would expect in a cyclic simple shear test with initial static shear offset (due to ground surface inclination).Namely,it converges towards the origin of effective stresses early during shaking and consequently stabilizes along the CSL,indicating contractive response and liquefaction related cyclic mobility.On the contrary,the response pattern next to the pile (Fig.3b)is completely different.In this case,the stress path moves away fromthe origin of effective stresses during shaking,indicating intense dilation and negative excess pore pressure build-up.Although peculiar,this response pattern is substantiated by the observations by González et al.,discussed in the introduction,emerging as the corner stone for replicating realistically the overall pile –soil interaction.2.3.Pile –soil interfaceAs seismic shaking evolves,large lateral spreading displace-ments accumulate at the ground surface,forcing the soil to ﬂow around the pile.This intense shearing is the major cause of the observed dilation,discussed in the previous paragraph.The afore-mentioned deformation mechanism is reproduced in the analysis with the installation of interface (slip and separation)elements between the soil and the pile.In that case,the net pressure (p )imposed to the pile by the soil has to be obtained through numerical integration of the normal and shear stresses correspond-ing to each interface node,with a special function developed for the present application using FLAC's programming language (FISH).A detail of the deformed mesh at the area around the pile head is shown in Fig.4a.Observe that the lique ﬁed sand is squeezed against the back side of the pile and it is relaxed in the front,while slippage occurs around the pile perimeter,in the direction of lateral spreading,as well as along the pile axis.Furthermore,Fig.4b demonstrates the potential effects of the pile –soil relative slip on p –y curves,for two analyses with and without interface elements (black and gray lines respectively).This comparison shows that slippage is essentially triggered at relatively large ground displacements,when the two curves start diverging,and results in stabilization or even slight decrease of soil pressures.Thus,ultimate soil pressures obtained without the use of interface elements are signi ﬁcantly overestimated (e.g.by about 70%in the presented case).Note that,interface elements in FLAC 3D exhibit an elastic-perfectly plastic response,de ﬁned by the following properties:elastic stiffness,cohesion and friction angle to control sliding,as wellasFig.3.Typical stress paths obtained from NTUA Sand,(a)at the free ﬁeld and (b)next to thepile.Fig.4.(a)Use of interface elements to simulate pile –soil relative slip,and (b)effect of interface elements on the evaluation of p –y curves.Y.K.Chaloulos et al./Soil Dynamics and Earthquake Engineering 55(2013)275–287277tensile strength to control separation and gap development.Selection of proper values for these parameters is based both on physical,as well as,οn numerical criteria.For instance,the elastic stiffness should be large enough to avoid excessive deformations before yielding,but at the same time it should remain lower than a certain limit so that the system does not become dysfunctional and unstable.Hence,a number of sensitivity analyses are required in order to select a proper stiffness value so that the error in computed soil pressures is minimized and,at the same time,the time step of the analysis is maintained at reasonable levels.Furthermore,for piles in sand,it is reasonable to assume that the interface has zero cohesion,while its friction angle varies between1/2φandφ(the friction angle of the sand)for steel and concrete piles respectively.Finally,it should be acknowledged that,despite the cohesionless nature of the soil,a large value was assigned to the tensile strength of the interface.This was essential in order to prevent separation and subsequent disrup-tion in the continuity of pore pressures,as a result of a possible separation of the two sides of the interface.Note that the effect of this assumption is minor since tensile forces were found to develop only in a very limited number of interface nodes(e.g.1–2out of105). For the same reason,separation associated with shear yielding,was also prohibited at the interface nodes.2.4.Soil permeabilityThe use of Darcy's coefﬁcient of soil permeability is a rather controversial issue for liquefaction related problems,as this para-meter has been established to quantifyﬂow under static conditions and through the pores of a stable soil skeleton.These conditions are not met in the present application,since the loading is dynamic and may cause alternating direction of the poreﬂuidﬂow,while excess pore pressure build-up and liquefaction leads to a rather unstable soil skeleton.Based on relevant recommendations from the literature,the following three different scenarios are considered herein with regard to the permeability coefﬁcient:(a)“Static”Permeability Coefﬁcient k¼6.6Â10À5m/s,determinedfrom constant head permeability tests under1g gravitational acceleration for D r¼40%(e.g.Arulmoli et al.[15]).(b)“Dynamic”Permeability Coefﬁcient k¼2.1Â10À5m/s,as pro-posed by Liu and Dobry[16]in order to take into account the alternatingﬂow direction within the pores of the sand skeleton during shaking.(c)Variable Permeability Coefﬁcient,k b¼k ini U½1þðαÀ1Þrβu ð1Þas proposed by Shahir et al.[17],in order to take into account that during liquefaction soil particles gradually loose contact, leading to the creation of additionalﬂow channels within the sand skeleton.All above scenarios are evaluated through comparison with experi-mental data in Section4.For that purpose,Eq.(1)was implemented in the analysis through a FISH function,which constantly updates the value of permeability at each zone and at each time step.The initial value of the permeability coefﬁcient in Eq.(1)was set to k ini¼2.1Â10À5m/s,while,based on Shahir et al.[17],it was further assumed thatα¼10andβ¼1.teral boundary conditions3.1.Tied-node boundariesNumerical analyses of inﬁnite slopes should replicate the actual ﬁeld conditions,namely that effective stresses and displacements in between any two lateral boundaries are uniform along the free ground surface and vary only with depth.For instance,foruniform Fig.5.Distribution of(a)shear and(b)horizontal effective stress after numerical establishment of static equilibrium for dry slopes using three different types of boundaries.Y.K.Chaloulos et al./Soil Dynamics and Earthquake Engineering55(2013)275–287278soil conditions and a Cartesian coordinate system xz ﬁtted to the free ground surface of an in ﬁnite slope (x is parallel to the ground surface),the static effective stresses vary linearly with z ,as:s z 0¼ðγ′cos θÞz s x 0¼k o s z0τxz ¼ðγ′sin θÞz ð2Þwhere θis the ground slope inclination relative to the horizontal,γ′is the effective unit weight of the soil,and k o is the geostatic earth pressure coef ﬁcient.In light of the above,and for small inclination angles θ,it is common practice (e.g.Cheng and Jeremic [18])to assume that the free ground surface is horizontal while the gravitational ﬁeld is inclined at an angle θrelative to the vertical.Thus,static effective stresses are computed by initially applying a vertical unit gravita-tional force equal to γ′cos θand subsequently superimposing to it a similar horizontal unit force equal to γ′sin θ.Furthermore,“tied-node ”conditions are imposed to the lateral boundaries which ensure that the vertical and horizontal displacements at opposite boundary nodes,with the same z elevation,are equal.The ef ﬁciency of tied-node boundaries,against the use of verticalrollers or hinges at the lateral boundaries of the in ﬁnite slope model,can be appreciated from Fig.5which compares numerical predictions of static shear and effective horizontal stresses in the case of a “dry ”in ﬁnite slope.In the common case of submerged slopes,considered in this paper,assuming an inclined (not vertical)gravitational ﬁeld implies that the phreatic surface is not horizontal anymore but it is rotated by an angle θso that it becomes perpendicular to the inclined unit gravitational force γ′.This transformation of the problem geometry is explained schematically in Fig.6.As a result,although effective stresses at the two lateral boundaries remain equal,hydrostatic pore pressures and total stresses are higher at the downslope than at the upslope boundary.This condition is perfectly acceptable in the free ﬁeld,where no boundary constraints are required in order to achieve equilibrium.However,in numerical analyses with “tied-node ”boundaries,the inclined phreatic surface and the associated total stress difference induce an unbalanced force in the direction of the slope gradient,in excess of the corresponding unit gravitational component γ′sin θ.Thus,the static stress ﬁeld is considerably distorted relative to that prevailing in the free ﬁeld [Eqs.(2a)–(2c)].This is shown in Fig.7,which presents the static shear and effective horizontal stress distributions within a submerged in ﬁnite slope obtained numeri-cally with tied-node boundaries.Observe that in ﬁnite slope con-ditions,i.e.uniform stress distribution along the ground surface,are now greatly violated,implying that the conventional tied-node boundary formulation needs to be revisited.In earthquake related problems,static equilibrium is followed by seismic excitation,directly applied at the base of the distorted model.Hence,the aforementioned effects of the tied node formulation on the static response of submerged slopes,are also disseminated into the subsequent seismic response.This issue is explicitly discussed and quanti ﬁed in Section 5,following the overall veri ﬁcation of the numerical methodology against experimental results.Note that the methodology described above for numerical analyses is also common for centrifuge and shaking table experi-ments using a laminar box container in order to simulate the seismic response of mildly inclined in ﬁnite slopes (e.g.Abdoun et al.[1]).Namely,in analogy with how numerical analyses are performed,a horizontally layered soil pro ﬁle is initially formed,at normal gravity,while the laminar box and the bucket of the centrifuge are in a horizontal position.Subsequently,before the onset of ﬂight,the laminar box is slightly tilted relativetoFig.6.(a)Physical Problem of the submerged slope and (b)numerical interpreta-tion with horizontal gravitational component and inclined phreaticsurface.Fig.7.Distribution of (a)shear stresses and (b)horizontal effective stresses after static equilibrium with the conventional tied nodes.Y.K.Chaloulos et al./Soil Dynamics and Earthquake Engineering 55(2013)275–287279the bucket in order to simulate the in ﬁnite slope.In this case,the role of tied node boundaries is played by the rectangular frames which constitute the walls of the laminar box.It is thus realized that the limitations of the tied-node bound-aries outlined before apply also to centrifuge experiments invol-ving submerged slopes and laminar box containers.For this reason,various efforts have been reported in the literature (e.g.[2,4,9])where the experimental con ﬁguration was altered in order to mitigate the aforementioned phreatic surface effects.Never-theless,none of these alternative con ﬁgurations has found wide acceptance until today,possibly because the associated experi-mental setup becomes considerably more complicated and may cause new sources of experimental error.3.2.Modi ﬁed tied nodesTo overcome the limitations identi ﬁed earlier,we should ﬁrst review the theoretical formulation for tied-node boundaries implemented in FLAC 3D.This is done with reference to the soil slice between tied nodes A and B in Fig.6b,and the associated body force diagram of Fig.8below.As described earlier,due to the inclined phreatic surface,hydrostatic pore pressures at grid point B are larger than at point A (i.e.u B 4u A ).Assuming further that the in ﬁnite slope is initially at equilibrium,it comes out that horizon-tal effective stresses at A and B are equal (i.e.s ′x ;B ¼s ′x ;A )while total stresses at B are larger than at A (i.e.s x ;B ¼s x ;A ).Assume now that tied-node boundaries are applied and the system is allowed to stabilize.For the above stress values,a solution cycle in FLAC 3D will proceed as follows:Step 1:Calculation of the total horizontal force,F ,at grid points A and B from the average total horizontal stresses.For positive downslope forces,accelerations and velocities,these forces are computed as:F A ¼Às xx ;A U A s F B ¼s xx ;B U A sð3Þwhere A s is the area corresponding to grid points A and B .Step 2:Calculation of acceleration,α,and velocity,v ,for each grid point from the corresponding total forces:αΑ¼F A =mv A ¼αΑU dt ¼ðÀs xx ;A U A s =m ÞU dt ð4ÞandαB ¼F B =mv B ¼αB U dt ¼ðs xx ;B U A s =m ÞU dtð5Þwhere m is the mass corresponding to each grid point and dt is the time step of the analysis.Step 3:At that point of the regular solution algorithm,the tied-node function,programmed in FISH,is called in order to modify the velocities at the boundary grid points by assigning to them an average common velocity:v av ¼v A þv B2ð6ÞNote that,due to the total stress difference at grid points A and B ,the velocity magnitude at B is larger than that at A (i.e.j v B j 4j v A j )and consequently the average velocity v av is not zero,as under dry slope conditions,but takes a positive value pointing in the down slope direction.Step 4:The solution cycle closes with the calculation of strains (based on v av )and new effective stresses (based on the adopted constitutive law).Steps 1–4are repeated until v av from Eq.(6)becomes zero.Since this velocity is the result of the difference in total stresses between corresponding points of the same elevation,equilibrium can only be achieved when the total stresses at the two (2)boundaries become equal.However,at that state,down slope effective stresses will be less than those upslope,violating the actual free-ﬁeld conditions.This is shown clearly in the numerical predictions of Fig.7b.Having understood the mechanisms triggered by the tied-node boundary conditions of submerged slopes,it is now feasible to modify their formulation so that the effective stress and deforma-tion conditions within the in ﬁnite slope remain uniform along the ground surface.The modi ﬁed tied-node formulation proposed herein is based on the idea that an upslope velocity should be applied at the lateral boundaries of the model,so that the average velocity v av computed from Eq.(6)is zero despite the totalstressFig.8.Equilibrium of the in ﬁnitesimal slice of soil around grid points A andB.Fig.9.Distribution of (a)shear stresses and (b)horizontal effective stresses after static equilibrium with the modi ﬁed tied nodes.Y.K.Chaloulos et al./Soil Dynamics and Earthquake Engineering 55(2013)275–287280difference between points A and B of the lateral boundaries. In other words,Eq.(6)should be re-written as:v av¼v Aþv Bþv up2ð7Þwherev up¼Àðv BÀv AÞð8ÞFurthermore,combining Eqs.(3)and(4),and taking also into account that s′xx,A¼s′xx,B,Eq.(8)yields:v up¼A s U dtmðs xx;AÀs xx;BÞ¼A sU dtmðu AÀu BÞð9ÞNote that all quantities in Eq.(9)can be estimated based on the geometry of the problem at hand,so that v up can be evaluated at the beginning of the analysis,prior to any stepping initiation.Fig.9a and9b,show the new predictions of shear and horizontal effective stresses for the submerged slope in Fig.6, obtained with the modiﬁed tied-node lateral boundaries outlined above.Observe that,as in the case of the dry slope presented in Fig.5,all numerical predictions are now compatible with the free ﬁeld conditions,i.e.the effective stress distributions parallel to the ground surface are uniform and agree well with the analytical solutions described by Eqs.(2a)–(2c).4.Methodology veriﬁcation4.1.Experimental set up and numerical modelThe numerical methodology described in the previous section is used herein to replicate the centrifuge test results of Gonzalez et al.[8].The purpose of this step is twofold:to verify the methodology against the basic response patterns observed in the experiment,and also to calibrate it in terms of the proper value of the permeability coefﬁcient,a parameter which is not yet well understood for liqueﬁed soils(see paragraph2.3).In the following presentation,all dimensions and measurements are given in prototype scale.The centrifuge test set up is shown in Fig.10a.Soil consists of a 6m thick layer of Nevada sand,placed at a relative density of D r¼40%,over a2m thick non-liqueﬁable layer of slightly cemented sand.The laminar box container has a mild inclination ofﬁve (5)degrees relative to the horizontal.The pile has a diameter D¼0.60m and a bending stiffness EI¼9000kNm2,while a sinu-soidal excitation of30cycles with period T¼0.5sec and peak acceleration a max¼0.30g is applied horizontally,at the base of the container.The test has been performed under50g centrifugal acceleration,and the system response has been monitored by densely placed instrumentation consisting of strain gauges,pore pressure transducers,accelerometers and linear variable differential transducers(LVDTs).The soil is saturated with metulose whose viscosity is50times that of water ensuring that,under the50g centrifugal acceleration,the permeability coefﬁcient of the D r¼40% Nevada sand will be equal to the actualﬁeld values reported in Section2.4.Note that the above scaling of the poreﬂuid viscosity does not necessarily ensure coincidence between model and prototype permeability coefﬁcients.According to Wood[19],for this to occur, the viscosity of the poreﬂuid used in the model should obey the following rule:nμ¼nρf U n gð10Þwhere nμis the model-to-prototype poreﬂuid viscosity ratio,while nρf and n g are the respective ratios for the poreﬂuid density and gravitational acceleration.In addition,in order to achieve coalescence between dynamic and diffusion time,the viscosity of the poreﬂuid used in the model should also satisfy the following condition:nμ¼n g UðnρU n GÞ0:5ð11Þwhere nρand n G are the model-to-prototype soil mass density and stiffness ratios.For the centrifuge test modeled in the present study,nρ¼n G¼1,since the same soil(Nevada Sand)is used for the model and the prototype setups,while nρf E1(metulose and water areﬂuids of very similar density).Hence,Eqs.(10)and(11) lead to a unique scaling for the viscosity of the poreﬂuid,with nμ¼50,and allow an“honest”presentation of the experimental test results in prototype scale.However,the above short discussion reveals also that,unless both Eqs.(10)and(11)are satisﬁed(e.g.when water is used as poreﬂuid in the model as well the prototype setups),a centrifuge simulation cannot necessarily reﬂect the actualﬁeld conditions,so that presentation of the experimental results in prototype scale might be misleading.Following the spirit of the paper,this issue is another potential pitfall regarding the accuracy of physical model-ing,which deserves special attention when planning small scale experiments or during interpretation of the relevant results.Fig.10b shows the variation of recorded lateral displacements with depth,at various instants during the test,and compares it to the maximum lateral displacement that is allowed by the laminar box container(heavy dashed line).It can be observed thatthe Fig.10.(a)Setup and instrumentation used in the experimental model(prototype units)and(b)variation of lateral displacements with depth measured at various test stages.Y.K.Chaloulos et al./Soil Dynamics and Earthquake Engineering55(2013)275–287281。

A New Approach for Filtering Nonlinear Systems

computational overhead as the number of calculations demanded for the generation of the Jacobian and the predictions of state estimate and covariance are large. In this paper we describe a new approach to generalising the Kalman ﬁlter to systems with nonlinear state transition and observation models. In Section 2 we describe the basic ﬁltering problem and the notation used in this paper. In Section 3 we describe the new ﬁlter. The fourth section presents a summary of the theoretical analysis of the performance of the new ﬁlter against that of the EKF. In Section 5 we demonstrate the new ﬁlter in a highly nonlinear application and we conclude with a discussion of the implications of this new ﬁlter1
Tቤተ መጻሕፍቲ ባይዱ
= = =
δij Q(i), δij R(i), 0, ∀i, j.
(3) (4) (5)

Chapter 9_Production and Cost in the Long Run

K MRTS L
MRTS diminishes. That is, as more and more labor is substituted for capital while holding output constant, the absolute value of K / L decreases. In Figure 9.1, when capital is plentiful relative to labor, the firm can discharge 10 units of capital but must substitute only 5 units of labor in order to keep output at 100.
OPTIMIZATION AND COST
An Expansion Path The Expansion Path and the Structure of Cost RETURNS TO SCALE
LONG – RUN COSTS Derivation of Cost Schedules from a Production Function Economies and Diseconomies of Scale Economies of Scope RELATION BETWEEN SHORT – RUN AND LONG – RUN COSTS – Long – Run Average Cost as the Planning Horizon – Restructuring Short – Run Costs
Chapter 9
Production and Cost in the Long Run
Long-run production decisions involve changing the level of employment of inputs that are fixed in the SR. Long-run cost of production: the cost of producing in the future when any scale of production facility can be employed. Economies of scale in production.

法布里珀罗基模共振英文

法布里珀罗基模共振英文The Fabryperot ResonanceOptics, the study of light and its properties, has been a subject of fascination for scientists and researchers for centuries. One of the fundamental phenomena in optics is the Fabry-Perot resonance, named after the French physicists Charles Fabry and Alfred Perot, who first described it in the late 19th century. This resonance effect has numerous applications in various fields, ranging from telecommunications to quantum physics, and its understanding is crucial in the development of advanced optical technologies.The Fabry-Perot resonance occurs when light is reflected multiple times between two parallel, partially reflective surfaces, known as mirrors. This creates a standing wave pattern within the cavity formed by the mirrors, where the light waves interfere constructively and destructively to produce a series of sharp peaks and valleys in the transmitted and reflected light intensity. The specific wavelengths at which the constructive interference occurs are known as the resonant wavelengths of the Fabry-Perot cavity.The resonant wavelengths of a Fabry-Perot cavity are determined bythe distance between the mirrors, the refractive index of the material within the cavity, and the wavelength of the incident light. When the optical path length, which is the product of the refractive index and the physical distance between the mirrors, is an integer multiple of the wavelength of the incident light, the light waves interfere constructively, resulting in a high-intensity transmission through the cavity. Conversely, when the optical path length is not an integer multiple of the wavelength, the light waves interfere destructively, leading to a low-intensity transmission.The sharpness of the resonant peaks in a Fabry-Perot cavity is determined by the reflectivity of the mirrors. Highly reflective mirrors result in a higher finesse, which is a measure of the ratio of the spacing between the resonant peaks to their width. This high finesse allows for the creation of narrow-linewidth, high-resolution optical filters and laser cavities, which are essential components in various optical systems.One of the key applications of the Fabry-Perot resonance is in the field of optical telecommunications. Fiber-optic communication systems often utilize Fabry-Perot filters to select specific wavelength channels for data transmission, enabling the efficient use of the available bandwidth in fiber-optic networks. These filters can be tuned by adjusting the mirror separation or the refractive index of the cavity, allowing for dynamic wavelength selection andreconfiguration of the communication system.Another important application of the Fabry-Perot resonance is in the field of laser technology. Fabry-Perot cavities are commonly used as the optical resonator in various types of lasers, providing the necessary feedback to sustain the lasing process. The high finesse of the Fabry-Perot cavity allows for the generation of highly monochromatic and coherent light, which is crucial for applications such as spectroscopy, interferometry, and precision metrology.In the realm of quantum physics, the Fabry-Perot resonance plays a crucial role in the study of cavity quantum electrodynamics (cQED). In cQED, atoms or other quantum systems are placed inside a Fabry-Perot cavity, where the strong interaction between the atoms and the confined electromagnetic field can lead to the observation of fascinating quantum phenomena, such as the Purcell effect, vacuum Rabi oscillations, and the generation of nonclassical states of light.Furthermore, the Fabry-Perot resonance has found applications in the field of optical sensing, where it is used to detect small changes in physical parameters, such as displacement, pressure, or temperature. The high sensitivity and stability of Fabry-Perot interferometers make them valuable tools in various sensing and measurement applications, ranging from seismic monitoring to the detection of gravitational waves.The Fabry-Perot resonance is a fundamental concept in optics that has enabled the development of numerous advanced optical technologies. Its versatility and importance in various fields of science and engineering have made it a subject of continuous research and innovation. As the field of optics continues to advance, the Fabry-Perot resonance will undoubtedly play an increasingly crucial role in shaping the future of optical systems and applications.。

OSRAM OSTAR Observation应用注意事项说明书

April 8, 2011page 1 of 10OSRAM OSTAR Observation Application NoteSummaryThis application note provides an overview of the general handling and functionality of the OSRAM OSTAR Observation. The im-portant optical and electrical characteristics are described and the thermal requirements for stable operation of the IR LED light source are addressed.In addition, the procedure for dimensioning an appropriate heat sink is illustrated by means of an example.Applications of the IR light source OSRAM OSTAR ObservationThere are various possibilities where our customers are using the OSRAM OSTAR Observation as IR light source:- Infrared illumination for cameras - General monitoring systems - IR data transfer- Driver assistance systems.Due to its compact and flat design together with its high light density, the OSRAM OSTAR Observation can be easily inte-grated in various applications. This opens up new application areas that were off limits to conventional IR devices.Construction of the OSRAM OSTAR ObservationDuring design of the OSRAM OSTAR Ob-servation, special attention was given to the thermal optimization of the module.The module core is formed from ten highly efficient semiconductor chips mounted on ceramic. For optimal heat transfer, the ce-ramic is directly mounted to the aluminum of the insulated metal core circuit board (base plate). This results in optimal heat dissipa-tion and additionally provides a sufficiently large area for a good thermal connection to the system heat sink where the OSRAM OSTAR module has to be attached to.With this construction, the light source itself exhibits a very low thermal resistance (R thJB ) between junction and base plate of 2.8 K/W.The frame surrounding the chips is available in black and white colour to enable a choice depending on the desired application.The black frame minimizes scattered light, which is important in imaging systems, whereas the white frame optimizes the total optical output power.Figure 1: Two frame colours are available for the OSRAM OSTAR Observation.April 8, 2011page 2 of 10Equipped with an ESD protection diode, the OSRAM OSTAR Observation possesses ESD protection up to 2 kV according to JESD22-A114-B.A thermistor (NTC EPCOS 8502) mounted to the base plate serves as a sensor for de-termining the temperature of the metal core board. The NTC temperature provides a good approximation of the average tempera-ture of the underside of the aluminum base plate. From this the junction temperature can be estimated (using R thJB ) and thus con-trolled.As a light source, semiconductors of the latest highly efficient thin film technology based on AlGaAs are employed. This pro-vides a nearly pure surface emitter with Lambertian radiation characteristics.All semiconductor chips are wired in series to achieve a constant intensity for all emit-ting surfaces.Tips for handling the OSRAM OSTAR ObservationIn order to protect the semiconductor chips from environmental influences such as mois-ture, they are encapsulated using a clear silicone.In addition, the silicone encapsulant allows an operation at a junction temperature of 145°C.Since this encapsulant is very elastic and soft, mechanical damage to the silicone should be minimized or avoided if at all pos-sible during processing (see also the appli-cation note "Handling of Silicone Resin LEDs“).This also applies to the black silicone en-capsulant for the connection contacts. Ex-cessive force on the cover can lead to spon-taneous failure of the light source (damageto the contacts).Figure 2: Areas of the silicone encapsu-lant of the OSRAM OSTAR Observation (shown in red hatch marks), which must not be damaged.In Figure 2, the corresponding locations are shown in red hatch marks.To prevent damaging or puncturing the en-capsulant the use of all types of sharp ob-jects should be avoided.Furthermore, it should be assured that the light source is provided with adequate cool-ing (see design example below) during op-eration. Even at low currents, prolonged operation without cooling can lead to over-heating, damage or even failure of the mod-ule.Electrical connection of the OS-RAM OSTAR ObservationFor easy electrical connection, the OSRAM OSTAR Observation is equipped with a 4-pin socket:Pin Assignment: Pin 1: Anode Pin 2: Thermistor Pin 3: Thermistor Pin 4: CathodeAs a mating plug, the SMD plug from ERNI (SMD214025.4-pins) is recommended.April 8, 2011page 3 of 10Mounting the OSRAM OSTAR Ob-servationSeveral mounting methods can be used for attaching the IR light source.When selecting an appropriate mounting method, make sure that a good heat transfer is provided between the OSRAM OSTAR Observation and the heat sink and that this is also guaranteed during operation.An insufficient or incorrect mounting can lead to thermal or mechanical problems dur-ing assembly.Generally, screws should be used for mount-ing the OSRAM OSTAR Observation.When mounting the module with M2 screws, a torque of 0.2 - 0.3 Nm should be used. In order to achieve a good thermal connection, the contact pressure should typically be in the range of 0.35 MPa.In addition to mounting with screws, the OSRAM OSTAR Observation can also be attached by means of gluing or clamping. When mounting with glue, care should be taken that the glue is both adhesive and thermally stable, and possesses a good thermal conductivity.When mounting a component to a heat sink, it should generally be kept in mind that the two solid surfaces must be brought into physical contact.Technical surfaces are never really flat or smooth, however, but have a certain rough-ness due to microscopic edges and depres-sions. When two such surfaces are joined together, contact occurs only at the surface peaks. The depressions remain separated and form air-filled cavities (Figure 3).DescriptionMaterial Advantages DisadvantagesThermally conductive pasteTypically silicone based, with heat conductiveparticlesThermally conductive compoundsImproved thermallyconductive paste – rub-bery film after curingThinnest connection with minimal pressureHigh thermal conductiv-ity No delaminationMaterial discharge at the edgesDanger of contamina-tion during mass pro-ductionPaste can escape and "creep" over timeConnections require curing process Phase Change Materi-als (PCM)Material of polyester or acrylic with lower glass transition temperature, filled with thermally con-ductive particlesEasy handling and mountingNo delaminationNo curingContact pressure re-quiredHeat pretreatment re-quiredThermally conductive elastomersSilicone plastic washer pads- filled with thermally conductive particles - often strengthened with glass fibers or di-electric filmsThermally conductive tapeDouble sided tape filled with particles for uniform thermal and adhesive propertiesNo leakage of materialCuring not requiredProblem with delamina-tionModerate thermal con-ductivityContact pressure re-quiredTable 1: Thermal Interface MaterialsApril 8, 2011page 4 of 10Figure 3: Heat flow with and without heat conductive material.Since air is a poor conductor of heat, these cavities should be filled with a thermally conductive material in order to significantly reduce the thermal resistance and improve the heat flow between the two adjacent sur-faces.Without an appropriate, optimally effective interface, only a limited amount of heat ex-change occurs between the two surfaces, eventually leading to overheating of the light source.To improve the heat transfer capability and reduce the thermal contact resistance, sev-eral materials are suitable.Thermally conductive pastes and com-pounds possess the lowest transfer resis-tance, but require a certain amount of care in handling.Elastomers and foils/bands are easy to use. With pretreated surfaces and appropriate contact pressure, a good thermal transfer can be realized.Table 1 shows an overview of the most commonly used thermally conductive mate-rials along with their most important advan-tages and disadvantages.Optical characteristics of the OS-RAM OSTAR ObservationWhen characterizing IR LEDs, the intensity is usually specified with two parameters - the total radiant flux Φe (units of mW) and the radiant intensity I e (units of mW/sr).The total radiant flux Φe of an LED describes the total radiated light power independent of direction. For the OSRAM OSTAR Observa-tion, this is shown in Figure 4, in relation to forward current.In contrast, the radiant intensity expresses the radiated power within a fixed solid angle (e.g. 0.01 sr ≙ ±3.2°) in the primary direction of radiation (optical axis).Figure 4: Relative total radiant flux in re-lation to forward current I F .The radiation characteristics (in the far field ) show the distribution of intensity dependent on angle and are shown for the OSRAM OSTAR Observation in Figure 5. This repre-sents a good approximation of a Lambertian source with a radiation angle of ±60°.In general, the brightness can be influenced with the help of appropriate secondary op-tics. That is, with the use of focusing optics, the light output within a particular angle can be significantly increased.April 8, 2011page 5 of 10Figure 5: Radiation characteristics with-out optics.The user should refrain from attempting to mount the primary optics to the silicone en-capsulant. This can lead to damage to the chip and especially to the bonding wires, thereby voiding the warranty provided by OSRAM.In the near field (at different operating cur-rents), the OSRAM OSTAR Observation exhibits the radiance images shown in Fig-ure 6.Figure 6: Radiance images in the near field at very low power (above) and at higher power (below).An especially homogeneous radiance is achieved through the black frame of the module - a particular advantage when using imaging optics.Optical safety regulationsDepending on the mode of operation, the OSRAM OSTAR Observation emits highly concentrated, invisible infrared radiation, which can be dangerous for the human eye. Products which contain these components must be handled according to the guidelines specified in IEC Standard 60825-1 and IEC 62471 "Photobiological Safety of Lamps and Lamp Systems“. Please see “Applica-tion Note Eye Safety” for more details.At high currents, one should always avoid looking at the optical path through a focus-ing lens, since the limits imposed by Laser Class 1M can be exceeded.Electrical characteristics and op-eration of the OSRAM OSTAR Ob-servationIn addition to optimized optical behavior, the new thin film AlGaAs technology also exhib-its improved electrical characteristics, when compared to traditional standard chip tech-nologies. These improvements lead to a significantly reduced forward voltage. It also enables higher forward currents for a given junction temperature.A typical current-voltage characteristic is shown in Figure 7.Care should be taken to observe the limiting conditions specified in the data sheet and at higher power, sufficient cooling should be provided.The OSRAM OSTAR Observation consists of a current-driven component, in which small voltage fluctuations at the input can lead to significant changes in current for theApril 8, 2011page 6 of 10device and thus to changes in the emitted output power. When selecting or developing suitable driver circuitry, it is therefore rec-ommended that appropriate current stabili-zation should also be provided. To find a suitable component for this purpose please see the manufacturer homepages linked on .Figure 7: Current-Voltage characteristic of the OSRAM OSTAR Observation.The efficiency of the OSRAM OSTAR Ob-servation module which results from the total radiated light power Φe and the electrical power P = V f x I f , is plotted in Figure 8. It is optimal at around 100 mA and de-creases at lower and higher currents.This is especially true for pulse operation at I f >100 mA, since the average optical power does not remain constant when the current is doubled and the duty cycle is halved.Figure 8: Efficiency in relation to forward current I f ; T B = 25°C, t pulse = 100µs.Thermal ConsiderationsIn order to achieve reliability and optimal performance for IR light sources such as the OSRAM OSTAR Observation, appropriate thermal management is necessary.Basically, there are two principle limitations for the maximum allowable temperature. First of all, for the OSRAM OSTAR Observa-tion, the maximum allowable base plate temperature T B of 125°C must not be ex-ceeded. Secondly, the maximum junction temperature is specified to be 145°C. Since these temperatures are dependent on the operating current and mode of operation (constant current or pulsed mode), the maximum allowable currents listed in the data sheet specify a T B of up to 125°C for DC operation. Thus, for example, the maxi-mum allowable constant current is 1 A for a base plate temperature T B = 85°C and is 650 mA at 110°C. The permissible pulse handling diagram shows the maximum cur-rent allowed for various pulse conditions with given pulse length t p and duty cycle D.April 8, 2011page 7 of 10Exceeding the maximum junction tempera-ture of 145°C can lead to irreversible dam-age to the LED and to spontaneous failure of the device.Due to underlying physical inter-dependencies associated with the function-ing of light emitting diodes, a change in the junction temperature T J - within the allowable temperature range - has an effect on several LED parameters.As a result, the forward voltage, radiant flux, wavelength and lifetime of LEDs are influ-enced by the junction temperature.Influence on forward voltage V f and optical power ΦeFor LEDs, an increase in junction tempera-ture leads to both a reduction of forward voltage V F (Figure 9), and a decrease in optical power Φe (Figure 10). The resulting changes are reversible. That is, the original default values return when the temperature change is reversed.For the application, this means that the lower the temperature of the semiconductor, the higher the light output will be.Influence on reliability and lifetimeIn general, with respect to aging, reliability and performance, continually driving the LEDs at their maximum allowable junction temperature is not recommended, since with an increase in temperature, a reduction in lifetime can be observed.Figure 9: Typical forward voltage in rela-tion to base plate temperature T B (I f = 1 A, t p = 10 ms).Figure 10: Relative optical power in rela-tion to base plate temperature for various pulsed currents (t p = 10 ms).April 8, 2011page 8 of 10Determination of the module tem-perature with the integrated NTCA good approximation of the base plate temperature TB can be determined from the measured resistance of the NTC and the curve given in the reference table (Fig-ure 12).Depending on the operating conditions, the corresponding junction temperature will be ΔT = R thJB x P D (P D = electrical power dissi-pation) higher. With appropriate feedback circuitry, T B and thus the junction tempera-ture can be regulated.Figure 11: Cross section of the OSRAM OSTAR Observation.Design ExampleIn the following example, the thermal re-quirements of the heat sink for the OSRAM OSTAR Observation are examined. In Fig-ure 13, an equivalent circuit for the different thermal resistances of the module is shown. Additional information is contained in the application note "Thermal Management of OSTAR-Projection Light Source".As a starting point for the thermal evaluation, an OSRAM OSTAR Observation module (10 Chips) is driven at an operating current of I f = 1000 mA and a maximum ambient tem-perature of T A = 50°C .From the given data and information from the data sheet, the requirements for the necessary cooling can be found by means ofthe following formula:Figure 12: Typical thermistor characteris-tics for the OSRAM OSTAR Observation (NTC EPCOS 8502).Where][][][][,)()(A I V V W P T T T K T f f Module D Safety mbient A unction J ⋅≈Δ−−=ΔWithT J(unction) = Max. Junction temperature (from data sheet: T J = 145°C)T B(aseplate) = Base plate temperatureT A(mbient) = Ambient temperature (T A = 50°C)ΔT Safety = Safety temperature range (typ.10 – 20K)V f = Forward voltage (from data sheet: V f = 15.5V)I f = Forward current (I f = 1A) Æ typ. P D, Module = 15.5 WApril 8, 2011page 9 of 10ΔT = Temperature change due to P D,ModuleR th,Interface = Thermal resistance of the transition mate-rial between the OSRAM OSTAR base plate and the cooler/heat sink (e.g. thermally conductive paste ≈ 0.1 K/W)R th,JB = Thermal resistance of the OSRAM OSTAR Observation (from data sheet: R th,JB = 2.8 K/W)R th,Heat sink = Thermal resistance of the cooler/heat sink to the environmentthe thermal resistances.In this example, the maximum thermal resis-tance required for cooling of the module can be found by:With the calculated thermal resistance value at hand, a corresponding heat sink can beselected from a manufacturer (see ). Using this setup at the given operating conditions the junction temperature of the module will be at 135°C. If a lower T J is desired, the safety temperature ΔT Safety has to be increased accordingly.In addition to a thermal evaluation by means of a simulation or a computed estimate, it is generally recommended to verify and safe-guard the design with a prototype and ther-mal measurements.ConclusionDeveloped for high power operation with pulsed currents of up to five Amperes, the OSRAM OSTAR Observation IR light source achieves a light output of several Watts, depending on operating parameters.Due to operation at high power levels, ap-propriate thermal management is particularly necessary in order to dissipate the accumu-lated heat and to assure the optimal per-formance and reliability of the module.When developing applications based on the OSRAM OSTAR Observation, it is generally recommended that in addition to thermal simulations, the design should be verified and safeguarded by means of a prototype and thermal measurements.April 8, 2011page 10 of 10Don't forget: LED Light for you is your place to be whenever you are looking for information or worldwide partners for your LED Lighting project.Author: Dr. Claus Jäger, Andreas StichABOUT OSRAM OPTO SEMICONDUCTORSOSRAM is part of the Industry sector of Siemens and one of the two leading lighting manufactur-ers in the world. Its subsidiary, OSRAM Opto Semiconductors GmbH in Regensburg (Germany), offers its customers solutions based on semiconductor technology for lighting, sensor and visu-alization applications. OSRAM Opto Semiconductors has production sites in Regensburg (Ger-many) and Penang (Malaysia). Its headquarters for North America is in Sunnyvale (USA), and for Asia in Hong Kong. OSRAM Opto Semiconductors also has sales offices throughout the world. For more information go to .All information contained in this document has been checked with the greatest care. OSRAM Opto Semiconductors GmbH can however, not be made liable for any damage that occurs in connection with the use of these contents.。

THE ARROW OF TIME

scopic view of the world as a system de generating toward complete disorder nor the microscopic view of the world as a changing but nonevolving system of in teracting particles and fields is required by fundamental phYSical laws. I submit that both views derive instead from aux iliary assumptions about the nature and origin of the universe. I propose to re place those assumptions with others that I believe are simpler and equally con sistent with observation. The resulting model of the universe, although differ ent from the one accepted by most physi cists, resolves the apparent contradiction between the historical and the thermo dynamic arrows of time, and it reconciles both with the almost time-symmetric character of physical laws at the micro scopic level. The theory implies that the universe is unfolding in time but not un raveling; on the contrary, it is becoming constantly more complex and richer in information. Irreversibility The historical and the thermodynamic arrows of time both derive from proc esses that always have the same direc tion; they are defined by events that can not be undone. What makes these proc esses irreversible? All phenomena can ultimately be described as .the interac tions of elementary particles; if the laws that govern those interactions do not distinguish between the past and the future, what is the source of the irre versibility we observe in the macroscopic world? One possibility is that the underlying microscopic laws are not in fact perfect ly time-symmetric. Evidence that a tem poral asymmetry exists at the level of subatomic particles can be found in the decay of the neutral K meson. One of

The Principles of Adsorption

The Principles of AdsorptionAdsorption is a critical process in many industries, from water purification to pharmaceuticals and beyond. In essence, it involves the attraction of molecules or particles to a surface or interface. Understanding the principles of adsorption is essential to optimizing these processes and achieving desired results.First, it is essential to understand the various types of adsorption. There are two main types: physical adsorption and chemical adsorption. Physical adsorption involves the attraction of molecules to a surface through van der Waals forces, while chemical adsorption involves a more significant interaction, with the formation of chemical bonds between the surface and adsorbing species.The strength of adsorption is determined by several factors, including temperature, pressure, and the nature of the adsorbent and adsorbate. The higher the temperature and pressure, the more significant the adsorption, but at some point, saturation can occur, and further adsorption becomes impossible. The nature of the adsorbent and adsorbate is also crucial. For example, an adsorbent with a high specific surface area will have a more significant capacity for adsorption, while a polar adsorbate will be more attracted to a polar surface.In addition to these factors, the kinetics of adsorption must also be considered. The rate of adsorption depends on the concentration of the adsorbate at the interface, the surface area of the adsorbent, and the mass transfer rate of the adsorbate to the interface.The principles of adsorption are applied in various ways in different industries. In water purification, activated carbon is used as an adsorbent to remove impurities such as chlorine and pesticides from drinking water. In the pharmaceutical industry, adsorption is used in the purification of drugs and the removal of impurities or contaminants. In the oil and gas industry, adsorption is utilized in processes such as natural gas purification and carbon dioxide capture.One area of particular interest in adsorption research is the development of new materials with improved adsorption properties. For example, graphene and other two-dimensional materials have been shown to have excellent adsorption capacity due to their high surface area. Metal-organic frameworks (MOFs) are also promising materials, with a high degree of tunability and the ability to target specific species for adsorption.In conclusion, understanding the principles of adsorption is essential for optimizing processes and achieving desired results in various industries. Factors such as temperature, pressure, the nature of the adsorbent and adsorbate, and the kinetics of adsorption all contribute to the strength and capacity of adsorption. Continued research and development of new materials with improved adsorption properties will further enhance the applications and effectiveness of this critical process.。

PROCESS AND APPARATUS FOR THE PURIFICATION OF WAST

专利名称：PROCESS AND APPARATUS FOR THE PURIFICATION OF WASTE WATER发明人：DE BRUIN, Lambertus MartinusMaria,KRAAN, Martinus Wilhelmus,DEKREUK, Merle Krista申请号：EP07709153.6申请日：20070109公开号：EP1976803A1公开日：20081008专利内容由知识产权出版社提供摘要：The invention relates to a process for the treatment of waste water with granular aerobic sludge, which waste water comprises organic matter and nutrients that have to be removed. According to the invention, the suspended material and sludge granules having poor settling ability are withdrawn from the reactor via a specifically chosen tap point. In this way a process is provided wherein the amount of suspended sludge in the effluent is drastically reduced, the selection for granular sludge is favoured, surplus sludge is withdrawn from the reactor, so that extensive post-treatment of the effluent becomes superfluous, resulting in considerable savings with regard to time and equipment. It is also proposed to periodically selectively withdraw the largest sludge granules in order to optimally control the size of the granules, to remove the sand fraction and to remove phosphate contained in the granules from the reactor.申请人：DHV B.V.地址：Laan 1914 No. 35 3813 EX Amersfoort NL国籍：NL代理机构：Kupecz, Arpad 更多信息请下载全文后查看。

Soil carbon pools and fluxes in urban ecosystems

Soil carbon pools and ﬂuxes in urban ecosystemsR.Pouyat a,*,P.Groﬀman b ,I.Yesilonis c ,L.Hernandez daNortheastern Research Station,c/o Baltimore Ecosystem Study,5200Westland Blvd.,Room 134,University of Maryland,Baltimore,MD 21227,USAbInstitute of Ecosystem Studies,Millbrook,NY 12545,USA cUniversity of Maryland,College Park,MD 20742,USAdNatural Resource Conservation Service,Staten Island,NY 10306,USA‘‘Capsule’’:Soil organic carbon pools are directly and indirectly aﬀected by urban land-use conversions and thesechanges can be observed in adjacent undisturbed forests.AbstractThe transformation of landscapes from non-urban to urban land use has the potential to greatly modify soil carbon (C)pools and ﬂuxes.For urban ecosystems,very little data exists to assess whether urbanization leads to an increase or decrease in soil C pools.We analyzed three data sets to assess the potential for urbanization to aﬀect soil organic C.These included surface (0–10cm)soil C data from unmanaged forests along an urban–rural gradient,data from ‘‘made’’soils (1m depth)from ﬁve diﬀerent cities,and surface (0–15cm)soil data of several land-use types in the city of Baltimore.Along the urban–rural land-use gradient,we found that soil organic matter concentration in the surface 10cm varied signiﬁcantly (P =0.001).In an analysis of variance,the urban forest stands had signiﬁcantly (P =0.02)higher organic C densities (kg m À2to 1m depth)than the suburban and rural stands.Our analysis of pedon data from ﬁve cities showed that the highest soil organic C densities occurred in loamy ﬁll (28.5kg m À2)with the lowest occurring in clean ﬁll and old dredge materials (1.4and 6.9kg m À2,respectively).Soil organic C densities for residential areas (15.5Æ1.2kg m À2)were consistent across cities.A comparison of land-use types showed that low density residential and institutional land-uses had 44and 38%higher organic C densities than the commercial land-use type,respectively.Our analysis shows that as adjacent land-use becomes more urbanized,forest soil C pools can be aﬀected even in stands not directly disturbed by urban land development.Data from several ‘‘made’’soils suggests that physical disturbances and inputs of various materials by humans can greatly alter the amount C stored in these soils.#2001Published by Elsevier Science Ltd.All rights reserved.Keywords:Soil organic carbon;Anthropogenic soils;Urban soils;Human modiﬁed soils;Baltimore Ecosystem Study1.IntroductionThe expansion of urban areas worldwide makes our understanding of the eﬀects of urbanization on soils increasingly important.Over 50%of the world’s popu-lation lives in urban areas (World Resources Institute,1996).In the lower 48states of the USA alone,urban areas have increased two-fold in area between 1969and 1994,and currently occupy 3.5%of the land base (Dwyer et al.,1998).On a global scale over 476,000ha of arable land is annually being lost to the expansion of urban areas (World Resources Institute,1996).The transformation of landscapes from primarily agricultural and forest uses to urbanized landscapes has the potential to greatly modify soil carbon (C)pools and ﬂuxes (Groﬀman et al.,1995;Pouyat et al.,1995b).While conversion of native ecosystems to agricultural use,and recovery from agricultural use,have been rela-tively well-studied,conversions to urban land uses have received little attention.We know,for example,that as agricultural practices have been abandoned in pre-viously forested areas in the United States,regrowth in these soils has resulted in a gradual recovery of above ground C pools (Houghton et al.,1999;Caspersen et al.,2000).Urban land conversions,however,often result in poor conditions for plant growth,and if the land is abandoned (an unlikely occurrence for large geo-graphical areas)recovery should be slower than on pre-viously agricultural lands (e.g.Clemens et al.,1984).On a global scale soil C pools are roughly three times larger than the C stored in all land plants (Schlesinger and Andrews,2000).At this scale soil C pools are pri-marily a function of the inputs of organic matter to the ecosystem (net primary productivity or NPP)and the average rate of decay within the ecosystem (soil hetero-trophic respiration),both of which are controlled by environmental factors such as soil temperature and moisture.Due to diﬀerences between sensitivities of decay rate and NPP to soil temperature and moisture,a0269-7491/01/$-see front matter #2001Published by Elsevier Science Ltd.All rights reserved.P I I :S 0269-7491(01)00263-9Environmental Pollution 116(2002)S107–S118/locate/envpol*Corresponding author.Fax:+1-410-455-8025.E-mail address:rpouyat@ (R.Pouyat).wide variation exists in organic soil C densities(kg mÀ2 to1m depth)on a global scale(Post et al.,1982). Current research eﬀorts are addressing the question of whether soil C pools will increase or decrease when global warming occurs(Kirschbaum,2000)and whether various land-use changes and their associated soil modiﬁcations will aﬀect soil C storage(Houghton et al., 1999).The long-term eﬀects of forest harvesting,for example,is negligible when regrowth of the forest occurs(Johnson,1992).Agricultural uses,however, have generally led to greater losses of soil organic C (Paul and Clark,1996).For urban ecosystems,very lit-tle data are available to assess whether urbanization leads to an increase or decrease in soil C pools.This paucity of data has made it problematic to predict or assess the regional eﬀects of land use change on soil C pools in various regions of the world(e.g.Howard et al.,1995;Ames and Lavkulich,1999).As land is converted to urban uses both direct and indirect factors can aﬀect soil C pools.Direct eﬀects include physical disturbances,burial or coverage of soil byﬁll material and impervious surfaces,and soil man-agement inputs(e.g.fertilization and irrigation).Pouyat and Eﬄand(1999)suggest that direct eﬀects often lead to ‘‘new’’soil parent material on which soil development then proceeds.Indirect eﬀects involve changes in the abiotic and biotic environment as areas are urbanized that can inﬂuence soil development in intact soils.Indirect eﬀects include,the urban heat island eﬀect(Oke,1995; Mount et al.,1999),soil hydrophobicity(White and McDonnell,1988;Craul,1992),introductions of exotic plant and animal species(Airola and Bucholz,1984; Steinberg et al.,1997),and atmospheric deposition of various pollutants(Grodzinski et al.,1984;Pouyat and McDonnell,1991;Lovett et al.,2000).Moreover,toxic, sub-lethal,or stress eﬀects of the urban environment on soil decomposers and primary producers can signiﬁcantly aﬀect soil Cﬂuxes(Pouyat et al.,1994,1997;Goldman et al.,1995;Groﬀman et al.,1995;Carreiro et al.,1999).In this paper we discuss the potential for soil dis-turbances and various urban environmental changes to aﬀect soil C pools andﬂuxes in urban ecosystems.Spe-ciﬁcally we address the following questions:(1)How large are existing soil organic C pools in urban ecosys-tems and how do they compare to the native ecosystems replaced by urbanization?and(2)How do soil organic carbon pools vary across diﬀerent land-use types in urban landscapes?To address these questions and explore the importance of urban indirect eﬀects(envi-ronmental factors)on soil C pools,we present a case study conducted in the New York City metropolitan area where the structure and function of unmanaged oak forests were compared across an urban–rural environmental gradient(McDonnell et al.,1993).To show how direct eﬀects,or soil disturbances,can modify soil C pools,we present data from newly described soil pedons(largely human‘‘made’’soils)collected as part of the New York City soil survey and other data mined from the literature.Finally,to explore at a city-wide scale the spatial variation in soil C pools resulting from these urban eﬀects,we present preliminary data from127plots in the city of Baltimore.2.Study areas2.1.New York City urban–rural gradientWe have been investigating forest soils along an urban-rural land-use gradient in the New York City metropolitan area since1990(Groﬀman et al.,1995; Pouyat et al.,1995a,b;McDonnell et al.,1997).This research includes detailed characterization of the gra-dient including the quantiﬁcation of land-use character-istics,soil chemical and physical properties,soil C and N dynamics,and ecological analysis of plant commu-nity structure and soil organism abundances.The study area includes contrasting land uses extend-ing from Bronx County,New York(New York City)to sites in Westchester County,New York,and Litchﬁeld County,Connecticut(McDonnell et al.,1993).The cli-mate of the region is characterized by warm humid summers and cold winters.Average annual air tem-peratures range from12.5 C in New York City to 8.5 C in northwestern Connecticut(NOAA,1985). Much of this diﬀerence in air temperature is attributed to the heat island eﬀect associated with New York City (Bornstein,1968).Precipitation is distributed evenly throughout the year for the entire study area and ranges from an annual average of108cm in New York City to 103cm in northwestern Connecticut(NOAA,1985). The Bronx,New York,and the study area to the north constitute the southern portion of the North-eastern Upland Physiographic Province(Broughton et al.,1966).The bedrock consists of highly metamor-phosed and dissected crystalline rocks that are com-posed of schist,granite and gneiss(Schuberth,1968). The soil types included in the study are well-drained, moderate to shallow,sandy loam soils situated on gently sloping terrain(Hill et al.,1980).A more detailed description of the study area is given in Medley et al. (1995)and Pouyat(1992).In a previous study,a20km wide by130km long belt transect was established along an urban–rural land-use gradient in the study area(Pouyat and McDonnell, 1991).The transect encompasses readily measurable diﬀerences in population density,road density,and automobile usage and corresponds closely to the loca-tion of gneiss and schist bedrock in the study area (Pouyat et al.,1995a).Forest stands were selected for the study using the following criteria:(1)location on upland sites on well-drained,moderate to shallow,S108R.Pouyat et al./Environmental Pollution116(2002)S107–S118sandy loam Inceptisols in the Dystrochrepts great group that vary only in depth to bedrock(Gonick et al.,1970; Hill et al.,1980);(2)oak-dominated forest with Quercus rubra L.and Quercus velutina Lam.,both of the sub-genus Erythrobalanus,as major components of the overstory(40–80%of total basal area);(3)minimum stand age of70years;and(4)no visual evidence of natural disturbance,such asﬁre or logging.By design,none of the study sites included a sig-niﬁcant proportion of non-native tree species.Measure-ments in previous studies,however,showed diﬀerences in abundances of non-native species of earthworms and soil organic horizon characteristics between plots.The urban,suburban,and rural plots had on average25.0, 7.0,and2.5individuals mÀ2of non-native species of earthworms,respectively(Steinberg et al.,1997).More-over,all of the urban plots and a few suburban plots have mull soil organic horizons(Pouyat,1992;Zhu and Carreiro,1999),which are characteristic of forest soils with high earthworm densities.In addition to variations in earthworm density(no.of individuals mÀ2),2–3 C higher temperatures and up to5-fold diﬀerences in heavy metaland total salt concentrations were measured in the upper 10cm of soil in the urban than in the suburban and rural stands(Pouyat et al.,1995a).Soil physical properties,such as bulk density and texture,did not diﬀer appreciably among stands(Table1).Data on organic matter and C concentrations in these soils have been reported in pre-vious studies(Groﬀman et al.,1995;Pouyat et al.,1995a). In this paper we analyze surface mineral soil and forest ﬂoor organic C pools from the same stands.2.2.New York City soil surveyNew York City is approximately80500ha in area and is mainly comprised of a group of islands,politi-cally grouped into counties or boroughs.These include Kings(Brooklyn)and Queens Counties that are both located on the western portion of Long Island,and New York(Manhattan),Bronx,and Richmond(Staten Island)Counties.Richmond County is separated from the other counties by the Hudson River.Surﬁcial geology consists of glacial drift(Pleistocene epoch)and post-glacial deposits(Schuberth,1968).The Harbor Hill terminal moraine extends across the south-ern part of the city crossing Queens,Brooklyn,and Staten Island.Glacial drift in Staten Island is redder in color(10YR or redder using Munsell system)than the drift in Queens and Brooklyn.Two glacial lakes formed on the back side of the terminal moraine submerging the island of Manhattan,the western portion of Staten Island and northern portion of Queens and Brooklyn. Outwash plains formed on the front side of the terminal moraine in Queens,Brooklyn and Staten Island. Structural geology of New York City is very complex and consists of eight bedrock formations(Schuberth,1968).These range from gneiss and schist formations in the Bronx to diabase and serpentine dominated forma-tions in Staten Island.Post-glacial formations consist of organic,eolian and anthropogenic materials.Eolian materials mostly occur on barrier islands along the Atlantic coast.These Barrier islands formed lagoons where organic deposits have accumulated during the past4000years.2.3.Baltimore City plotsThe Baltimore metropolitan area has hot humid summers and cold winters with average annual air tem-peratures ranging from14.5 C in Baltimore City to 12.8 C in the surrounding area(NRCS,1998).This diﬀerence in air temperature between the city and sur-rounding areas is attributed to the heat island eﬀect associated with Baltimore City(Brazel et al.,2000). Precipitation is distributed relatively evenly throughout the year for the entire study area and ranges from an annual average of107.5cm in Baltimore City to104cm in the surrounding metropolitan area(NRCS,1998). Baltimore City lies within two physiographic prov-inces,the Piedmont Plateau and the Atlantic Coastal Plain.The north–northeast trending Fall Line separates the two provinces,dividing the city in half.Most of the city is characterized by nearly level to gently rolling uplands,dissected by narrow stream valleys.Old igneous and metamorphic rocks underlie the Piedmont Plateau in the City of Baltimore.Much younger,poorly consolidated sediments underlie the Coastal Plain in the city.Four soil Associations make up most of the City’s area.These include:(1)Urban Land-Legore Associa-tion,which are very deep,nearly level to moderately Table1Means(ÆS.E.)of soil chemical and physical properties(10cm)by land-use type along an urban–rural transect in the New York City metropolitan area,spring1989aSoil properties Means by land-useRural Suburban UrbanPH 4.7 4.6 4.5 Conductivity(S mÀ1)0.081(0.002)0.115(0.003)0.131(0.004) Bulk density(Mg mÀ2)0.88(0.02)0.91(0.02)0.87(0.02) Sand(%)75(0.61)74(0.94)74(0.63) Clay(%)9.2(0.37)10.1(0.47)9.6(0.34) Cu(mg kgÀ1)14.8(1.94)16.0(0.68)31.6(1.14) Ni(mg kgÀ1)14.0(0.60)17.8(1.06)22.0(1.25) Pb(mg kgÀ1)26.9(0.86)36.7(2.06)110.0(4.12) Ca(mg kgÀ1)28.3(6.2)61.2(9.8)175.6(23.7) Mg(mg kgÀ1)11.3(0.69)13.4(0.76)29.8(2.8) K(mg kgÀ1)48.2(2.7)54.5(2.7)65.3(3.9) N(g kgÀ1) 1.97(0.14) 2.31(0.11) 2.63(0.11) Organic matter(g kgÀ1)73(4.3)83(2.6)97(3.3) a Values are the mean of nine plots(four composite samples per plot)for each land-use type(adapted from Pouyat,1992).R.Pouyat et al./Environmental Pollution116(2002)S107–S118S109sloping,well drained upland soils that are underlain by semibasic or mixed basic and acidic rocks(34%of the land area,of which43%is urban land);(2)Urban Land-Joppa-Sassafras Association,which are very deep, somewhat excessively drained and well-drained upland soils that are underlain by sandy or gravelly sediments (18%of the land area,of which40%is urban land);(3) Urban Land-Sunnyside Association,which are very deep,nearly level to moderately sloping,well-drained upland soils that are underlain by unstable clayey sediment(24%of land area,of which54%is urban land);and(4)highly disturbed soils that vary in slope position,drainage,and origin that make up24%of the land area(NRCS,1998).Based on the percentage of cover of these soil associations,the total land cover of the city that is made up of highly disturbed soils(urban land and Udorthents)is approximately60%,though the coverage of undisturbed soil map units by imper-vious surfaces,such as roads,is not included in this estimation.3.Materials and methods3.1.New York City urban–rural gradientTwenty-seven plots were established along the urban–rural land-use gradient.Each plot was20Â20m and consisted of sixteen5Â5-m(0.025ha)quadrats, four of which were randomly selected for sampling. Mineral soil,or soil below the O horizon,was sampled to a depth of10cm using a standard2-cm diameter stainless steel sampling probe.Approximately10cores were composited for each quadrat.Two5Â5-cm cores were taken per plot to determine bulk density.Mineral soil samples were air dried,ground and anic matter concentration of mineral soil was determined by loss on ignition(450 C for4h).The entire organic soil layer was sampled together to a depth that reached the mineral soil surface,using a15Â15cm(225cm2)tem-plate for each anic layer samples were separated into O1(corresponding to O i horizon)and O2(corresponding to O a and O e horizons)layers,oven dried at66 C to constant weight,weighed and ground in a Wiley Mill with a20-mesh stainless steel screen. Subsamples of soil and forestﬂoor were analyzed for total C and N using a Carlo Erba NA1500Analyzer. Carbon amounts(kg C mÀ2)in the forestﬂoor O-lay-ers were calculated from concentration(g kgÀ1dry mass)and dry mass(kg dry mass mÀ2)anic C densities(kg C mÀ2)in the surface mineral soil(0–10 cm depth)were calculated by multiplying concentration (mg C gÀ1soil)by the areal density(g soil cmÀ2;Gar-ten et al.,1999).The aerial density of the soil was cal-culated as the product of the bulk density(g soil cmÀ3) and the depth of the sample(10cm).Coarse fractions were estimated using pedon data from the New York City Soil Survey Database,which ranged between2 and3%in the mineral surface horizons for the soils in this study.The mineral soil organic C pools are expressed on a kg mÀ2basis.Mineral soil organic C amounts were added to forestﬂoor amounts to derive the forest soil organic pool(O layers+surface10cm mineral soil).Urban environmental eﬀects were evaluated statisti-cally in two ways.First,soil organic matter and C con-centrations,and mineral and forest soil(mineral+O1+ O2)organic C densities were regressed against distance to the urban core.Second,data were combined into urban,suburban,and rural land-use classes(nine plots per each land use)and subjected to a one-way analysis of variance(ANOVA)to test for diﬀerences in organicC amounts between land-use types.3.2.New York City soil surveyThe New York City soil survey is being conducted at two scales:a reconnaissance soil survey of the entire city and high intensity soil surveys of speciﬁed areas(Her-nandez and Galbraith,1997;Hernandez,1999).The reconnaissance survey was conducted as a medium intensity soil survey(order3)utilizing the standard classiﬁcation system,but with these added features:a higher level of analysis(at the series level),mapping at a smaller scale(1:62,500),and use of proposed new soil series.The intensive soil surveys(order1)were con-ducted using large mapping scales(1:4800and1:6000). Large-scale surveys have been completed at South Latourette Park in Staten Island(Hernandez and Gal-braith,1997)and are currently underway at Gateway National Recreation Area.The United States National Cooperative Soil Survey (NCSS)Program has prepared soil maps for much of the country,however,this eﬀort has focused primarily on agricultural areas.Mapping and classiﬁcation of urban areas requires the establishment of new soil series. In the New York City soil survey,35new soil series have been established for human disturbed soils.Only 10of the35pedons,however,have been completely described and were used in this analysis.Soil series names of highly disturbed soils have been established by deﬁning a speciﬁc range of characteristics.Soil features that have been used to develop soil series concepts include the type of transported material,anthro-pogeomorphic processes,thickness ofﬁll material,bulk density,organic carbon distribution,base saturation, trace elements,soil temperature,presence of a diag-nostic horizon after disturbance,and amount and kind of human artifacts(Hernandez,1999).The following sampling procedures have been devel-oped for the New York City soil survey.More speciﬁc sampling procedures and laboratory analysis methodsS110R.Pouyat et al./Environmental Pollution116(2002)S107–S118can be found in the Soil Survey Manual(Soil Survey Staﬀ, 1993)and Soil Survey Laboratory Methods Manual(Soil Survey Staﬀ,1996).The soil series type location was carefully selected using transect observations,landscape models,and Major Land Resource Area(MLRA)soil survey documentation.The selected pedon for sampling represents the central concept of the series.After selecting the site,a1.5m2soil pit was excavated with a backhoe or by hand,to either a depth of1.8m or contact to bed-rock,whichever is shallower.Soil horizons,or zones of uniform morphological characteristics,were identiﬁed and were described using NRCS guidelines for describ-ing and sampling soils(Schoeneberger et al.,1998).A representative sample from each horizon was collected.A subsample was used to measure organic C concentra-tions using FeSO4Titration(method6A1c;Soil Survey Staﬀ,1996).For those cases where organic C data were not available,organic C concentration was calculated by multiplying organic matter concentration by the Van Bemmelen factor,or0.58(Soil Survey Staﬀ,1992).In the ﬁeld,the20-to75-mm fraction was sieved,weighed,and discarded.Samples taken to the laboratory were sieved and weighed to determine the<20-mm fraction.Weight percentages of the>2-mm fractions were estimated from volume estimates of the>20-mm fractions and weight determinations of the<20-mm fractions by procedure 3Blb(Soil Survey Staﬀ,1996).Undisturbed clods were collected for bulk density and micromorphological analysis.Clods were obtained in the same pit face as the mixed,representative sample. Four bulk density clods were collected from each hori-zon.Two of the clods were used in the primary analysis, while the third clod was reserved for a rerun,if needed. The fourth clod was collected for preparation of thin sections and micromorphological examination(data not included in our analysis).We calculated the density of C in a horizon of unit area(1m2)as:c¼c f B D1À 2mmðÞVwhere c is carbon density, 2mm is the fraction of mate-rial larger than2mm diameter,B D is bulk density,c f is the fraction by mass of organic C,and V is the volume of the horizon(Post et al.,1982).Data for the soil horizons were summarized to report soil C density on a m2basis to a1m depth.For this paper,pedon data from disturbed soils were assigned into‘‘made’’and residential soil categories. Pedon data of disturbed soil proﬁles in urban areas were located in the literature(Short et al.,1986;Jo and McPherson,1995;Stroganova et al.,1998;Evans et al., 2000)and added to the New York City data set.The made soils categories were further subdivided based on the origin of theﬁll material(loamyﬁll,cleanﬁll,refuse, old dredge,and recent dredge materials).3.3.Baltimore City plotsPlots were located in Baltimore City by a stratiﬁed random design.Seven land-use types were delineated using1994Digital Orthophoto Quarter Quads(Mary-land Department of Natural Resources)and were weighted based on the aerial coverage of each type. These land-use types included commercial;industrial; institutional;transportation right-of-ways;high,medium, and low density residential;and forest.A grid was laid over the land-use map and200plots were randomly located on the grid.At each sample point a circular plot with an11.35m radius was established.In a previous study,plant spe-cies,vegetation structure,and other measurements were recorded in each plot(D.Nowak,unpublished data). During the summer of2000,127of the original200 plots were sampled for soils.Fewer plots were sampled because many sample points landed on impervious sur-faces,or permission was not granted to collect soil samples at a number of private residences.If more than one type of cover occurred within a plot,samples were stratiﬁed by the cover types present.Cover types included tree,managed grass,unmanaged herbaceous,and‘‘no vegetation’’categories.Two techniques were used to acquire soil samples within the plots:an undisturbed 5-cm bulk density core(three per cover type)and a composite soil sample to a depth of15cm with a2-cm diameter stainless steel sampling probe.Typically,10–15cores(2cm)were sampled in a grid pattern and composited,depending on the amount of soil surface that was exposed.The composite sample was air-dried and sieved in the lab with a2-mm mesh sieve.The undisturbed cores were used to determine soil bulk density.The dry sieved samples were used to measure various soil properties,of which organic matter content will be reported anic matter concentration of mineral soil was determined by loss on ignition(450 C for4h).Soil organic C was estimated by multiplying organic matter content by0.58(Soil Survey Staﬀ,1992). Organic C amounts(kg C mÀ2)in the mineral soil were calculated the same as for the urban-rural gradient study described above except the depth used was15cm and the fraction of coarse fragments(>2mm)were not factored into these calculations.For this analysis,we present soil organic C data for each of the land use categories used to stratify the sampling.4.Results4.1.New York City urban–rural gradientForestﬂoor mass and soil organic C varied widely among oak forests along the urban–rural transect.A signiﬁcant(P=0.01)relationship existed between soilR.Pouyat et al./Environmental Pollution116(2002)S107–S118S111organic matter concentration(g kgÀ1)in the surface10 cm and distance from the urban core with approxi-mately23%of the variation among plots explained by distance.Unlike soil organic matter,both organic C density and concentration in the surface10cm of soil were not statistically signiﬁcant when regressed against distance(P=0.07and0.06,respectively).The trend for all three measurements was to decrease from the urban to the rural end of the transect with negative slopes (À0.22,À0.13,andÀ0.01for soil organic matter con-centration,C concentration,and C density,respectively). Regressions of total forestﬂoor,O1,and O2layer mass against distance to the urban core were not statis-tically signiﬁcant.However,when plots exhibiting mull humus layers were excluded from the regression the relationship between total forestﬂoor mass and distance was statistically signiﬁcant(P=0.0004;Fig.1).The regression between distance and soil organic matter is similar to the regression with forestﬂoor mass(absent mull humus plots),where mass decreased from the urban to the rural end of the transect with a negative slope(À0.03).Distance from the urban core explained 54%of the variation of forestﬂoor mass(absent mull humus plots)among plots(Fig.1).ANOVA-based comparisons of organic C densities (kg C mÀ2)of urban,suburban,rural land use types showed contrasting trends for diﬀerent horizons. Mineral soil organic C densities were approximately 30%higher(P=0.03)in urban compared to suburban and rural stands(Fig.2).Forestﬂoor organic C den-sities showed an opposite,but statistically insigniﬁcant trend(P=0.43)as the mineral soil densities(Fig.2). The result of these contrasting responses was that dif-ferences among the land-use types in forest soil organic C densities(soil+forestﬂoor organic C)were not sta-tistically signiﬁcant(P=0.16).To explore the importance of earthworm activity on surface soil organic C pools,we compared C pools between mull and mor soil humus types.Mineral soil organic C densities were approximately39%higher (P=0.02)in mull than in mor humus types(Fig.3). In contrast,forestﬂoor organic C densities were more than2-fold higher(P=0.001)in the mor than in the mull humus types(Fig.3).No diﬀerences occurred in forest soil(mineral+O layer)organic C between mor and mull humus types,suggesting that while earthworms inﬂuence distribution of organic matter in the soil proﬁle,they do not aﬀect total forest soil C content(Fig.3).4.2.New York City soil surveySoil organic C densities to a1m depth varied widely across disturbed or‘‘made’’soil types.Variation insoil Fig.1.Plot of forestﬂoor mass as a function of distance to the urbancore for data collected along urban-rural transect in New York Citymetropolitan area.Values are the mean of nine forest plots(fourcomposite samples perplot).Fig.2.Means(ÆS.E.)of organic C densities(kg mÀ2)for urban,suburban,and rural forest stands.Bars are grouped by mineral soil(0–10cm),forestﬂoor(O1+O2),and total forest soil(forestﬂoor+mineral soil).Values are the means of nine forest plots(four compositesamples perplot).Fig.3.Means(ÆS.E.)of organic C densities(kg mÀ2)for mull andmor humus types.Bars are grouped by mineral soil(0–10cm),forestﬂoor(O1+O2),and total forest soil(forestﬂoor+mineral soil).Valuesfor mull and mor humus types are the means of nine and18forestplots,respectively(four composite samples per plot).S112R.Pouyat et al./Environmental Pollution116(2002)S107–S118。

An all-sky study of compact, isolated high-velocity clouds

a r X i v :a s t r o -p h /0206306v 1 18 J u n 2002Astronomy &Astrophysics manuscript no.(will be inserted by hand later)An all–sky study of compact,isolated high–velocity cloudsV.de Heij 1,R.Braun 2,and W.B.Burton 1,31Sterrewacht Leiden,P.O.Box 9513,2300RA Leiden,The Netherlands 2Netherlands Foundation for Research in Astronomy,P.O.Box 2,7990AA Dwingeloo,The Netherlands 3National Radio Astronomy Observatory,520Edgemont Road,Charlottesville,Virginia 22903,U.S.A.Received mmddyy/accepted mmddyy Abstract.We combine the catalog of compact high–velocity H i clouds extracted by de Heij et al.(2002)from the Leiden/Dwingeloo Survey in the northern hemisphere with the catalog extracted by Putman et al.(2002)from the Parkes HIPASS data in the southern hemisphere,and analyze the all–sky properties of the pact high–velocity clouds are a subclass of the general high–velocity cloud phenomenon which are isolated in position and velocity from the extended high–velocity Complexes and Streams down to column densities below 1.5×1018cm −2.Objects satisfying these criteria for isolation are found to have a median angular size of less than one degree.We discuss selection eﬀects relevant to the two surveys;in particular the crucial role played by obscuration due to Galactic H i .Five principal observables are deﬁned for the CHVC population:(1)the spatial deployment of the objects on the sky,(2)the kinematic distribution,(3)the number distribution of observed H i column densities,(4)the number distribution of angular sizes,and (5)the number distribution of H i linewidth.Two classes of models are considered to reproduce the observed properties.The agreement of models with the data is judged by extracting these same observables from simulations,in a manner consistent with the sensitivities of the observations and explicitly taking account of Galactic obscuration.We show that models in which the CHVCs are the H i counterparts of dark–matter halos evolving in the Local Group potential provide a good match to the observables.The best–ﬁtting populations have a maximum HI mass of 107M ⊙,a power-law slope of the HI mass distribution in the range −1.7to −1.8,and a Gaussian dispersion for their spatial distributions of between 150and 200kpc centered on both the Milky Way and M 31.Given its greater mean distance,only a small fraction of the M 31sub–population is predicted to have been detected in present surveys.An empirical model for an extended Galactic halo distribution for the CHVCs is also considered.While reproducing some aspects of the population,this class of models does not account for some key systematic features of the population.Key words.ISM:atoms –ISM:clouds –Galaxy:evolution –Galaxy:formation –Galaxies:dwarf –Galaxies:Local Group1.IntroductionSince the discovery of the H i high–velocity clouds byMuller et al.(1963),diﬀerent explanations,each with itsown characteristic distance scale,have been proposed.Itis likely that not all of the anomalous–velocity H i rep-resents a single phenomenon,in a single physical state.Determining the topology of the entire population ofanomalous–velocity H i is not a simple matter,and thetask is all the more daunting to carry out on an all–sky basis because of disparities between the observationalsurvey material available from the northern and south-ern hemispheres.The question of distance remains themost important,because the principal physical parame-ters depend on distance:mass varying as d 2,density as2V.de Heij,R.Braun,and W.B.Burton:An all–sky study of CHVCs(2001)to lie within the distance range8<d<10kpc. If,as seems plausible,the other large complexes also lie at distances ranging from several to some50kpc,they will have been substantially aﬀected by the radiation and gravitationalﬁelds of the Milky Way.Another category of anomalous H i high–velocity clouds are the compact,isolated high–velocity clouds dis-cussed by Braun&Burton(1999).CHVCs are distinct from the HVC complexes in that they are sharply bounded in angular extent at very low column density limits,i.e. below1.5×1018cm−2(de Heij et al.2002).This is an order of magnitude lower than the critical H i column density of about2×1019cm−2,where the ionized fraction is thought to increase dramatically due to the extragalactic radiation ﬁeld.For this reason,these objects are likely to provide their own shielding to ionizing radiation.Although not se-lected on the basis of angular size,such sharply bounded objects are found to be rather compact,with a median angular size of less than1degree.An analogy of the CHVC ensemble with that of the dwarf galaxy population in the Local Group is sugges-tive,and illustrates the hypothesis that is under discussion here.Some few Local Group dwarf galaxies also extend over large angles.The Sgr Dwarf Spheroidal discovered by Ibata et al.(1994)spans some40◦;presumably it was once a rather conventional dwarf,but its current proximity to the Milky Way accounts for its large angular size.This proximity has fundamentally distorted its shape,and will determine its further evolution.The streams of stars found in the halo of the Galaxy by Helmi et al.(1999)probably represent even more dramatic examples of the fate which awaits dwarf galaxies which transgress into the sphere of the Galaxy’s dominance.Analogous stellar streams have been found in M31by Ibata et al.(2001)and by Choi et al.(2002),as well as in association with Local Group dwarf galaxies by Majewski et al.(2000),indicating that accretion(and subsequent stripping)of satellites is an on-going process.If a selection were to be made of the dwarf galaxy population in the Local Group on the basis of an-gular size,the few large–angle systems which would be selected,namely the LMC and SMC,and the Sgr dwarf, and—depending on theﬂexibility of the selection crite-ria—perhaps the ill–fated coherent stellar streams in the Galactic halo,would represent systems nearby,of large an-gular extent,and currently undergoing substantial evolu-tion.Those systems selected on the basis of being compact and isolated,on the other hand,would represent dwarf galaxies typically at substantial distances and typically at a more primitive stage in their evolution.Regarding dis-tance and evolutionary status they would diﬀer from the nearby,extended objects,although at some earlier stage the distinction would not have been relevant.The possibility that some of the high–velocity clouds might be essentially extragalactic has been considered in various contexts by,among others,Oort(1966,1970, 1981),Verschuur(1975),Eichler(1976),Einasto et al. (1976),Giovanelli(1981),Bajaja et al.(1987),Wakker& van Woerden(1997),Braun&Burton(1999),and Blitz et al.(1999).It is interesting to note that the principal ear-lier arguments given against a Local Group deployment, most eﬀectively in the papers cited above by Oort and Giovanelli,were based on the angular sizes of the few large complexes and on the predominance of negative velocities in the single hemisphere of the sky for which substantial observational data were then available.The more com-plete data available now,however,show about as many features at positive velocities as at negative ones.It is also interesting to note that the papers by Eichler and by Einasto et al.cited above consider distant high–velocity clouds as possible sources of matter,including dark mat-ter,fueling continuing evolution of the Galaxy.Blitz et al. (1999)revived the suggestion that high–velocity clouds are the primordial building blocks fueling galactic growth and evolution,and argued that the extended complexes owe their angular extent to their proximity.Braun&Burton(1999)identiﬁed CHVCs as a subset of the anomalous–velocity gas that might be character-istic of a single class of HVCs,whose members plausibly originated under common circumstances and share a com-mon subsequent evolutionary history.They emphasized the importance of extracting a homogenous sample of in-dependently conﬁrmed objects from well–sampled,high–sensitivity H i surveys.The spatial and kinematic distri-butions of the CHVCs were found by Braun&Burton to be consistent with a dynamically cold ensemble spread throughout the Local Group,but with a net negative velocity with respect to the mean of the Local Group galaxies.They suggested that the CHVCs might repre-sent the low–circular–velocity dark matter halos predicted by Klypin et al.(1999)and Moore et al.(1999)in the context of the hierarchical structure paradigm of galactic evolution.These halos would contain no,or only a few, stars;most of their visible matter would be in the form of atomic hydrogen.Although many of the halos would already have been accreted into the Galaxy or M31,some would still populate the Local Group,either located in the farﬁeld or concentrated around the two dominate Local Group galaxies.Those passing close to either the Milky Way or M31would be ram–pressure stripped of their gas and tidally disrupted by the gravitationalﬁeld.Near the Milky Way,the tidally distorted features would corre-spond to the high–velocity–cloud complexes observed.The quality and quantity of survey material is impor-tant to interpretation of the CHVC population,which is a global one.The observational data entering this anal-ysis involved merging two catalogs of CHVCs,one based on the material in the Leiden/Dwingeloo Survey(LDS) of Hartmann&Burton(1997),and the other based on the H i Parkes All–Sky Survey(HIPASS)described by Barnes et al.(2001).Both of these surveys were searched for anomalous–velocity features using the algorithm de-scribed by de Heij et al.(2002,Paper I).This algorithm led to the LDS catalog of de Heij et al.for the CHVCs at declinations north of−30◦,and to the HIPASS catalog of Putman et al.(2002)for those atδ<0◦.The surveys overlap in the declination range+2◦<δ<−30◦,allowingV.de Heij,R.Braun,and W.B.Burton:An all–sky study of CHVCs3estimates of the relative completeness of the catalogs.Weare able to predict how a survey with the LDS parameters would respond to the CHVCs detected by HIPASS,andvice versa.In the subsequent simulations,we are able to sample the simulated material as if it were being observed by one of these surveys.This paper is organized as follows.We describe theapplication of the algorithm in§2.1.In§2.2we discuss various observational selection eﬀects,and indicate how to account for these.We address obscuration by our ownGalaxy in§2.2.1,the consequences of the diﬀering ob-servational parameters of the LDS and HIPASS data in §2.2.2,and the resulting diﬀering degrees of completeness of the LDS and HIPASS catalogs in§2.2.3.We discuss the observable all–sky properties of the CHVC ensemble in§3,including the spatial deployment(in§3.1),the kinematic deployment(in§3.2),and the distributions of H iﬂux and angular size(in§3.3).We then attempt to repro-duce these properties by considering models,ﬁrst based on Local Group distributions as discussed in§4and then on distributions within an extended Galactic Halo as dis-cussed in§5;these simulations are sampled as if being observed with the LDS and HIPASS programs.We dis-cuss the conclusions that can be drawn from this analysis in§6.2.Observations representing CHVCs over theentire sky2.1.Identiﬁcation criteriaA full description of the cloud extraction algorithm is given in Paper I;the most salient aspects of the algorithm are the following:–All pixels in the HIPASS and LDS material above the1.5σlevel(as appropriate to the particular survey)were assigned to a local intensity maximum;each pixel was assigned to the same maximum as its brightest neighboring pixel.If the local maximum was brighter than3σ,then the local maximum and the pixels which were assigned to it were considered to constitute a cloudlet.–Adjacent cloudlets were merged into clouds if the brightest enclosing contour for the two cloudlets either exceeds40%of the brighter peak or if the brightness exceeds10σ.–Those merged cloudlets for which the peak tempera-ture exceeds5σwere deemed clouds,and were en-tered in the catalog pending further consideration of their deviation velocity,as described below.The value of the noise that was used for this selection was de-termined locally,whereas for the other steps a preset noise value was used.To minimize the inﬂuence of noise on the peak detec-tion,the data wereﬁrst smoothed along both the spatial and spectral axes.The RMSﬂuctation level noted above refers to that within the smoothed data.Once the rel-evant pixels were assigned to a cloud,unsmoothed data were used to determine the cloud properties.The Paper I search algorithm led to the identiﬁca-tion of sub–structure within extended anomalous–velocity cloud complexes as well as to the identiﬁcation of sharply bounded,isolated sources.Because of the importance to the present analysis of selecting only isolated objects,we comment on the determination of the degree of isolation. In order to determine the degree of isolation of the clouds that were found by the algorithm,velocity–integrated im-ages were constructed of10◦by10◦ﬁelds centered on each general catalog entry.The range of integration in these moment maps extended over the velocities of all of the pixels that were assigned to a particular cloud.The CHVC classiﬁcation then depended on the column density distribution at the lowest signiﬁcant contour level(about 3σ)of1.5×1018cm−2.We demanded that this contour satisfy the following criteria:(1)that it be closed,with its greatest radial extent less than the10◦by10◦im-age size;and(2)that it not be elongated in the direction of any nearby extended emission.Since some subjectivity was involved in this assessment,two of the authors(VdH and RB)each independently carried out a complete clas-siﬁcation of all sources in the HIPASS and LDS catalogs. Identical classiﬁcation was given to about95%of the sam-ple,and consensus was reached on the remaining5%after re–examination.A slightly diﬀerent criterion for isolation was employed by Putman et al.2002in their analysis of the HIPASS sample of HVCs.Rather than employing the column den-sity contour at aﬁxed minimum value to make this assess-ment,they employed the contour at25%of the peak N HI for each object.Since the majority of detected objects are relatively faint,with a peak column density near12σ,the two criteria are nearly identical for most objects.Only for the brightest∼10%of sources might the resulting clas-siﬁcations diﬀer.We have reclassiﬁed the entire HIPASS HVC catalog with the absolute N HI criteria above,and have determined identical classiﬁcations for1800of the 1997objects listed.Given that the agreement in classiﬁca-tion is better than90%,we have chosen to simply employ the Putman et al.classiﬁcations in the current study.In this way the analysis presented here can be reproduced from these published sources.High–velocity clouds are recognized as such by virtue of their anomalous velocities.Although essentially any physical model of these objects would predict that they also occur at the modest velocities characteristic of the conventional gaseous disk of the Milky Way,at such ve-locities the objects would not satisfy our criterion for isola-tion.In our analysis,only anomalous–velocity objects with a deviation velocity greater than70km s−1were consid-ered.As deﬁned by Wakker(1990),the deviation velocity is the smallest diﬀerence between the velocity of the cloud and any Galactic velocity,measured in the Local Standard of Rest reference frame,allowed by a conventional kine-matic model in the same direction.The kinematic and4V.de Heij,R.Braun,and W.B.Burton:An all–sky study of CHVCs Fig.1.Fraction of a homogeneously distributed sample of test clouds that is not blended with Galactic emission.The sample was distributed on the sky with a Gaussian velocity distribution in the Local Standard of Rest system(upper panel), the Galactic Standard of Rest system(middle),and the Local Group Standard of Rest(lower).The average velocity of the Gaussian velocity distributions is−50km s−1for each of the simulations;the velocity dispersion is240km s−1for the LSR representation,and110km s−1for the GSR and LGSR ones,in rough accordance with the observed situation.The obscuration by Galactic emission was simulated by removing that part of the sample for which the deviation velocity (measured with respect to the LSR)is less than70km s−1,based on a model of Galactic kinematics which incorporates observed properties of the gaseous disk,including its warp andﬂare.The apparent structures are a consequence of the non–uniform obscuration in position and velocity.The appearance of the H i Zone of Avoidance diﬀers when material is considered in diﬀerent reference frames.Fig.2.Demonstration of the inﬂuence of Galactic obscuration on observable properties of a CHVC population.The left–hand panels show the distributions of measured(after obscuration)versus actual mean velocity and velocity dispersion determined from a series of1000simulations of a population of200objects.The upper panel on the left represents a population with a Gaussian velocity distribution in the GSR reference frame,withµ=−50km s−1and σ=115km s−1;the lower left–hand panel represents a population with a Gaussian velocity distribution in the LGSR frame,withµ=−55km s−1andσ=105km s−1.Obscuration removes about30%of the population and leads to both an overestimate of the dispersion and a more negative estimate of the mean velocity.The right–hand panels illustrate the degree to which populations in the GSR and LGSR frames could be distinguished via their statistical parameters.Measured and actual parameter diﬀerences between the GSR and LGSR frames are contrasted for1000 simulated populations of200objects,half deﬁned in the GSR frame and half in the LGSR frame.All populations have a mean velocity of−50km s−1and a dispersion of110km s−1in their reference frame.The measured parameter diﬀerences for the observed CHVC sample are indicated by the dashed lines.While the mean velocity does not provide signiﬁcant distinguishing capability between the GSR and LGSR frames,the velocity dispersion does.Fig.3.Demonstration of the eﬀects of diﬀering resolution and sensitivity in the LDS and HIPASS data on the extracted CHVC samples.upper left:Comparison of the velocity FWHM for the CHVCs found in the HIPASS catalog(black line)with the velocity widths of those found in the LDS catalog(red line).The velocity resolution of the LDS is1.03km s−1,but26km s−1for HIPASS;after degrading the LDS data to the HIPASS velocity resolution, the dashed red line is obtained.upper right:Comparison of the angular FWHM for the CHVCs in the HIPASS catalog(black line)with those in the LDS catalog(dashed line).The angular resolution of the LDS is36′,but15′for HIPASS;the dashed black line shows the HIPASS values after convolving with the LDS beam.lower left:Comparison of the totalﬂux of CHVCs found in the HIPASS catalog(black line)with those in the LDS catalog(red line).After compensating the HIPASS detection rates for the lower LDS sensitivity,the dashed black histogram is obtained.The compensation was performed using the relative detection rates in the region of survey overlap(see Fig.4).lower right:Comparison of the totalﬂuxes for the semi–isolated objects(:HVCs and?HVCs)entered in the HIPASS catalog (black line)with those in the LDS listing(red line).Compensation of the HIPASS detections to the corresponding LDS sensitivity gives the dashed black line.Fig.4.Indication of the degree of completeness of the CHVC catalog extracted from the LDS by de Heij et al.(2002).The solid curve shows the fraction of external galaxies with the indicated peak H i brightness temperatures that were shown by Hartmann&Burton(1997)to have been detected in the LDS.Dashed lines show the expected completeness for sensitivities that are25%better or worse,respectively,than that of the LDS.The histogram indicates the fraction of HIPASS sources from the catalog of Putman et al.(2002)within each temperature range that are also found in the LDS,in the declination zone−30◦<δ<0◦where the two surveys overlap.Fig.5.Spatial deployment of CHVCs over the sky.upper panel:Distribution of the cataloged CHVCs,with triangles representing the LDS sample of de Heij et al.(2002)atδ>0◦and diamonds representing the HIPASS sample of Putman et al.(2002)at southern declinations.Filled circles correspond to the Local Group galaxies listed by Mateo(1998). Red symbols indicate positive LSR velocities and black symbols negative velocities.The background grey–scale shows H i column depths from an integration of observed temperatures over velocities ranging from V LSR=−450km s−1 to+400km s−1,but excluding all gas with V DEV<70km s−1.lower panel:Smoothed relative densityﬁeld of the CHVCs,accounting for the diﬀerent observational parameters of the LDS and HIPASS catalogs.The cataloged CHVCs are each represented by a Gaussian with a true–angle dispersion of20◦;the totalﬂux of the Gaussian is set to unity for the LDS objects and to the likelihood of observing such an object in an LDS–like survey for the HIPASS sources.The grey-scale is calibrated in object number per steradian.Contours are drawn at relative densities of−60%,−30%,0%(in white)and30%,60%,90%(in black).A signiﬁcant over–density of CHVCs in the southern hemisphere remains after accounting for the diﬀerent observational parameters.V.de Heij,R.Braun,and W.B.Burton:An all–sky study of CHVCs5Fig.6.Kinematic deployment of CHVCs identiﬁed in the LDS(triangles)and in the HIPASS(diamonds)data,plotted against Galactic longitude for three diﬀerent kinematic reference frames,namely the LSR(upper),the GSR(middle),and the LGSR(lower panel).Theﬁlled circles show the kinematic deployment with longitude of the Local Group galaxies listed by Mateo(1998).The mean velocities and the dispersions in velocity of the CHVCs and Local Group galaxies are listed in Table2for the three reference frames.Fig.7.Kinematic deployment of CHVCs identiﬁed in the LDS(triangles)and in the HIPASS(diamonds)data,plotted against Galactic latitude in the three diﬀerent kinematic reference frames,as in Fig.6.Theﬁlled circles show the kinematic deployment with latitude of the Local Group galaxies listed by Mateo(1998).The mean velocities and the dispersions in velocity of the CHVCs and Local Group galaxies are listed in Table2for the three reference frames.Fig.8.Smoothed distributions of velocity and velocity dispersion of the CHVC ensemble.The panels on the left show the average velocity in the LSR(upper),GSR(middle),and LGSR(lower)reference frames,respectively.The panels on the right show the velocity dispersions,similarly arranged.Individual CHVCs in the ensemble were convolved with a Gaussian of true–angle dispersion of20◦.White contours for the velocity and dispersionﬁelds are at drawn at values of0,50,...km s−1;black ones are drawn at−50,−100,...km s−1.These smoothed representations of the observed situation can be compared with similarly sampled and smoothed representations of simulations,as described in the text.spatial properties of the conventional Galactic HI were de-scribed by a thin gaseous disk whose properties of volume density,vertical scale–height,kinetic temperature,and ve-locity dispersion,remain constant within the solar radius; at larger galactocentric distances the gaseous diskﬂares and warps,as described in Paper I,following Voskes& Burton(1999).The gas exhibits circular rotation with aﬂat rotation curve constant at220km s−1.Synthetic H i spectra were calculated for this model Galaxy,and then deviation velocities were measured from the extreme–velocity pixels in these spectra for which the intensity exceeded0.5K.Selection against objects at V dev<70 km s−1in the LSR frame introduces systematic eﬀects,as discussed in the following subsection.Although compactness was not explicitly demanded of these isolated objects,the67CHVCs found in the LDS survey and the179CHVCs found in the HIPASS data have a small median angular size,amounting to less than 1◦FWHM.2.2.Selection eﬀects and completenessThe CHVC samples used in this analysis were not ex-tracted from a single,homogeneous set of data,nor are they free of selection eﬀects,nor are they complete.We discuss below how we attempt to recognize and,insofar as possible,to account for some inevitable limitations.2.2.1.Systematic consequences of obscuration byH i in the Milky WayThe inevitable obscuration that follows from our perspec-tive immersed in the gaseous disk of the Milky Way,and which motivates the use of the deviation velocity,will dis-criminate against some CHVC detections.We may extend an analogy of the optical Zone of Avoidance to the21–cm regime.The optical Zone of Avoidance refers to extinction of light by dust in the Milky Way,and thus traces a band with irregular borders but roughly deﬁned by|b|<5◦;kinematics are irrelevant in the optical case,since the ab-sorption is broad–band.The HI searches for galaxies in the optical Zone of Avoidance carried out by,among oth-ers,Henning et al.(1998)in the north,and by Henning et al.(2000)and Juraszek(2000)in the south,were conﬁned to|b|<5◦.In the21–cm regime,extinction due to high–optical–depth foreground H i is largely negligible,but confusion due to line blending occurs at all latitudes.The analogous H i zone refers to a certain range in velocity,of varying width depending on l and on b,but present to some ex-tent everywhere:near zero LSR velocity,the“H i Zone of Avoidance”covers the entire sky.The nearby LSB galaxy Cep I was discovered during CHVC work(Burton et al. 1999);although it is at a relatively substantial latitude, b=8.◦0,its velocity of V hel=58km s−1locates it within the H i obscuration zone.Because of the strong depen-dence on velocities measured with respect to the Local Standard of Rest,the zone of obscuration is distorted upon transformation to a diﬀerent kinematic reference frame. (We note that the Magellanic Stream and the HVC com-plexes plausibly also discriminate against CHVC detec-tions,but because these extended features are smaller in scale and more conﬁned in velocity than the Galaxy,we do not consider them further here.)The relationship between the diﬀerent velocity refer-ence systems used to characterize the CHVC kinematics is given by the equations below.v LSR=v HEL+9cos(l)cos(b)+12sin(l)cos(b)+7sin(b)(1)v GSR=v LSR+0cos(l)cos(b)+220sin(l)cos(b)+0sin(b)(2)v LGSR=v GSR−62cos(l)cos(b)+40sin(l)cos(b)−35sin(b)(3) Note that a typographical error is present(the sign of the coeﬃcient of sin(b))in the version of Eqn.1that is published in Braun&Burton(1999).The inﬂuence of the obscuration by the modeled Galaxy is illustrated by Fig.1,where the integral6V.de Heij,R.Braun,and W.B.Burton:An all–sky study of CHVCsFig.9.Summary of the observed spatial,kinematic,angular size,andﬂux properties of the CHVC ensemble.The three panels arranged across the top of theﬁgure show sky projections,as follows:left:Smoothed densityﬁeld of the CHVC population.A Gaussian with a dispersion of20◦(true angle)was drawn at the location of each CHVC;the volume of the Gaussian is unity for both LDS and HIPASS sources–thus in this case the observations are shown directly,i.e. HIPASS sources are not weighted by the likelihood with which they would be observed in a LDS–like survey.middle: Smoothed velocityﬁeld of the population in the Galactic Standard of Rest frame.right:Smoothed velocity dispersion ﬁeld.The grey–scale bar for the left–hand panel is labeled in units of CHVC per steradian;the other two bars are labeled in units of km s−1.Contours are drawn at relative densities of−60%,−30%,0%(in white)and30%,60%, 90%(in black).White contours for the velocity and dispersionﬁelds are at drawn at values of0,50,...km s−1; black ones are drawn at−50,−100,...km s−1.The two panels in the middle row of theﬁgure show the kinematic distribution of the observed CHVC ensemble,representing V GSR plotted against l and b,as indicated.Delta functions at the observed coordinates were convolved with a Gaussian with an angular dispersion of20◦and velocity dispersion of20km s−1.The two lower panels show,respectively,the observed peak H i column density distribution of the CHVC population and the observed angular size distribution.Fig.10.Variation of heliocentric velocity versus the cosine of the angular distance between the solar apex and the direction of the object;CHVCs from the de Heij et al.(2002)LDS compilation atδ>0◦are plotted as triangles;those from the Putman et al.(2002)HIPASS compilation,as diamonds.The CHVCs with b<−65◦are plotted in red.Local Group galaxies,from the review of Mateo(1998),are indicated byﬁlled circles.The solid line represents the solar motion of V⊙=316km s−1towards l=93◦,b=−4◦as determined by Karachentsev&Makarov(1996).Dashed lines give the 1σ(V)envelope(±60km s−1,following Sandage1986)encompassing most galaxiesﬁrmly established as members of the Local Group.exp(−(V−µ)2/2σ2)dV is plotted.The range of inte-gration extends over all velocities which deviate more than70km s−1from any Local Standard of Rest(LSR) velocity allowed by the Galactic model described above;µis the average velocity andσis the standard deviation of the test clouds.The panels in Fig.1show the fraction of a population of clouds,homogeneously distributed on the sky and with a Gaussian velocity distribution relative to a particular reference frame,that are not obscured by virtue of being coincident with H i emission from the Milky Way. The upper panel of theﬁgure represents a model in which the Gaussian velocity distribution is with respect to the Local Standard of Rest frame,with an average velocity of−50km s−1and dispersion of240km s−1,in rough agreement with the measured CHVC values in this frame. In this case the obscuration is simply proportional to the velocity width of the obscuring emission.The obscuration at high latitudes is quite uniform since the infalling pop-ulation is always displaced from V LSR=0km s−1,where the obscuring gas resides.The middle panel presents a model wherein the Gaussian velocity distribution is with respect to the Galactic Standard of Rest(GSR)frame with an average velocity of−50km s−1and dispersion of 110km s−1,in rough agreement with the CHVC values in this frame.The low–latitude obscuration is similar to that in the LSR model,although more strongly modulated since the velocity dispersion is smaller.The high–latitude obscuration is quite strongly modulated since the infall ve-locity in the GSR frame overlaps with V LSR=0km s−1in the plane approximately perpendicular to the direction of rotation,(l,b)=(90◦,0◦).Broad apparent maxima in unobscured object density are centered near l=90◦and l=270◦.The lower panel presents a model in which the Gaussian velocity distribution is with respect to the Local Group Standard of Rest(LGSR),again using an average velocity of−50km s−1and dispersion of110km s−1.The pattern of obscuration is very similar to that of the GSR case,although the maxima in unobscured object density are slightly shifted with respect to b=0◦.These results indicate that caution must be exercised in interpreting ap-parent spatial concentrations of detected objects without properly accounting for the distortions introduced by the H i Zone of Avoidance.We have also considered how the measured statistics of a distribution,namely the mean velocity and dispersion, are inﬂuenced by the non–completeness caused by obscu-ration.Figure2shows the distribution of the errors in the average velocity and dispersions for1000simulations, each involving200test clouds;one set of simulations was run with the GSR as the natural reference frame,and a second set was run with the LGSR as the natural frame. After removing the test clouds that have an LSR devia-tion velocity less than70km s−1,the velocity dispersion of the simulated ensemble was measured for both the GSR and the LGSR velocity systems,and compared with what would have been determined if there had been no obscu-ration by the Galaxy.The upper left–hand panel in Fig.2refers to test ob-jects with a Gaussian distribution in V GSR with a dis-persion of115km s−1and average of−50km s−1.The measured dispersion exceeds the true one by9km s−1, whereas a more negative average velocity is inferred by12km s−1.The lower left–hand panel is based upon test samples with a Gaussian distribution in V LGSR with a dispersion of105km s−1and an average of−55km s−1. The diﬀerences between the measured and true disper-sion and average velocity,of6km s−1and−5km s−1, respectively,are smaller than for the GSR system.From。

1、下载文档前请自行甄别文档内容的完整性，平台不提供额外的编辑、内容补充、找答案等附加服务。
2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
3、如文档侵犯您的权益，请联系客服反馈,我们会尽快为您处理(人工客服工作时间：9:00-18:30)。

a r X i v :a s t r o -p h /0001155v 1 10 J a n 2000A&A manuscript no.(will be inserted by hand later)Key words:ISM:abundances—ISM:individual(NGC7023)—reﬂection nebulae—stars:individual(HD200775)—stars:pre-main-sequence—infrared:ISM:lines and bands1.IntroductionNGC7023is a prototypical photodissociation region that hasbeen largely studied in the last10years.It is a reﬂection nebulailluminated by the Herbig B3Ve star HD200775.HD200775islocated in a cavity of the molecular cloud whose sharp edges de-lineate perfectly the optical nebula(Fuente et al.1992,Rogerset al.1995,Fuente et al.1998).Observations of CI and OIIby Chokshi et al.(1988)reveal that a dense PDR is formedin the walls of this cavity with its peak located50”NW fromthe star.Chokshi et al.(1988)carried out theﬁrst model forthis region and estimated a UVﬁeld of G0(λ>912˚A)=2.4103in units of the Habingﬁeld and a density of n=104cm−3.Because of its edge-on geometry and proximity(d∼440pc),this PDR turned out to be one of the best sites to study thephysical and chemical processes taking place in a PDR.Fig.1shows the column density map of HI as derived from the[HI]21cm line superposed on the integrated intensity map of the13CO J=1→0line.The HI peak is shifted≈20′′relative tothe molecular ridge,showing the layered structure expected ina PDR(Fuente et al.1998).Very high angular resolution images of the region in theExtended Red Emission(ERE)and vibrationally excited H2lines show that the PDR has aﬁlamentary structure with verybright thinﬁlaments located50′′NW and70′′SW from the star(Sellgren et al.1992,Lemaire et al.1996).Hereafter,we willrefer to these positions as NW PDR and SW PDR respectively.The NW PDR is spatially coincident with the peak of the CIand OII lines and has been extensively studied in atomic andmolecular lines(Chokshi et al.1988;Fuente et al.1993,1996,1997;Lemaire et al.1996,1999;Martini et al.1997;Gerin et al.1998).Fuente et al.(1993,1997)observed that the CN/HCNand CO+/HCO+abundance ratios increases by a factor10and>100respectively towards this position relative to the valuesin the molecular cloud.Both,the CN/HCN and CO+/HCO+abundance ratios are expected to increase in the lower extinc-tion layers of PDRs(Fuente et al.1993,Sternberg&Dalgarno1995).An interferometric image of the HCO+line towards theNW PDR show that the PDR has also aﬁlamentary structure in molecular emission with high densityﬁlaments(n>105 cm−3)embedded in a more diﬀuse medium(n∼104cm−3) (Fuente et al.,1996).The HCO+ﬁlaments are spatially coin-cident with theﬁlaments seen in the near-infrared continuum images and in the emission of the vibrational H2lines.The SW PDR is fainter by a factor2-3than the northern one and has been less studied in both,molecular and atomic lines.Gerin et al.(1998)reported observations of this PDR in CO and CI lines.We have observed a strip which joins the NW PDR,the star and SW PDR using the SWS and L WS instrument on board of ISO.These observations have provided a very important information about the extent and spatial distribution of the warm H2and the atomic species within the PDR.2.ObservationsThe observations were made using the Long-Wavelength Spec-trometer(L WS)(Clegg et al.1996,Swinyard et al.1996)and the Short-Wavelength Spectrometer(SWS)(de Graauw et al. 1996)on board the Infrared Space Observatory(ISO)(Kessler et al.1996).The L WS observations were carried out in full grat-ing scan mode(L WS01AOT)which provides coverage of the 43-90.5µm range with a spectral resolution of0.29µm and of the90.5-196.5µm range with a resolution of0.60µm with ten independent detectors.Since the instrumental beam size is 74′′-90′′,we have observed only three positions of the nebula, the NW PDR,the star and the SW PDR.In Table1,we give the coordinates of the observed positions.The data were pro-cessed using version6of the oﬀ-line pipeline(OLP V6).The uncertainty in the calibration is of about30%(Swinyard et al.1996)We have observed one HI recombination line(Brα),ﬁve H2 pure rotational lines(S(0),S(1),S(2),S(3),S(4)and S(5)v=0–0) and four ionic and atomicﬁne structure lines([SIV]10.5µm, [FeII]26.0µm,[SIII]33.5µm and[SiII]34.8µm)using SWS AOT02.This mode provides a spectral resolution ofλ/∆λ∼1000-2000.All the observed lines are unresolved at this spectral resolution.The aperture of the instrument in these line ranges from14′′x20′′to20′′x33′′.In order to study the spatial distribution of these lines we have observed5positions along the strip that joins the NW PDR,the star and the SW PDR. These positions are marked in Fig.1,the observed spectra are shown in Fig.2and absolute coordinates are given in Table1. Data reduction are carried out with version7.0of the OﬀLine Processing routines and the SWS Interactive Analysis at the ISO Spectrometer Data Center at the Max-Planck Institut f¨u r Extraterrestrische Physik.The uncertainties in the calibration are of about20%at short wavelengths(2-20µm)and of about30%at longer wavelengths(Salama et al.1997).3.ResultsThe L WS and SWS observational results are shown in Table2 and3.The spectra of the SWS observations are shown in Fig.2.In order to study the spatial distribution of the diﬀerent lines,we have plotted in Fig.3the normalized intensities of the[HI]21cm line(Fuente et al.1996,1998),the H2S(1)v=0–0and[SiII]34.8µm lines as a function of the distance from the star.3.1.Recombination linesThe Brαline has only been detected towards the star.Con-tinuum emission at3.6cm and6cm was detected by Skinner et al.(1993).They interpreted the emission as arising in an ionized anisotropic ter,Nisini et al(1995)observed the Paβ,Brγ,Brα,Pfγand Pfβrecombination lines with an aperture of10′′and modeled the wind with a mass loss rate of 3.910−7M⊙yr−1and a terminal velocity of280kms−1.We have measured an intensity of the Brαline of1.410−3erg s−1 cm−2sr−1which agrees within the calibration uncertainties with the data of Nisini et al.(1995).3.2.The CII and OI linesThe[CII]157.7µm and the[OI]63and145.6µm lines have been detected towards all the observed positions.The intensi-ties and line intensity ratios are shown in Table2.The emis-sion of the three lines is maximum in the NW PDR and de-creases towards the south,being the SW PDR fainter by a factor of2than the NW PDR.Along the observed strip,the OI(63µm)/OI(145.6µm)and OI(63µm)/CII(157.7µm)ratios are uniform and take the values∼10and∼2respectively. This means that at the scale of the L WS observations,the physical conditions of the nebula are quite uniform.High an-gular resolution observations show that this PDR is formed by high densityﬁlaments immersed in a more diﬀuse medium (Sellgren et al.1992,Lemaire et al.1996,Fuente et al.1996). Because of the low angular resolution of the L WS observations, they are very likely tracing the more extended diﬀuse medium.3.3.Other ionic and atomicﬁne structure lines:SiIIThe spatial distribution of the SiII line is very diﬀerent from that of the CII,OI and HI lines(see Fig.3).While the CII,OI and HI lines peak towards the NW PDR,the SiII line peaks towards the star,with the intensity towards the star∼3times larger than towards the NW PDR.This diﬀerent spatial dis-tribution cannot be explained by the diﬀerent angular resolu-tion of the CII and OI observations.The[HI]21cm line map which has an angular resolution similar to that of the SiII line presents a ring-like morphology with the peaks towards the NW and SW PDRs and a minimum towards the star,just in the peak of the SiII line.The implications of the peculiar spa-tial distribution of the SiII emission are discussed in Section 4.1.We have not detected the[SIII]33.480µm,[SIV]10.51µm and[FeII]25.9µm lines at any position(see Table3for upper limits).3.4.H2rotational linesWe have observed the v=0–0H2rotational lines from S(0)to S(5)towards the positions shown in Fig.1.The intensities of the H2lines for all positions are shown in Table2.Towards the positions NW1,SW1and the star,we have detected only the S(1)line which prevents from any excitation analysis.The most intense H2emission is found towards the NW and SW PDRs.These positions are spatially coincident with theﬁla-ments observed in H2ﬂuorescent emission which are located in the interface between the atomic and the molecular gas. The H2emission is much weaker towards the HI clump(posi-tion NW1)and the SW1position,suggesting that this region is mainly atomic.In contrast with the spatial distribution ofTable1.ObservationsNW PDR21h01m32s.568◦10′27′′.5 NW121h01m34s.568◦10′05′′.0 Star21h01m36s.968◦09′48′′.3 SW121h01m36s.968◦09′10′′.0 SW PDR21h01m32s.368◦08′45′′.0Table2.Intensitites of the OI and CII linesOI(63µm)7.5±0.1 3.9±0.8 3.8±1.3OI(145.6µm)0.9±0.10.4±0.10.3±0.1CII(157.7µm) 2.8±0.1 1.8±0.1 1.7±0.1OI(63µm)/OI(145.6µm)8.39.012.7OI(63µm)/CII(157.7µm) 2.7 2.2 2.3Lineλ(µm)NW PDR NW1Star SW1SW PDR(x10−5erg s−1cm−2sr−1)Brα 4.051≤2.3≤1.0141.7±1.5≤0.7≤1.0H2S(5) 6.90926.4±3.0≤4.8≤7.1≤3.39.2±3.3H2S(4)8.02515.0±0.2≤3.9≤5.0≤1.4 5.6±1.1H2S(3)9.66540.8±1.3≤4.4≤3.3≤1.621.2±1.5H2S(2)12.27924.0±3.3≤10≤7.4≤8.016.5±2.7H2S(1)17.03521.4±0.7 1.7±0.4 1.8±0.2 1.6±0.410.0±0.7H2S(0)28.219 4.8±4.8≤4.4≤3.5≤3.3≤4.5SIV10.510≤2.3≤2.9≤3.1≤1.8≤1.7FeII25.988≤2.3≤2.0≤1.4≤2.1≤2.2SIII33.480≤3.3≤6.6≤8.3≤1.7≤3.3SiII34.814 3.4±1.3 4.7±0.58.6±0.1 1.8±0.3≤3.41We have used the version number90.04which uses the MICEinterface developed by Henrik Spoon at MPEcan be explained by excitation eﬀects and a diﬀerent deple-tion of Fe and Si on grains.The critical density of the[FeII] 26.0µm line is an order of magnitude larger than that of the [SiII]34.8µm line(Hollenbach&Mc Kee1979).On the other hand,Si is more easily released to the gas phase than Fe.Fe is expected to be signiﬁcantly depleted in the HII region and the PDR although the Si has been released to the gas phase(Soﬁa et al.1994).In standard PDR models,the depletion of Fe is assumed to be a factor of10larger than that of Si(δF e=-2.0) and the intensity of the FeII(26.0µm)lines is then an order of magnitude lower than the intensity of the SiII(34.8µm)line (Burton et al.1992).It is well known that the abundance of gas phase sili-con in the interstellar medium increases by grain-grain colli-sions and/or sputtering in shocks(Mart´ın-Pintado et al.1992, Caselli et al.1997,Bachiller&P´e rez-Gutierrez1997).Pho-todesorption of Si in grain mantles by UV radiation has re-cently been proposed as a mechanism to explain the large SiO abundance in some PDRs(Walmsley et al.1999).Our data put some constraints on the eﬃciency of photodesorption to release Si to the gas phase.In NGC7023,Si is heavily depleted(δSi= -1.3)in gas at an extinction of2mag,i.e.,silicon is mainly in solid form in the molecular gas where the emission of SiO is expected to arise.The large bipolar cavity associated with HD 200775proves the existence of an energetic bipolar outﬂow in a previous stage of the stellar evolution(Fuente et al.1998). The large gas phase Si abundance in the positions closest to the star could be a relique of this stage.If it were the case,the eﬃciency of photodesorption to return the Si to the gas phase would be even lower.4.2.A gradient in the ortho-to-para-H2ratioThe S(0),S(1),S(2),S(3),S(4)and S(5)v=0–0H2rotational lines have been detected towards the NW PDR and the S(1),S(2),S(3),S(4)and S(5)lines towards the SW PDR.In a previous paper(Fuente et al.1999),we have studied in detail the H2rotational lines towards the NW PDR.In this paper we will concentrate in the observations towards the SW PDR.In Fig.5we show the rotational diagram of the H2lines towards the SW PDR.Note that the errors in Fig.5are en-tirely dominated by the calibration uncertainties(15%at6.909µm,25%at8.025µm,25%at9.665µm,25%at12.279%, 20%at17.035µm and30%at28.219µm).The data have been corrected for dust attenuation using the value for the dust opacity derived from the L WS01spectra(A v=0.3mag) and the extinction curve of Draine&Lee(1984).The extinc-tion for the S(0),S(1),S(2),S(3),S(4)and S(5)lines amounts to0.006,0.013,0.013,0.027,0.013and0.006.Like in the case of the NW PDR(see Fuente et al.1999),the rotational di-agram of the SW PDR shows that the ortho-H2levels have systematically lower N u/g u values(where N u and g u are the column densities and degeneracies of the upper levels of the transitions)than the adjacent J-1and J+1para-H2levels pro-ducing a“zig-zag”distribution.In fact,the ortho-levels seem to deﬁne a curve which is oﬀset from that of the para-levels by more than a factor of2(see Fig.5).This oﬀset cannot be ex-plained by calibration uncertainties and extinction eﬀects.The calibration uncertainties are at most30%.Since the values of the extinction at17.03µm,12.3µm and8.025µm are very similar(Draine&Lee1984),a correction for dust attenuation cannot push the S(1),S(2)and S(4)points to the same curve.This”zig-zag”distribution cannot be due to the diﬀerent aper-tures of the instrument for the diﬀerent lines.In the case of a point-like source,the S(1)and S(2)lines should be corrected by a factor1.35relative to the S(3)line and the oﬀset between ortho-and para-curve would increase.The”zig-zag”distribution observed in the rotational dia-gram can only beﬁtted by assuming a non-equilibrium OTP ratio,i.e.,diﬀerent from the OTP equilibrium value at the gas kinetic temperature.In this case,only the rotation temper-ature between the levels of the same symmetry constitute an estimate of the gas kinetic ing the ortho-levels, we have derived a rotation temperature of∼450K between the S(1)and S(3)lines and of∼630K between the S(3)and S(5)lines.For the para-levels we have derived a rotation tem-perature of∼430K using the S(2)and S(4)lines.Then we can conclude that the H2rotational lines are arising in gas with ki-netic temperatures ranging from∼400to700K.These kinetic temperatures are similar to those obtained in the NW PDR, but the intensities of the lines are about a factor of2lower.A change in the incident UVﬁeld and/or the density would imply a variation of both the line ratios(rotation temperatures)and the line intensities.Since the line ratios are similar to those of the NW PDR and the line intensities are signiﬁcantly lower, we conclude that the SW PDR has the same excitation condi-tions as the NW PDR(G0=104,n=106cm−3)but a beam ﬁlling factor of∼0.5.All PDR models which include both, chemical and thermal balance in the gas,assume aﬁxed value of3for the OTP ratio,which is the equilibrium value for gas kinetic temperatures>100K.In order to determine the value of the OTP ratio in the SW PDR,we have corrected the in-tensities predicted by the model of Burton et al.(1992)for the eﬀect of diﬀerent values of the OTP ratio assuming that the total amount of H2molecules at each temperature and the line ratios between the levels of the same symmetry remain unchanged.This correction is only valid if the excitation of the H2lines is mainly collisional which is the expected case for the low rotational transitions.In Fig.6,we compare the observational data for the S(0),S(1),S(2)and S(3)transitions with the model for G0=104,n=106cm−3and aﬁxed OTP ratio of3,and the same model corrected for OTP ratios of1.5 and2.We obtain that the rotation diagram of the SW PDR is wellﬁtted with an OTP ratio of1.5(see Fig.6).Martini et al.(1997)derived an OTP ratio of2.3±0.5towards the SW PDR based on the H2vibrational lines.Because of the opti-cal depth eﬀects in the excitation of the ortho-and para-H2 vibrational lines,this value is just a lower limit to the actual value of the OTP ratio(Draine&Bertoldi1996,Sternberg& Neufeld1999).Then we can conclude that the OTP ratio is larger than2.3±0.5in the less shielded layers of the PDR(A v <0.7mag)where the vibrational lines arises.The gas kinetic temperature in these layers ranges from several hundreds to paring the OTP ratio derived from the rotational lines with that derived from the vibrational lines,we conclude that the OTP ratio increases across the PDR.It takes values ∼1.5at temperatures of∼400-700K and values close to3 at temperatures>∼700K.The same behavior,a non-equilibrium OTP ratio for the gas with kinetic temperatures∼400K and a gradient of the OTP ratio across the PDR was found in the NW PDR by Fuente et al.(1999).They discussed that the most likely expla-nation for this behavior is the existence of an advancing pho-todissociation front.An important problem remains in this ex-planation,the velocity of the photodissociation front required to have a non-equilibrium OTP ratio is too large(∼107n−1 kms−1).However,since the expected behavior of the OTP ratio in an advancing photodissociation front is in agreement with that observed in our data(the cooler gas is further from the OTP equilibrium value than the warm gas)and there are many uncertainties in this estimate(it is not known the OTP ratio at which the H2is formed within the PDR,the gas is clumpy and small dense clumps can be evaporating into the PDR,...), we considered that this was the most likely ter, Lemaire et al.(1999)have obtained a very high spectral resolu-tion spectrum of the H2S(1)v=1→0line at2.121µm towards the NW PDR.The H2line is seen shifted in velocity relative to that of the HI and the CO rotational lines.They argue that this velocity diﬀerence is a clear proof of the presence of dynamical eﬀects in this PDR and estimated a(projected)velocity of1 kms−1for the photodissociation front.So far,very few mod-els have been developed for non-equilibrium PDRs(Bertoldi& Draine1996,St¨o rzer&Hollenbach1998).These models con-sider regions illuminated by O stars in which a ionization front is combined with a photodissociation front and assume aﬁxed value for the OTP ratio of3.They do not give any informa-tion about the eﬀect of an advancing photodissociation on the ortho-to-para-H2ratio.But they predict how the existence of a photodissociation front aﬀects the thermal structure of the region.Following the model by St¨o rzer&Hollenbach(1998), because of the advection of molecular gas through the PDR, the intensities of the H2rotational lines and vibrational lines can be enhanced by a factor of3.Assuming a velocity of1 kms−1for the photodissociation front,the intensities of the H2rotational lines observed in the NW PDR can be wellﬁtted with a density of n∼105cm−1and a UVﬁeld of G0=104.A non-equilibrium OTP ratio has to be assumed to explain the ”zig-zag”distribution.4.3.Absence of molecular emissionTowards NW PDR we have tentatively detected a weak line centered at153.927µm with a S/N ratio of∼3.A weak line centered at153.637µm appears also in the L WS spectrum towards the star position with a S/N ratio of5.Taking into account that the spectral resolution of the L WS at these fre-quencies is0.6µm,both lines can be identiﬁed with the CO J=17→16line(153.267µm).However,since other CO lines have not been detected towards these positions,we cannot dis-card the possibility of a misidentiﬁcation or a spurious detec-tion.If conﬁrmed,this line would implied the existence of a gas with T k>200K and n>105cm−3in the vicinity of the star.However the density of the bulk of the molecular gas in the PDR must be at least an order of magnitude lower.Fe-derman et al.(1997)estimated based on the ultraviolet ab-sorption lines of CH,CH+,C2and CN that the density of the PDR in front of the star is∼200cm−3.Furthermore,the OI(63µm)/CII(157.7µm)ratio is consistent with densities of 104cm−3.We have not detected any OH,CH and CH2line towards the observed positions.Emission of the OH lines have been detected in several sources(NGC7027:Liu et al.1996;R CrA and LkHα234:Giannini et al.1999).ISO-L WS observations towards a sample of11Herbig Ae/Be stars shows that the OH lines are only detected towards the stars associated with with the largest mean densities(n>105cm−3)as derived from the OI(63µm)/CII(157µm)ratio.The non-detection of OH in this nebula is consistent with the idea that the bulk of the molecular gas is formed by the low density component traced by the OI(63µm)/CII(157.7µm)ratio.A similar situation is found for CH and rge CH column densities have been detected in absorption.The data by Federman et al.(1997)shows that the CH column density in front of HD200775is3.2±0.21013cm−2.Even if we assume excitation temperatures as low as T ex=20K,we should have detected intensities of the order of10−9erg s−1cm−2for the 180.93and149.09µm lines.The non-detection of these lines can only be explained invoking the excitation conditions.The beamﬁlling factor of the dense gas(n>105cm−3)is very low. Another possibility is that the CH abundance decreases for high densities.This would also explain the lack of CH emission in other sources.In fact,far to our knowledge,emission of CH has not been detected towards any source(just a tentative detection towards NGC7027).5.Summary:A model for the nebulaIn this paper we present the ISO-SWS and L WS observations towards a strip crossing the star and the most intense PDRs in the reﬂection nebula NGC7023.These observations altogether with previous maps in molecular lines and the[HI]21cm line have allowed toﬁgure out a model of the nebula.We can consider the nebula as composed by three diﬀerent regions.Theﬁrst one is formed by the star,the small HII region around it and the low density gasﬁlling the cavity.Federman et al.(1997)based on studies of absorption lines derived that the density of the gas in front of the star is of a few102cm−3.A slightly larger value,n=103cm−3,was found by Gerin et al. (1998)based on observations of the CO rotational lines towards the star.The Brαand[SiII]34.8µm lines peak in this region but the[HI]21cm and H2rotational lines have a minimum in emission.The Brαemission is well explained as arising in the stellar ionized wind modeled by Nisini et al.(1995).However, the contribution of the stellar wind to the intensity of the[SiII] 34.8µm line is expected to be negligible.The[SiII]34.8µm line is better explained as arising in the diﬀuse gasﬁlling the cavity in which at least20%–30%of the silicon is in gas phase.The second region is formed by the walls of the cavity in which the star is immersed.Clumpy PDRs have been formed in these walls with high densityﬁlaments(n∼105–106cm−3) immersed in a more diﬀuse medium(n∼104cm−3).These high densityﬁlaments are located almost symmetrically∼50′′NW and∼70′′SW(NW PDR and SW PDR)from the star and constitute the peaks of the[CII]157.7µm,[OI]63.2and 145.6µm,[HI]21cm and H2rotational lines emission.The NW and SW PDRs have very similar excitation conditions although the SW PDR is a factor of2weaker than the NW PDR.This diﬀerence in intensity is very likely due to a diﬀerent beam ﬁlling factor.In both,the NW and SW PDRs,the intensities of the H2rotational lines can only beﬁtted assuming an ortho-to-para-H2ratio lower than3.Since the rotation temperatures of the H2lines ranges from400to700K,this implies the existence of a non-equilibrium OTP ratio in the region.The comparison between the vibrational and pure rotational H2lines suggests that the OTP ratio increases across the PDR from values as low as∼1.5for gas with temperatures∼400–700K to values close to3in the gas located at a visual extinction A v<0.7 mag(T k>∼700K).This non-equilibrium OTP ratio has beeninterpreted in terms of an advancing photodissociation front. The intensity of the[SiII]34.8µm line observed towards the NW PDR is consistent with the predictions of PDR models with a standard silicon depletion ofδSi=-1.3,i.e.,only5% of the silicon is in gas phase in these PDRs.The third region is the molecular gas.We have not detected the OH,CH and CH2molecular lines towards this source.This is consistent with the weakness of these lines in other sources and reveals that the beamﬁlling factor of the dense gas(n≥105cm−3)in the L WS aperture is low.The CO J=17→16line has tentatively been detected towards the star. Acknowledgements.We would like to thank the referee F. Bertoldi for his helpful comments and suggestions.We are also grateful to J.Black for fruitful discussions on the chem-istry of this region.This work has been partially supported by the Spanish DGES under grant number PB96-0104and the Spanish PNIE under grant number ESP97-1490-E.N.J.R-F acknowledges Consejer´ıa de Educaci´o n y Cultura de la Co-munidad de Madrid for a pre-doctoral fellowship. ReferencesBachiller R.P´e rez-Gutierrez M.,1997,ApJ487,L93 Bertoldi F&Draine B.T.,1996,ApJ458,222Burton M.G.,Hollenbach D.J.,Tielens A.G.G.M.,1992,ApJ 399,563Caselli P.,Hartquist T.W.,Havnes O.,1997,A&A322,296 Clegg et al.,1996,A&A315,L38Chokshi A.,Tielens A.G.G.M.,Werner M.W.,Castelaz M.W., 1988,ApJ334,803Draine B.T.&Lee H.M.,1984ApJ285,89Draine B.T.&Bertoldi F.,1996ApJ468,269Hollenbach D.&McKee C.F.,1979,ApJ342,306 Federman S.R.,Knauth D.C.,Lambert D.L.,Anderson B.-G., 1997,ApJ489,758Ferland G.J.,1996,Hazy,a Brief Introduction to Cloudy,Uni-versity of Kentucky,Department of Physics and Astronomy, Internal Report.Fuente A.,Mart´ın-Pintado J.,Cernicharo J.,Brouillet N.,Du-vert G.,1992,A&A237,471Fuente A.,Mart´ın-Pintado J.,Cernicharo J.,Bachiller R., 1993,A&A276,473Fuente A.,Mart´ın-Pintado J.,Neri R.,Rogers C.,Moriarty-Schieven G.,1996,A&A310,286Fuente A.&Mart´ın-Pintado J.,1997,ApJ477,L107 Fuente A.,Mart´ın-Pintado J.,Rodr´ıguez-Franco A.,Moriarty-Schieven G.D.,1998,A&A339,575Fuente A.,Mart´ın-Pintado J.,Rodr´ıguez-Fern´a ndez,N.J., Rodr´ıguez-Franco A.,de Vicente P.,Kunze D.,1999,ApJ 518,L45Gerin M.,Phillips T.G.,Keene J.,Betz A.L.,Boreiko R.T., 1998,ApJ500,329Giannini T.,Lorenzetti D.,Tommasi E.,et al.,1999,A&A 346,617de Graauw et al.,1996,A&A315,L49Kaufman M.J.,Wolﬁre M.G.,Hollenbach D.,Luhman M.L., 1999,preprintKessler M.F.et al.,1996,A&A315,L27Kurucz,R.L.,1979,ApJS40,1Lemaire J.L.,Field D.,Gerin M.,Leach S.,Pineau des ForetsG.,Rostas F.,Rouan D.,1996,A&A308,895Lemaire J.L.,Field D.,Maillard J.P.,Pineau des Forets G., Falgarone E.,Pijpers F.P.,Gerin M.,Rostas F.,1999,A&A 349,253Liu X.-W.,Barlow M.J.,Nguyen-Q-Rieu,et al.1996,A&A 315,L257Lorenzetti D.,Tommasi E.,Giannini T.et al.,1999,A&A346, 604Mart´ın-Pintado J.,Bachiller R.,Fuente A.,1992,A&A254, 315Martini P.,Sellgren K.,Hora J.L.,1997,ApJ484,296Nisini B.,Milillo A.,Saraceno P.,Vitali F.,1995,A&A302, 169Rogers C.,Heyer M.H.,Dewdney P.E.,1995,ApJ442,694 Salama A.et al.,1997,ESA SP-419,17Sellgren K.,Werner M.W.,Dinerstein H.L.,1992,ApJ400,238 Skinner S.L.,Brown A.,Stewart R.T.,1993,ApJS87,217 Soﬁa U.,Cardelli J.A.,Savage B.D.,1994,ApJ430,650 Sternberg A.&Dalgarno A.,1995,ApJS99,565Sternberg A.&Neufeld D.,1999,ApJ516,371van Steenberg M.E.&Shull J.M.,1988,ApJ330,942St¨o rzer H.&Hollenbach D.,1998,ApJ495,853Swinyard B.M.et al.,1996,A&A315,L43Walmsley C.M.,Pineau de Forˆe ts G.,Flower D.R.,1999,ApJ 342,542This article was processed by the author using Springer-Verlag L a T E X A&A styleﬁle1990.Fig.1.Scheme of the observed region.The HI total column density map(contours)is superposed on the integrated intensity map of the 13CO J=1→0map(grey scale)(Fuente et al.1998).Theﬁlaments observed in H2ﬂuorescent emission are also shown(Lemaire et al.1996). The observed positions are indicated byﬁlled squares and the star.。