Research Article Modeling the Impacts of Boreal Deforestation on...

Hindawi Publishing CorporationAdvances in MeteorologyVolume 2013 Article ID 486962 9 pageshttpdxdoiorg1011552013486962

Research ArticleModeling the Impacts of Boreal Deforestation onthe Near-Surface Temperature in European Russia

Zhihui Li123 Xiangzheng Deng13 Qingling Shi13 Xinli Ke4 and Yingcheng Liu5

1 Institute of Geographic Sciences and Natural Resources Research Chinese Academy of Sciences Beijing 100101 China2University of Chinese Academy of Sciences Beijing 100049 China3 Center for Chinese Agricultural Policy Chinese Academy of Sciences Beijing 100101 China4College of Land Management Huazhong Agricultural University Wuhan 430070 China5 Faculty of Resource and Environmental Science Hubei University Wuhan 430062 China

Correspondence should be addressed to Xiangzheng Deng dengxzccapgmailcom

Received 9 July 2013 Revised 2 September 2013 Accepted 26 September 2013

Academic Editor Burak Guneralp

Copyright copy 2013 Zhihui Li et alThis is an open access article distributed under the Creative CommonsAttribution License whichpermits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Boreal deforestation plays an important role in affecting regional and global climate In this study the regional temperature variationinduced by future boreal deforestation in European Russia boreal forest region was simulated based on future land cover changeand the Weather Research and Forecasting (WRF) model This study firstly tested and validated the simulation results of the WRFmodelThen the land cover datasets in different years (2000 as baseline year 2010 and 2100) was used in theWRFmodel to explorethe impacts of boreal deforestation on the near-surface temperature The results indicated that the WRF model has good abilityto simulate the temperature change in European Russia The land cover change in European Russia boreal forest region whichwill be characterized by the conversion from boreal forests to croplands (boreal deforestation) in the future 100 years will lead tosignificant change of the near-surface temperatureThe regional annual temperature will decrease by 058∘C in the future 100 yearsresulting in cooling effects to some extent and making the near-surface temperature decrease in most seasons except the spring

1 Introduction

According to the fourth assessment report of Intergov-ernmental Panel on Climate Change (2007 IPCC AR4)there is a probability of more than 90 percent that humanactivities have affected the climate [1] mainly through twoapproaches fossil fuel burning and land cover change Thereis a consensus among the scientists that fossil fuel burningcan lead to increase in the greenhouse gas concentration inthe atmosphere and further results in the global warmingwhile the impacts of land cover change on the climate systemat the local regional and global scales have become one ofthe research hotspots Terrestrial land cover is an importantcomponent of the climate system It is the most direct sourcenot only of the atmospheric heat but also of the atmosphericmoisture Therefore land cover change will directly affectthe surface-atmosphere interactions and further influencethe atmospheric thermodynamic characteristics The landuse activities significantly changed the regional land cover

and exerted great impacts on the climate system at regionalscale including temperature evapotranspiration precipita-tion wind and air pressure The impacts of land coverchange on climate can be divided into two major categoriesthat is biogeochemical and biogeophysical impacts [2] Thebiogeochemical processes mainly refer to greenhouse gasemissions caused by the land cover change changing thegas composition of the atmosphere and thereby affectingthe climate The biogeophysical processes directly affectthe physical parameters that determine the absorption anddisposition of energy at the earthrsquos surface

Global deforestation plays an important role in regulat-ing climate through biogeophysical effects [3] At presentdeforestation is one of the most important types of landcover change and the global land cover change associatedwith deforestation has always been taken as the main reasonfor the climate change in different regions The albedofor forests is relatively lower than that for grasslands or

2 Advances in Meteorology

croplands thus forests can absorb more solar radiation [4]And the transpiration of forests is more exuberant than thatof herbaceous plants in the growing season and will releasemore latent heat Forest cover change may cause the changeof the albedo and evapotranspiration and consequently leadsto climate changeThese effects on climate are not the same atdifferent latitudes Tropical forests have strong evapotranspi-ration deforestation in tropical forests such as the Amazonregion will lead to lower evaporate rates and make the localclimate become drier and warmer [5ndash8] Deforestation athigh latitudes that is deforestation in boreal forests hasdifferent effects on the climate compared with that in tropicalregions [3 9 10]

There have beenmany researches focusing on the impactsof the high-latitude boreal deforestation on climate changeAs boreal forest is the largest continuous terrestrial ecosystemin the world boreal forest has the potential to influencethe climate by altering the radiation budget Bonan foundthat loss of boreal forests provided a positive feedback forglaciation whereas boreal forest expansion during the mid-Holocene amplified warming [3] Bathiany et al clarifiedthat in the next 100 years the deforestation at the northernlatitudes (45∘Nndash90∘N) will lead a decrease of 025∘C inglobal annual mean temperature while the afforestation hadequally large warming effects combining both biochemicaland biophysical effects [9] And through latitude-specificlarge-scale deforestation experiment Bala et al revealed thatthe difference of the global average temperature between thestandard case without deforestation and the experiment withboreal deforestation at high latitude in year 2100 is minus08∘C[11] Based on observations Lee et al found that for thesite pairs at 45∘N the mean annual temperature differenceinduced by deforestation is 085 plusmn 044∘C (plusmnmean standarddeviation) [12] Overall most researches clarified that athigher latitudes boreal deforestation will result in coolingeffects due to the boreal forests with lower albedo beingreplaced by other types of vegetation with higher albedosuch as crops and grasslands [13] Boreal deforestation willlead to a large increase in albedo especially in winter [14]the alteration of albedo is a major factor driving climatechange which will alter surface heat balance Also borealdeforestation can alter surface heat balance by alteringevaporative heat transfer caused by evapotranspiration fromvegetation and by changes in surface roughness [15] Unlikethe tropical forests removal of the boreal forest vegetation hasa larger effect that resulted from the strong albedo feedbackthan from the evapotranspiration change on the surfaceradiative balance In boreal forest region the considerableincrease in albedo due to deforestation will result in coolingeffects since the albedo plays a more important role thanthe vegetation transpiration [16ndash18] Replacing the forestvegetation with other types of vegetation or bare ground thatwill be snow covered increases the albedo considerably Theland surface responds by absorbing less net radiation as moreincoming solar radiation is reflected from the surfaceThe airtemperature at the surface will cool considerably as there isless energy absorbed at the surface [19]

As to the selection of climate models and experimentdesigns to detect the climate impact of deforestation current

researches on the impacts of boreal forests change on climatechange mainly focus on large-scale experiments based onglobal or regional climate models with assumption thatthe boreal forests are being replaced by other types ofvegetation or bare lands [10 19] while only a few researchesimplemented simulation of the climate effects of future landcover change based on the scenario analysis with a mesoscalenumerical model The earliest researches on the regionalclimate effects generally used the global circulation models(GCMs) and carried out the sensitivity test with the force-response method [20 21] that is representing the land usechanges with the changes of the land surface parameters(eg albedo roughness and evapotranspiration)The GCMshave been widely used in the study of climate changehowever their coarse resolution is inappropriate to reveal theland surface-atmosphere interactions for simulating regionalclimate variability especially for complex terrain so there aresome bias and uncertainties in the simulation of the regionalclimate change with GCMs [22 23] Then regional climatemodels (RCMs) were developed during the late 1980s andthe early 1990s [24] and have become an important toolin the regional climate simulations among which WeatherResearch and Forecasting (WRF)model represents the recentadvances of RCMs that combine the expertise and experiencefor mesoscale meteorology and land-surface and climatescience developed over the last several decades [25]TheWRFmodel was specifically designed for high resolution applica-tions and provides an ideal tool for assessing the value of highresolution regional climate modeling and some researchershave identified theWRF to be superior to other RCMs [23 2627] In addition although the experiments with assumptionthat the boreal forest be completely replaced by other typesof vegetation can help understand the impact mechanismof boreal deforestation experiments based on land coverscenarios can relatively be related to real human activities andprovide reference for future land use management

European Russia with high coverage of boreal forest hasgone through intensive human activities The largest partof the boreal forests is located in Russia and about halfof the boreal forests are still primary with very limitedimpacts from forestry and other human activities Whilein Scandinavia and western Russia (European Russia) theboreal forests are most intensely managed and influenced byhuman beings with only patches of old-growth forests remainin reserves Some researchers have shown that EuropeanRussia has gone through fluctuant forest cover change [28ndash30] Baumann et al found that the temperate forests inEuropean Russia underwent substantial change during 1990ndash2010 with a decrease rate of 1 in total area between 1990and 1995 and an increase rate of 14 between 2005 and 2010which may be caused by the forest regrowth on abandonedfarmlands [28] Hansen et al reported that Russia has thethird largest area of gross forest cover loss (GFCL) RussiarsquosGFCL is geographically widespread due to deforestation inthe European and far-eastern parts of the country and forestfires throughout Siberia [29] Potapov et al indicated thatthe forest cover in the central part of European Russia wasbetween 16 and 50 (average forest cover of 36) The lowforest cover within these regions is a result of a long history

Advances in Meteorology 3

01

0

Accu

mul

ated

frac

tion

of co

nver

sion

135∘W 150

∘W 165∘W

70∘N

50∘N 30

∘N

45∘E 60

∘E 75∘E 90

∘E

Figure 1 Accumulated fraction of conversion from forests tocroplands between 2000 and 2100 in the study areaThe black box isthe boundary of the study area

of conversion of forests to croplands The forest cover is thelowest (3ndash14) in eleven administrative regions located in theforest-steppe interface where the expansion of croplands wascoupled with the natural forest fragmentation [30]

Thus in this study the European Russia that has expe-rienced fluctuant land cover change and serious forest lossdue to the intensive human activities was selected as a typicalarea to detect the impacts of deforestation on the near-surfacetemperature (Figure 1) And the near-surface temperature inEuropean Russia was simulated with the Weather Researchand Forecasting Model (WRF) on the basis of the land coverdatasets of years 2000 2010 and 2100 which were derivedfrom the scenarios of land cover change The results cancontribute to the understanding of biogeophysical effects ofboreal forest change on the climate and provide constructivesuggestions on the future climate mitigation and land usemanagement of European Russia

2 Data and Methodology

21 Data Processing The data used in this study include landcover data atmospheric forcing data and meteorologicalobservation data The accuracy and resolution of land coverdata are significant to the climate simulation [31] In thisstudy the 1 km resolution land cover data of the USGSclassification system in year 2000 derived from USGS LandCover Institute which include 24 land cover types wereused as baseline data The land cover data of years 2010 and2100 were predicted with the data of land conversion amongdifferent land cover types The land conversion data werederived from land use scenario that was simulated with themodel that combines Asia-Pacific Integrated Model with theglobal economy model (AIMCGE) based on RepresentativeConcentration Pathways 6 (RCP6) In the AIMCGE modelland was treated as a production factor for agriculture live-stock forestry and biomass energy production Urban landuse increased due to population and economic growth while

croplands area expanded due to increasing food demandTheland cover data and underlying surface change data during1500ndash2100 can be obtained through data fusion at 05∘times 05∘resolution [32]

As there is a difference in the spatial resolution and classi-fication types between the land cover data of the USGS classi-fication system and the RCP-based land conversion data itis necessary to reclassify the land conversion data to theUSGS classification and convert the spatial resolution to beof higher resolution (1ndash10 km) as the requirement of theWRFmodel As it is well acknowledged that RCMs with spatialresolution at or coarser than 30 km are unable to produceaccurate climate forecasts [33] Higher resolution allowedthe model to include the regional features and predict theregional climate with more accuracy For example WRFsimulations can be done at a resolution of 4 km which allowsmany small scale features such as mountains coastlines andother land use categories to be represented more accuratelywhich is important for our purposes [34] In this study weset the resolution of the land cover data to be 5 km in theWRF model Taking the processing of land cover data ofyear 2100 as an example first the accumulated fraction ofdifferent kinds of land conversion in each 05∘times05∘ grid from2000 to 2100 was calculated (Figure 1) Then the dominantconversion type (with maxima conversion amount) of eachgrid was identified and thereafter whether the land covertype of grids changed or not was identified through settingthreshold value of the conversion rate The threshold valueswere mainly set to reveal the conversion trend as to eachtype of conversion the threshold value was set to be the 50thpercentile of the conversion rateThe land cover data in years2010 and 2100 were further obtained on the basis of the landcover data of the USGS classification in year 2000 and thedata of land conversion during 2000ndash2010 and 2000ndash2100and finally these underlying surface data were transformedto grid data of 5 km times 5 km through resampling

As to the atmospheric forcing data the fifth phase of theCoupled Model Intercomparison Project (CMIP5) producesa state-of-the-artmultimodel dataset designed to advance ourknowledge of climate variability and climate change [35 36]The model output which is being analyzed by researchersworldwide underlies the Fifth Assessment Report of IPCCIt provides projections of future climate change on two timescales near term (out to about 2035) and long term (outto 2100 and beyond) Model output of RCP6 such as airtemperature specific humidity sea level pressure eastwardwind northward wind and geopotential height from 2000 to2100 was used as the atmospheric forcing dataset of theWRFmodel And the meteorological observation data which wereused to comparewith the simulated temperature in this studywere derived from European Climate Assessment amp Dataset(httpecaknminldailydatapredefinedseriesphp)

22 The WRF Model and Scenario-Based Experiment Design

221WRFModel With the development of the climatemod-els and land surface process models the numerical sim-ulation has become a widely used approach to study theinfluence of land cover change on climate The WRF model


is a next generation mesoscale numerical weather predictionsystem that has been developed as a collaborative effortby the National Center for Atmospheric Research (NCAR)the National Oceanic and Atmospheric Administration theNational Centers for Environmental Prediction (NCEP) andthe Forecast Systems Laboratory (FSL) theAir ForceWeatherAgency (AFWA) the Naval Research Laboratory the Univer-sity of Oklahoma and the Federal Aviation Administration(FAA) It is a state-of-the-art atmospheric simulation systembased on the Fifth-Generation Penn StateNCAR MesoscaleModel (MM5) [37] This mesoscale model has been widelyused in the simulation of the global climate [38ndash40] andregional climate [41] In this paper the WRF model basedon the Eulerian mass solver was used to analyze the impactsof land cover change on the near-surface temperature in thestudy area In a WRF simulation each grid point has a landcover type based on the land cover dataset being used for themodel runThe properties (surface albedo surface emissivitymoisture availability and surface roughness length) of eachland cover type depend on the land surface model used in theWRF runThe land surfacemodel is the component that takescare of the processes involving land-surface interactionsFor the WRF runs the parameterization scheme of physicalprocesses in the model was set USGS classification datasetwas used to specify land cover types and their properties Tosimulate land cover change conditions all the land cover datawere projected based on scenarios

222 Experiment Design The Advanced Research WRF(ARW-WRF Edition 33) was used in this studyThe Lambertprojection was used with the two standard parallels bothbeing 57∘N and central meridian 48∘EThe spatial resolutionwas set to be 5 km The study area is located in EuropeanRussia and it contains 192 grid points in the east-westdirection and 174 grid points in the north-south direction

The parameterization scheme of physical processes in themodel (Table 1) is as follows The Microphysics parameter-ization Scheme adopted the scheme introduced by Lin etal [42] The cumulus parameterization scheme adopted theGrell-Devenyi ensemble schemeThe boundary layer processscheme was Yonsei University (YSU) scheme The long-waveradiation scheme and short-wave radiation scheme bothwerethe Community Atmosphere Model (CAM) scheme and theland surface process scheme was Noah land surface modelThe boundary buffer was set to be 4 layers of grid points andthe boundary conditions adopted the relaxation schemeThetime interval of the model integration was set to be 5 minutesand that of the radiation process and cumulus convectionwas30 minutes and 5 minutes respectively There were 27 layersin the vertical direction and the atmospheric pressure at thetop layer was 50 hPa

The simulation schemes in this study are as follows(Table 2) The simulation was implemented with the landcover data in three years as the underlying surface Theland cover dataset of year 2000 (Figure 2) with the UnitedStates Geological Surveyrsquos (USGS) classification system wasused as the baseline underlying surface data in this studyThe land use datasets of years 2010 and 2100 were based onthe land usecover conversion information that was derived

Table 1 Parameterization scheme of physical processes in themodel

Classification of schemes Scheme optionMicrophysics parameterizationscheme

Bulk microphysics schemesintroduced by Lin et al [42]

Cumulus parameterization scheme Grell-Devenyi ensembleBoundary layer process scheme YSULong-wave radiation scheme CAM long-wave RadiationShort-wave radiation scheme CAM short-wave RadiationLand Surface process scheme Noah land surface model

Table 2 Design of the simulation scheme

Simulation Period offorcing data

Land cover data used in theWRF model

Control simulationyear 2000 as thebaseline year

2000 Land cover data of year2000 (USGS Classification)

Sensitivity simulationI year 2010 2010

Land cover data of year2010 (USGS ClassificationBased on Scenario)

Sensitivity simulationII year 2100 2100


from land use scenario which was simulated with AIMCGEbased on RCP6 [32] Then the underlying surface datasetsof each year were used as the input data of the WRF modelto simulate the impact of land cover change on the climatechange The simulations in three years were all implementedwith the meteorological forcing data as displayed in Table 2Themonthly and seasonal simulation results were comparedFirstly the simulation results of monthly temperature in year2010 were used to validate the ability of the WRF modelto simulate the temperature change in the study area thenspatial difference between the near-surface temperature inthe winter of years 2000 and 2010 was analyzed at last thecomparison between the monthly near-surface temperaturesof years 2010 and 2100 was implemented

23 Land Cover Change in the Study Area of EuropeanRussia The main land cover types in the study area are thecroplands (four types of croplands in USGS classificationwith ldquoDryland Cropland and Pasturerdquo (DCP) and ldquoCroplandWoodland Mosaicsrdquo (CWM) dominating) and forests (fourtypes of forests in USGS classification with ldquoEvergreenNeedleleaf Forestrdquo (ENF) and ldquoMixed Forestrdquo (MF) dominat-ing) accounting for about 53 and 44 of the total land arearespectively The conversion types in the region are mainlydominated by the conversion between croplands and forests

From 2000 to 2010 the land cover conversion is domi-nated by the conversion from croplands to forests (mainlyfrom CWM type to MF type in the northern part of thestudy area) and conversion from forests to croplands (mainlyfrom MF type to DCP type in the southern part of thestudy area) in different parts of the study area Statisticsof the dominant conversion type of each grid show that


Urban and built-up landDryland cropland and pastureIrrigated cropland and pastureMixed drylandirrigated cropland and pastureCroplandgrassland mosaicCroplandwoodland mosaicGrasslandShrublandMixed shrublandgrasslandSavannaDeciduous broadleaf forestDeciduous needleleaf forestEvergreen broadleaf forestEvergreen needleleaf forestMixed forestWater bodiesHerbaceous wetlandWooded woodlandBarren or sparsely vegetatedHerbaceous tundraWooded tundraMixed tundraBare ground tundraSnow or ice

60∘09984000 N

55∘09984000 N

60∘09984000 N

55∘09984000998400998400N

55∘09984000 E50

∘09984000 E45

∘09984000 E

55∘09984000 E50

∘09984000 E45

∘09984000 E40

∘09984000 E

998400998400

998400998400 998400998400 998400998400 998400998400

998400998400

998400998400

998400998400998400998400998400998400

Figure 2 Land cover of the study area in year 2000

grids dominated by the conversion from croplands to forestsaccount for 2618 of all the grids in the study area andgrids dominated by the conversion from forests to croplandsaccount for 2780 of all the grids Besides integrated withother types of land conversion the percentage of forests in thetotal land area will decrease from 4440 to 4407 whilethat of the croplands will increase from 5304 to 5337The conversion between the forests and croplands will bein balance overall during 2000ndash2010 However the spatialheterogeneity of the conversion between them will lead tothe change of their spatial distribution For example thecroplands in the northern part of study area have a tendencyto be converted to forests while the scattered forests in thesouthern part tend to be converted to croplands (Figure 3)

According to the statistics of conversion type gridsthe conversion from forests to croplands will dominate inthe study area by 2100 The grids presenting the conversion

60∘09984000 N

55∘09984000 N

60∘09984000 N

55∘09984000998400998400N

55∘09984000 E50

∘09984000 E45

∘09984000 E

55∘09984000 E50

∘09984000 E45

∘09984000 E40

∘09984000 E

998400998400

998400998400 998400998400 998400998400 998400998400

998400998400

998400998400

998400998400998400998400998400998400



from forests to croplands as the primary conversion type willaccount for 374 of all the gridsThe croplands coverage willaccount for 7227 of the total land area while the forestscoverage will decrease to 2531 (Figure 4) It is indicatedthat there is a strong tendency of conversion from forests tocroplands in this region in the future 100 years

The coverage percentage of different land cover type inyears 2000 2010 and 2100 which was calculated based on theaccount of grids of each land cover type is shown in Table 3

3 Results

31 Validation of the Simulation Result To assess the abilityof the WRF model to simulate the temperature change inthe study area we compared the simulated values of monthlyaverage temperature with the ground-based observed val-ues of year 2010 First the daily average temperature was


60∘09984000 N

55∘09984000 N

60∘09984000 N

55∘09984000998400998400N

55∘09984000 E50

∘09984000 E45

∘09984000 E

55∘09984000 E50

∘09984000 E45

∘09984000 E40

∘09984000 E

998400998400

998400998400 998400998400 998400998400 998400998400

998400998400

998400998400

998400998400998400998400998400998400



simulated based on the WRF model then it was used tocalculate monthly average temperature to compare it withthe ground-based observed data As shown in Figure 5the change trends of both simulated and observed monthlyaverage temperatures in the study area were relatively similarand the difference between the observed and simulatedvalues generally fluctuates around zero To examine whetherthe difference was significant or not a paired t-test wasconducted to test the difference between simulated values andobserved values The null hypothesis was that there was nosignificant difference between those two samples and the Pvalue of the paired t-test was 09146 at the significance levelof 005 indicating that there was no significant differencebetween the simulated values and observation values Thusit can be identified that the WRF model has the ability tosimulate the temperature change

0

1

2

05

1015202530

1 2 3 4 5 6 7 8 9 10 11 12

Difference between observed and

simulation

Simulationobserved

Simulation temperature of year 2010Observed temperature of year 2010

minus5

minus10

minus15

minus20

minus1

minus2

minus3

minus4

temperature

temperature

Observed temperature minus simulation temperature

(∘C)

(∘C)

Figure 5 Comparison of simulated and observed values of themonthly average near-surface temperature at 2 meters above theground of the study area

0

002

004

006

008

01

012

014

016

018

023

60∘09984000 N

55∘09984000 N55

∘09984000998400998400N

60∘09984000 N998400998400

40∘09984000 E998400998400

45∘09984000 E998400998400

50∘09984000 E998400998400

55∘09984000 E998400998400

998400998400

998400998400

55∘09984000 E998400998400

50∘09984000 E998400998400

45∘09984000 E998400998400

Figure 6 Difference of simulated monthly average temperature inthe winter between year 2000 and 2010 (unit ∘C) in the study area

32 Analysis of Temperature Change As a result of the con-version between croplands and forests from 2000 to 2010the simulatedmonthly average temperature during thewinter(December January and February) increased in most of theparts of the study area compared with that in year 2000(Figure 6) The spatial pattern of the temperature changeduring winter corresponded to the land cover and land coverchange The average temperature in the winter generallyincreased in the regions which are mainly covered by crop-lands and where the boreal forest expands In the southernpart of the study area which is mainly covered by DCPtype cropland the forests were converted to croplands andthe temperature increment was relatively higher The averagetemperature increment generally declines as the distance toregions mainly covered by the forests decreases In generalthe land cover change in the study area is almost in balancefrom 2000 to 2010 there was very slight change in the average


Table 3 Coverage percentage of different land cover type ()

USGS land cover classification Coverage percentage2000 2010 2100

(1) Urban and built-up land 065 059 062(2) Dryland cropland and pasture 3107 2859 3221(3) Irrigated cropland and pasture 000 000 000(5) Croplandgrassland mosaic 107 078 101(6) Cropwoodland Mosaic 2090 2400 3905Sum of cropland 5304 5337 7227(7) Grassland 018 012 003(8) Shrub land 001 001 001(10) Savanna 012 007 000(11) Deciduous broadleaf forest 024 006 001(12) Deciduous broadleaf forest 027 013 002(14) Evergreen needleleaf forest 987 854 676(15) Mixed forest 3385 3533 1851(16) Water bodies 174 174 174Sum of Forests 4422 4407 2530(18) Wooded wetland 000 000 000(19) Barren or sparsely vegetated 002 002 001(21) Wooded tundra 000 000 000(22) Mixed tundra 001 001 001Total 100 100 100

temperature in the winter (no more than 0023∘Cyear) andthe spatial distribution of temperature change corresponds toland cover and land cover change

The land cover change in the study area in year 2100is mainly characterized by the conversion from forests tocroplands and it will lead to change of the near-surfacetemperature The annual temperature will decline while themonthly average temperature will increase from Februaryto June and decrease from July to January (Figure 7) Inthe boreal forest region the temperature change mainlyresults from the change of albedo due to snow masking [9]As snow covers the surface and boreal forests are convertedto croplands the albedo of the land surface will increaseand the net surface solar radiation will reduce thus it willlead to the cooling effect which offsets the warming effectdue to the decrease of evaporation-transpiration Thereforenear-surface temperature during the winter will change mostintensively decreasing by 181∘C on average In the northernhemisphere solar radiation begins to strengthen in June andbecomes the strongest around July and August The coniferforests (needle-leaf forests) have lower evapotranspirationrate (defined as ratio of latent heat flux to available energy)than the deciduous broadleaf forests in the summer whichcan lead to the higher rates of sensible heat flux [3] From 2010to 2100 most evergreen needleleaf forests will be convertedto croplands and thus the cooling effect due to the albedoincrease is also stronger than the warming effect due to thedecrease of evapotranspiration in the previous needle-leafforest area during July and August and consequently makesthe average temperature decrease by 030∘C While in thespring snowwillmelt and the solar radiation is not very high

the warming effect due to the evapotranspiration reductionof evergreen needleleaf forests cannot be offset by the coolingeffect due to the increase of surface albedo and consequentlythe monthly average temperature will increase by 086∘C inthe spring In general the result indicated that the high albedoresulting from deforestation of boreal forests has negativeimpacts on the temperature that is boreal deforestation willmake the temperature decrease especially during the snowseason

4 Conclusions and Discussions

In this paper the WRF model was used to study the impactsof boreal deforestation on the near-surface temperature inthe European Russia Based on the analysis of the land coverdata of years 2000 2010 and 2100 it was shown that theland cover change in the study area was mainly characterizedby the conversion between forests and croplands During2000ndash2010 the croplands increased by 033 and the forestsdecreased by 015 while other land cover types changedslightly The conversion from croplands to forests mainlyappeared in the northern part which was previously domi-nated by forests while the conversion from scattered foreststo croplands mainly appeared in the southern part which wasmainly dominated by croplands By year 2100 there will besignificant land cover change the percentage of forests in thetotal land area will decrease from 4422 to 253 indicatingthat about half of the forests will be coveted to croplands

The results indicate that the simulation result with theWRF model can match the monthly average temperaturechange with the observed temperature in the study area


0

5

10

15

20

25

30

minus5

minus10

minus15

minus20 minus3

minus25

minus2

minus15

minus1

minus05

0

05

1

15

1 2 3 4 5 6 7 8 9 10 11 12

Temperature difference

between year 2010 and

Simulation

Simulation temperature of year 2010 (TEM2010)Simulation temperature of year 2100 (TEM2100)

2100

temperature (∘C)

(∘C)

Temperature difference TEM2100 minus TEM2010

Figure 7 Comparison of simulatedmonthly average temperature at2 meters above the ground in year 2010 and year 2100 in the studyarea

The land cover change in the study area which was mainlycharacterized by the conversion between boreal forests andcroplands will lead to significant change of the near-surface temperature especially in year 2100 It will makethe regional near-surface temperature decrease by 058∘Cin future 100 years (00058∘Cyear on average) with thetemperature change varying greatly in different seasons Thetemperature changes most drastically in the winter with anaverage decrease of 181∘C And the temperature will decreaseby 030∘C in the summer while in the spring it will increaseby 087∘C The temperature change is mainly due to theincrease of surface albedo caused by boreal deforestationwhich offsets the warming effect due to evapotranspirationreduction of forests

The results of this study basically reveal the impactsof boreal deforestation on regional climate changes whichare closely related to human activities but there are stillsome uncertainties about the influence of the future landcover change since only biogeophysical effects were takeninto consideration in the WRF model without consideringthe biochemical effects through some other key influencingfactors such as carbon emission Since the impacts of landcover change are very complex it is still necessary to carryout more indepth researches on a series of issues such asthe improvement of the climate model and combination ofbiochemical and biophysicalmodels which take both the bio-geophysical and biogeochemical processes into account Inaddition there are still some aspects that should be improvedto more accurately detect the impact of deforestation onclimate changes First it is still necessary to further explorethe data integration between the land cover data of USGSclassification and the RCP-based land conversion data sincethere is some difference between them Although the maintrend of land conversion can be identified there still existsome uncertainties about its impacts on the future climate

change Second there are many factors that may influencethe climate change and it is necessary to do some sensitivityexperiments to detect the main factors and minimize theuncertainty

Conflict of Interests

The authors declare that they have no conflict of interestsregarding the publication of this paper

Acknowledgments

This research was supported by the Key Project funded by theChinese Academy of Sciences (no KZZD-EW-08) NationalKey Programme for Developing Basic Science in China (no2010CB950900) and the External Cooperation Program ofthe Chinese Academy of Sciences (no GJHZ1312)

References

[1] S Solomon D Qin M Manning et al Climate Change 2007The Physical Science Basis Cambridge University Press NewYork NY USA 2007

[2] J J Feddema K W Oleson G B Bonan et al ldquoAtmosphericscience the importance of land-cover change in simulatingfuture climatesrdquo Science vol 310 no 5754 pp 1674ndash1678 2005

[3] G B Bonan ldquoForests and climate change forcings feedbacksand the climate benefits of forestsrdquo Science vol 320 no 5882pp 1444ndash1449 2008

[4] G B Bonan Ecological Climatology Concepts and ApplicationsCambridge University Press 2002

[5] M H Costa and J A Foley ldquoCombined effects of deforestationand doubled atmospheric CO2 concentrations on the climate ofAmazoniardquo Journal of Climate vol 13 no 1 pp 18ndash34 2000

[6] R E Dickinson and A Henderson-Sellers ldquoModelling tropicaldeforestation a study of GCM landsurface parametrizationsrdquoQuarterly Journal vol 114 no 480 pp 439ndash462 1988

[7] N Gedney and P J Valdes ldquoThe effect of Amazonian defor-estation on the northern hemisphere circulation and climaterdquoGeophysical Research Letters vol 27 no 19 pp 3053ndash3056 2000

[8] A Henderson-Sellers R E Dickinson T B Durbidge P JKennedy K McGuffie and A J Pitman ldquoTropical deforesta-tion modeling local- to regional-scale climate changerdquo Journalof Geophysical Research vol 98 no 4 pp 7289ndash7315 1993

[9] S Bathiany M Claussen V Brovkin T Raddatz and V GaylerldquoCombined biogeophysical and biogeochemical effects of large-scale forest cover changes in the MPI earth system modelrdquoBiogeosciences vol 7 no 5 pp 1383ndash1399 2010

[10] E L Davin and N de Noblet-Ducoudre ldquoClimatic impact ofglobal-scale Deforestation radiative versus nonradiative pro-cessesrdquo Journal of Climate vol 23 no 1 pp 97ndash112 2010

[11] G Bala K Caldeira M Wickett et al ldquoCombined climate andcarbon-cycle effects of large-scale deforestationrdquo Proceedings ofthe National Academy of Sciences of the United States of Americavol 104 no 16 pp 6550ndash6555 2007

[12] X Lee M L Goulden D Y Hollinger et al ldquoObserved increasein local cooling effect of deforestation at higher latitudesrdquoNature vol 479 no 7373 pp 384ndash387 2011

[13] R A Pielke Sr G Marland R A Betts et al ldquoThe influence ofland-use change and landscape dynamics on the climate system


relevance to climate-change policy beyond the radiative effectof greenhouse gasesrdquo Philosophical Transactions of the RoyalSociety A vol 360 no 1797 pp 1705ndash1719 2002

[14] P J Lawrence and T N Chase ldquoInvestigating the climateimpacts of global land cover change in the community climatesystemmodelrdquo International Journal of Climatology vol 30 no13 pp 2066ndash2087 2010

[15] E Ellis and R Pontius ldquoLand-use and land-cover changerdquoEncyclopedia of Earth Environmental Information Coali-tion National Council for Science and the EnvironmentWashington DC USA 2007 httpwwweoearthorgarticleLanduse and land-cover change

[16] G B Bonan D Pollard and S L Thompson ldquoEffects of borealforest vegetation on global climaterdquo Nature vol 359 no 6397pp 716ndash718 1992

[17] V Brovkin A Ganopolski M Claussen C Kubatzki and VPetoukhov ldquoModelling climate response to historical land coverchangerdquoGlobal Ecology and Biogeography vol 8 no 6 pp 509ndash517 1999

[18] B Govindasamy P B Duffy andK Caldeira ldquoLand use changesand Northern Hemisphere coolingrdquo Geophysical Research Let-ters vol 28 no 2 pp 291ndash294 2001

[19] P K Snyder C Delire and J A Foley ldquoEvaluating the influenceof different vegetation biomes on the global climaterdquo ClimateDynamics vol 23 no 3-4 pp 279ndash302 2004

[20] R A Pielke Sr R Avissar M Raupach A J Dolman XZeng andA S Denning ldquoInteractions between the atmosphereand terrestrial ecosystems influence on weather and climaterdquoGlobal Change Biology vol 4 no 5 pp 461ndash475 1998

[21] I S Kang K Jin B Wang et al ldquoIntercomparison of the cli-matological variations of Asian summermonsoon precipitationsimulated by 10 GCMsrdquo Climate Dynamics vol 19 no 5-6 pp383ndash395 2002

[22] V Brovkin M Claussen E Driesschaert et al ldquoBiogeophysicaleffects of historical land cover changes simulated by six Earthsystem models of intermediate complexityrdquo Climate Dynamicsvol 26 no 6 pp 587ndash600 2006

[23] E T Yu H J Wang and J Q Sun ldquoA quick report on adynamical downscaling simulation over China using the nestedmodelrdquo Atmospheric and Oceanic Science Letters vol 3 no 6pp 325ndash329 2010

[24] H Von Storch E Zorita and U Cubasch ldquoDownscaling ofglobal climate change estimates to regional scales an applica-tion to Iberian rainfall in wintertimerdquo Journal of Climate vol 6no 6 pp 1161ndash1171 1993

[25] J M Jin N L Miller and N Schlegel ldquoSensitivity study offour land surface schemes in the WRF modelrdquo Advances inMeteorology vol 2010 Article ID 167436 11 pages 2010

[26] P Caldwell H-N S Chin D C Bader andG Bala ldquoEvaluationof a WRF dynamical downscaling simulation over CaliforniardquoClimatic Change vol 95 no 3-4 pp 499ndash521 2009

[27] C V Srinivas D Hariprasad D V Bhaskar Rao Y AnjaneyuluR Baskaran and B Venkatraman ldquoSimulation of the Indiansummer monsoon regional climate using advanced researchWRF modelrdquo International Journal of Climatology vol 33 no5 pp 1195ndash1210 2013

[28] M Baumann M Ozdogan T Kuemmerle K J WendlandE Esipova and V C Radeloff ldquoUsing the Landsat record todetect forest-cover changes during and after the collapse of theSovietUnion in the temperate zone of EuropeanRussiardquoRemoteSensing of Environment vol 124 pp 174ndash184 2012

[29] M C Hansen S V Stehman and P V Potapov ldquoQuantificationof global gross forest cover lossrdquo Proceedings of the NationalAcademy of Sciences of the United States of America vol 107 no19 pp 8650ndash8655 2010

[30] P Potapov S Turubanova and M C Hansen ldquoRegional-scaleboreal forest cover and change mapping using Landsat datacomposites for European Russiardquo Remote Sensing of Environ-ment vol 115 no 2 pp 548ndash561 2011

[31] FWu J Y ZhanHM Yan C C Shi and JHuang ldquoLand covermapping based onmultisource spatial datamining approach forclimate simulation a case study in the farming-pastoral ecotoneof North Chinardquo Advances in Meteorology vol 2013 Article ID520803 12 pages 2013

[32] G C Hurtt L P Chini S Frolking et al ldquoHarmonizationof land-use scenarios for the period 1500ndash2100 600 years ofglobal gridded annual land-use transitions wood harvest andresulting secondary landsrdquo Climatic Change vol 109 no 1 pp117ndash161 2011

[33] J Jin and N L Miller ldquoAnalysis of the impact of snow ondaily weather variability in mountainous regions using MM5rdquoJournal of Hydrometeorology vol 8 no 2 pp 245ndash258 2007

[34] R Mawalagedara and R J Oglesby ldquoThe climatic effects ofdeforestation in south and southeast Asiardquo DeforestationAround the World 2011

[35] N S Keenlyside M Latif J Jungclaus L Kornblueh and ERoeckner ldquoAdvancing decadal-scale climate prediction in theNorth Atlantic sectorrdquo Nature vol 453 no 7191 pp 84ndash882008

[36] K E Taylor R J Stouffer and G A Meehl ldquoAn overview ofCMIP5 and the experiment designrdquo Bulletin of the AmericanMeteorological Society vol 93 no 4 pp 485ndash498 2012

[37] M A Hernandez-Ceballos J A Adame J P Bolivar and B ADe LaMorena ldquoAmesoscale simulation of coastal circulation inthe Guadalquivir valley (southwestern Iberian Peninsula) usingtheWRF-ARWmodelrdquoAtmospheric Research vol 124 pp 1ndash202013

[38] J Chen P Zhao H Liu and X Guo ldquoModeling impactsof vegetation in western China on the summer climate ofnorthwestern ChinardquoAdvances in Atmospheric Sciences vol 26no 4 pp 803ndash812 2009

[39] A Garcia-R T Schoenemeyer A Jazcilevich D G Ruiz-Suarez and V Fuentes-Gea ldquoImplementation of the multiscaleclimate chemistry model (MCCM) for Central Mexicordquo inProceedings of the International Conference on Air Pollution pp71ndash78 July 2000

[40] A D Jazcilevich A R Garcıa and L G Ruız-Suarez ldquoAmodeling study of air pollution modulation through land-usechange in the Valley of Mexicordquo Atmospheric Environment vol36 no 14 pp 2297ndash2307 2002

[41] M Mohan and S Bhati ldquoAnalysis of WRF model performanceover subtropical region of Delhi Indiardquo Advances in Meteorol-ogy vol 2011 Article ID 621235 13 pages 2011

[42] Y L Lin R D Farley andH D Orville ldquoBulk parameterizationof the snow field in a cloud modelrdquo Journal of Climate andApplied Meteorology vol 22 no 6 pp 1065ndash1092 1983

Submit your manuscripts athttpwwwhindawicom

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ClimatologyJournal of

EcologyInternational Journal of


EarthquakesJournal of


Hindawi Publishing Corporationhttpwwwhindawicom

Applied ampEnvironmentalSoil Science

Volume 2014

Mining


Journal of

Hindawi Publishing Corporation httpwwwhindawicom Volume 2014

International Journal of

Geophysics

OceanographyInternational Journal of


Journal of Computational Environmental SciencesHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal ofPetroleum Engineering


GeochemistryHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

Atmospheric SciencesInternational Journal of


OceanographyHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Advances in


MineralogyInternational Journal of


MeteorologyAdvances in

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Paleontology JournalHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

ScientificaHindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Geological ResearchJournal of


Geology Advances in


croplands thus forests can absorb more solar radiation [4]And the transpiration of forests is more exuberant than thatof herbaceous plants in the growing season and will releasemore latent heat Forest cover change may cause the changeof the albedo and evapotranspiration and consequently leadsto climate changeThese effects on climate are not the same atdifferent latitudes Tropical forests have strong evapotranspi-ration deforestation in tropical forests such as the Amazonregion will lead to lower evaporate rates and make the localclimate become drier and warmer [5ndash8] Deforestation athigh latitudes that is deforestation in boreal forests hasdifferent effects on the climate compared with that in tropicalregions [3 9 10]

There have beenmany researches focusing on the impactsof the high-latitude boreal deforestation on climate changeAs boreal forest is the largest continuous terrestrial ecosystemin the world boreal forest has the potential to influencethe climate by altering the radiation budget Bonan foundthat loss of boreal forests provided a positive feedback forglaciation whereas boreal forest expansion during the mid-Holocene amplified warming [3] Bathiany et al clarifiedthat in the next 100 years the deforestation at the northernlatitudes (45∘Nndash90∘N) will lead a decrease of 025∘C inglobal annual mean temperature while the afforestation hadequally large warming effects combining both biochemicaland biophysical effects [9] And through latitude-specificlarge-scale deforestation experiment Bala et al revealed thatthe difference of the global average temperature between thestandard case without deforestation and the experiment withboreal deforestation at high latitude in year 2100 is minus08∘C[11] Based on observations Lee et al found that for thesite pairs at 45∘N the mean annual temperature differenceinduced by deforestation is 085 plusmn 044∘C (plusmnmean standarddeviation) [12] Overall most researches clarified that athigher latitudes boreal deforestation will result in coolingeffects due to the boreal forests with lower albedo beingreplaced by other types of vegetation with higher albedosuch as crops and grasslands [13] Boreal deforestation willlead to a large increase in albedo especially in winter [14]the alteration of albedo is a major factor driving climatechange which will alter surface heat balance Also borealdeforestation can alter surface heat balance by alteringevaporative heat transfer caused by evapotranspiration fromvegetation and by changes in surface roughness [15] Unlikethe tropical forests removal of the boreal forest vegetation hasa larger effect that resulted from the strong albedo feedbackthan from the evapotranspiration change on the surfaceradiative balance In boreal forest region the considerableincrease in albedo due to deforestation will result in coolingeffects since the albedo plays a more important role thanthe vegetation transpiration [16ndash18] Replacing the forestvegetation with other types of vegetation or bare ground thatwill be snow covered increases the albedo considerably Theland surface responds by absorbing less net radiation as moreincoming solar radiation is reflected from the surfaceThe airtemperature at the surface will cool considerably as there isless energy absorbed at the surface [19]

As to the selection of climate models and experimentdesigns to detect the climate impact of deforestation current

researches on the impacts of boreal forests change on climatechange mainly focus on large-scale experiments based onglobal or regional climate models with assumption thatthe boreal forests are being replaced by other types ofvegetation or bare lands [10 19] while only a few researchesimplemented simulation of the climate effects of future landcover change based on the scenario analysis with a mesoscalenumerical model The earliest researches on the regionalclimate effects generally used the global circulation models(GCMs) and carried out the sensitivity test with the force-response method [20 21] that is representing the land usechanges with the changes of the land surface parameters(eg albedo roughness and evapotranspiration)The GCMshave been widely used in the study of climate changehowever their coarse resolution is inappropriate to reveal theland surface-atmosphere interactions for simulating regionalclimate variability especially for complex terrain so there aresome bias and uncertainties in the simulation of the regionalclimate change with GCMs [22 23] Then regional climatemodels (RCMs) were developed during the late 1980s andthe early 1990s [24] and have become an important toolin the regional climate simulations among which WeatherResearch and Forecasting (WRF)model represents the recentadvances of RCMs that combine the expertise and experiencefor mesoscale meteorology and land-surface and climatescience developed over the last several decades [25]TheWRFmodel was specifically designed for high resolution applica-tions and provides an ideal tool for assessing the value of highresolution regional climate modeling and some researchershave identified theWRF to be superior to other RCMs [23 2627] In addition although the experiments with assumptionthat the boreal forest be completely replaced by other typesof vegetation can help understand the impact mechanismof boreal deforestation experiments based on land coverscenarios can relatively be related to real human activities andprovide reference for future land use management

European Russia with high coverage of boreal forest hasgone through intensive human activities The largest partof the boreal forests is located in Russia and about halfof the boreal forests are still primary with very limitedimpacts from forestry and other human activities Whilein Scandinavia and western Russia (European Russia) theboreal forests are most intensely managed and influenced byhuman beings with only patches of old-growth forests remainin reserves Some researchers have shown that EuropeanRussia has gone through fluctuant forest cover change [28ndash30] Baumann et al found that the temperate forests inEuropean Russia underwent substantial change during 1990ndash2010 with a decrease rate of 1 in total area between 1990and 1995 and an increase rate of 14 between 2005 and 2010which may be caused by the forest regrowth on abandonedfarmlands [28] Hansen et al reported that Russia has thethird largest area of gross forest cover loss (GFCL) RussiarsquosGFCL is geographically widespread due to deforestation inthe European and far-eastern parts of the country and forestfires throughout Siberia [29] Potapov et al indicated thatthe forest cover in the central part of European Russia wasbetween 16 and 50 (average forest cover of 36) The lowforest cover within these regions is a result of a long history


01

0

Accu

mul

ated

frac

tion

of co

nver

sion

135∘W 150

∘W 165∘W

70∘N

50∘N 30

∘N

45∘E 60

∘E 75∘E 90

∘E


































60∘09984000 N

55∘09984000 N

60∘09984000 N

55∘09984000998400998400N

55∘09984000 E50

∘09984000 E45

∘09984000 E

55∘09984000 E50

∘09984000 E45

∘09984000 E40

∘09984000 E

998400998400

998400998400 998400998400 998400998400 998400998400

998400998400

998400998400

998400998400998400998400998400998400




60∘09984000 N

55∘09984000 N

60∘09984000 N

55∘09984000998400998400N

55∘09984000 E50

∘09984000 E45

∘09984000 E

55∘09984000 E50

∘09984000 E45

∘09984000 E40

∘09984000 E

998400998400

998400998400 998400998400 998400998400 998400998400

998400998400

998400998400

998400998400998400998400998400998400





3 Results



60∘09984000 N

55∘09984000 N

60∘09984000 N

55∘09984000998400998400N

55∘09984000 E50

∘09984000 E45

∘09984000 E

55∘09984000 E50

∘09984000 E45

∘09984000 E40

∘09984000 E

998400998400

998400998400 998400998400 998400998400 998400998400

998400998400

998400998400

998400998400998400998400998400998400




0

1

2

05

1015202530

1 2 3 4 5 6 7 8 9 10 11 12


simulation

Simulationobserved


minus5

minus10

minus15

minus20

minus1

minus2

minus3

minus4

temperature

temperature


(∘C)

(∘C)


0

002

004

006

008

01

012

014

016

018

023

60∘09984000 N

55∘09984000 N55

∘09984000998400998400N

60∘09984000 N998400998400

40∘09984000 E998400998400

45∘09984000 E998400998400

50∘09984000 E998400998400

55∘09984000 E998400998400

998400998400

998400998400

55∘09984000 E998400998400

50∘09984000 E998400998400

45∘09984000 E998400998400














0

5

10

15

20

25

30

minus5

minus10

minus15

minus20 minus3

minus25

minus2

minus15

minus1

minus05

0

05

1

15

1 2 3 4 5 6 7 8 9 10 11 12



Simulation


2100

temperature (∘C)

(∘C)








Acknowledgments


References






















































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in


01

0

Accu

mul

ated

frac

tion

of co

nver

sion

135∘W 150

∘W 165∘W

70∘N

50∘N 30

∘N

45∘E 60

∘E 75∘E 90

∘E


































60∘09984000 N

55∘09984000 N

60∘09984000 N

55∘09984000998400998400N

55∘09984000 E50

∘09984000 E45

∘09984000 E

55∘09984000 E50

∘09984000 E45

∘09984000 E40

∘09984000 E

998400998400

998400998400 998400998400 998400998400 998400998400

998400998400

998400998400

998400998400998400998400998400998400




60∘09984000 N

55∘09984000 N

60∘09984000 N

55∘09984000998400998400N

55∘09984000 E50

∘09984000 E45

∘09984000 E

55∘09984000 E50

∘09984000 E45

∘09984000 E40

∘09984000 E

998400998400

998400998400 998400998400 998400998400 998400998400

998400998400

998400998400

998400998400998400998400998400998400





3 Results



60∘09984000 N

55∘09984000 N

60∘09984000 N

55∘09984000998400998400N

55∘09984000 E50

∘09984000 E45

∘09984000 E

55∘09984000 E50

∘09984000 E45

∘09984000 E40

∘09984000 E

998400998400

998400998400 998400998400 998400998400 998400998400

998400998400

998400998400

998400998400998400998400998400998400




0

1

2

05

1015202530

1 2 3 4 5 6 7 8 9 10 11 12


simulation

Simulationobserved


minus5

minus10

minus15

minus20

minus1

minus2

minus3

minus4

temperature

temperature


(∘C)

(∘C)


0

002

004

006

008

01

012

014

016

018

023

60∘09984000 N

55∘09984000 N55

∘09984000998400998400N

60∘09984000 N998400998400

40∘09984000 E998400998400

45∘09984000 E998400998400

50∘09984000 E998400998400

55∘09984000 E998400998400

998400998400

998400998400

55∘09984000 E998400998400

50∘09984000 E998400998400

45∘09984000 E998400998400














0

5

10

15

20

25

30

minus5

minus10

minus15

minus20 minus3

minus25

minus2

minus15

minus1

minus05

0

05

1

15

1 2 3 4 5 6 7 8 9 10 11 12



Simulation


2100

temperature (∘C)

(∘C)








Acknowledgments


References






















































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in
























60∘09984000 N

55∘09984000 N

60∘09984000 N

55∘09984000998400998400N

55∘09984000 E50

∘09984000 E45

∘09984000 E

55∘09984000 E50

∘09984000 E45

∘09984000 E40

∘09984000 E

998400998400

998400998400 998400998400 998400998400 998400998400

998400998400

998400998400

998400998400998400998400998400998400




60∘09984000 N

55∘09984000 N

60∘09984000 N

55∘09984000998400998400N

55∘09984000 E50

∘09984000 E45

∘09984000 E

55∘09984000 E50

∘09984000 E45

∘09984000 E40

∘09984000 E

998400998400

998400998400 998400998400 998400998400 998400998400

998400998400

998400998400

998400998400998400998400998400998400





3 Results



60∘09984000 N

55∘09984000 N

60∘09984000 N

55∘09984000998400998400N

55∘09984000 E50

∘09984000 E45

∘09984000 E

55∘09984000 E50

∘09984000 E45

∘09984000 E40

∘09984000 E

998400998400

998400998400 998400998400 998400998400 998400998400

998400998400

998400998400

998400998400998400998400998400998400




0

1

2

05

1015202530

1 2 3 4 5 6 7 8 9 10 11 12


simulation

Simulationobserved


minus5

minus10

minus15

minus20

minus1

minus2

minus3

minus4

temperature

temperature


(∘C)

(∘C)


0

002

004

006

008

01

012

014

016

018

023

60∘09984000 N

55∘09984000 N55

∘09984000998400998400N

60∘09984000 N998400998400

40∘09984000 E998400998400

45∘09984000 E998400998400

50∘09984000 E998400998400

55∘09984000 E998400998400

998400998400

998400998400

55∘09984000 E998400998400

50∘09984000 E998400998400

45∘09984000 E998400998400














0

5

10

15

20

25

30

minus5

minus10

minus15

minus20 minus3

minus25

minus2

minus15

minus1

minus05

0

05

1

15

1 2 3 4 5 6 7 8 9 10 11 12



Simulation


2100

temperature (∘C)

(∘C)








Acknowledgments


References






















































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in


60∘09984000 N

55∘09984000 N

60∘09984000 N

55∘09984000998400998400N

55∘09984000 E50

∘09984000 E45

∘09984000 E

55∘09984000 E50

∘09984000 E45

∘09984000 E40

∘09984000 E

998400998400

998400998400 998400998400 998400998400 998400998400

998400998400

998400998400

998400998400998400998400998400998400




0

1

2

05

1015202530

1 2 3 4 5 6 7 8 9 10 11 12


simulation

Simulationobserved


minus5

minus10

minus15

minus20

minus1

minus2

minus3

minus4

temperature

temperature


(∘C)

(∘C)


0

002

004

006

008

01

012

014

016

018

023

60∘09984000 N

55∘09984000 N55

∘09984000998400998400N

60∘09984000 N998400998400

40∘09984000 E998400998400

45∘09984000 E998400998400

50∘09984000 E998400998400

55∘09984000 E998400998400

998400998400

998400998400

55∘09984000 E998400998400

50∘09984000 E998400998400

45∘09984000 E998400998400














0

5

10

15

20

25

30

minus5

minus10

minus15

minus20 minus3

minus25

minus2

minus15

minus1

minus05

0

05

1

15

1 2 3 4 5 6 7 8 9 10 11 12



Simulation


2100

temperature (∘C)

(∘C)








Acknowledgments


References






















































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in












0

5

10

15

20

25

30

minus5

minus10

minus15

minus20 minus3

minus25

minus2

minus15

minus1

minus05

0

05

1

15

1 2 3 4 5 6 7 8 9 10 11 12



Simulation


2100

temperature (∘C)

(∘C)








Acknowledgments


References






















































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in









































Volume 2014

Mining


Journal of



Geophysics







Journal of




Advances in











Geology Advances in

Research Article Modeling the Impacts of Boreal Deforestation on...

Documents

Transcript of Research Article Modeling the Impacts of Boreal Deforestation on...