Spatial And Temporal Variations Of Urban Vegetation And Soil Carbon Storage

Urbanization together with climate change has become the biggest problems worldwide,and carbon emissions in urban areas are sharply increasing.Urban vegetation as the most important green infrastructures has received considerable interests in mitigating carbon emissions amongst scientists and urban planners.The quantifications of urban vegetation carbon storage and its spatial and temporal variations were important to precisely estimate and evaluate its carbon sinks functions,and be good for the scientific and rational planning of vegetation arrangements.This study,we used the combination of one GF-1 images with 2 m resolution and 5 series Landsat TM/OLI remote sensing images with 30 m resolution,field studies in 2014 with 271 plots,laboratory analysis,and Invest model to estimate total amount of carbon stocks in urban forests,wetlands,and cultivated lands.We try to explore the spatial variations of carbon stocks in urban forests and possible influencing factors,and to be clear of carbon stocks changes during urban sprawl and driving forces.The main conclusions are as follows:(1)Characteristics of urban vegetation in Harbin.The total green coverage of forestlands,wetlands,and cultivated lands within forth ring road of Harbin was about 210 km2,which takes 35.5% of total area in Harbin.The green coverage increased along with ring roads,which showed an increasing trend along urban-rural gradient.Urban forests in Harbin were rich in tree species,and a total of 131 tree species were identified,which belongs to 33 families and 69 genera.The average DBH was 17.8 cm,average tree height was 8.7 m,average canopy diameter was 50 m2,and average under branch height was 2.3 m.Middle mature and young forests were the main body of urban forests in Harbin.Tree densities were averaged for 902 tree·ha-1 and increased along ring roads.Urban wetlands in Harbin were mainly beach land,shoal,islands,wetland parks,paddy fields that mainly covered by Salix spp.,Phragmites australias,Typha orientalis,Echinochloa crusgalli,Polygonum,and Miscanthus.(2)Urban vegetation carbon densities and associations with community structure characteristic parameters.Averaged tree and soil C stock density(C stocks per unit tree cover)for Harbin city were 7.71(±7.69)kgC·m-2 and 5.48(±2.86)kgC·m-2,respectively.They were higher than those of other Chinese cities(Shenyang and Changchun),but were much lower than local natural forests.The tree C stock densities varied 2.3-to 3.2-fold among forest types,administrative districts,and ring road-based urban-rural gradients,and these variations were correlated with species numbers and Shannon-wiener index.The richer the tree species composition,the lower the tree carbon stocks was.In comparison,soil organic C(SOC)densities varied by much less(1.4 to 1.5 folds),and were urbanization dependent,which were closely related to the urban-rural gradient data based on ring-roads and settlement history patterns.A city’s ability to provide C-related ecosystem services increases as it ages,especially soils.The accumulation rate of SOC in soils was 15.4 g·C·m-2·year-1.(3)Urban vegetation carbon stocks and associations with landscape patterns.Small patches were the main body of urban forest in Harbin.Mean patch size of forest was 0.31 ha,and patches that cover an area of less than 0.5 ha took 72% of all patches.Urban forests in Harbin city are in a much higher fragmentation level than local cities or other cities in the world.Landscape metrics in different forest types,urban-rural gradients or administrative districts were highly different and these variations were strongly associated with carbon variations.Total area(TA)of forest landscape was the key factor related with total amounts of trees and soils carbon storage(ton),and in low tree cover regions,the most effective way to promote carbon storage was to increase TA.Relations between carbons tocks per unit area(ton ha-1)and landscape metrics were much more instructive.Strong and positive associations were found between mean patch size(AREA_MN,p < 0.05)and tree biomass carbon density,and big patch was more conductive to tree carbon storage.Soil carbon density was negatively related with landscape shape index(LSI,p < 0.01),and LSI was significantly increased along with ring-road related(p < 0.05)and urban settlement history related urban-rural gradients.The increased LSI and dispersion and shape complex of patches in rural regions,may impact carbon storage in forest soils.(4)Temporal changes of urban vegetation carbon stocks from 1985 to 2014.During 30 years urbanization,about 65% and 15% of newly emerging urban lands were converted from cultivated lands and wetlands,two of the fast decreasing land covers during urban sprawl,while landscape of forestlands was increased along with urbanization.Carbon density in forestlands was highest in all land covers(139.26 tons ha-1),and was 2.2 and 1.8 folds higher than cultivated lands and wetlands respectively.The total carbon stocks in Harbin decreased 0.16 Tg during 30 year urban sprawl,of which,forestlands increased 0.54 Tg,and cultivated lands and wetlands decreased 0.70 Tg and 0.30 Tg respectively.Carbon stocks differences between cultivated lands,wetlands and urban lands and the sharply shrink of wetlands and cultivated lands during urbanization were the main reason of carbon losses in Harbin,and carbon gains in forestlands could compensate this carbon losses to some degree.Soil carbon stocks took over 60% of total carbon stocks,and total carbon losses were mainly occurred in soils.Proportions of carbon stocks in aboveground biomass and forest lands were all negatively correlated with Euclidean Nearest-Neighbor Distance(ENN_MN)of forestlands,and total carbon stocks was negatively correlated with mean patch size(AREA_MN)of urban lands.The main conclusions of this study could not only help to improve the scientific understanding and estimation precision of carbon storage in urban vegetation,but also conductive to the understanding the positive or negative effects of urbanization on biomass and soil carbon storage.Urbanization process accompanied with carbon storage accumulation in forest soils,and losses of carbon storage in urban areas,particularly in wetlands and cultivated lands.Associations between carbon storage and landscape metrics could provide some information about the promotion of carbon storage in urban area.Possible strategies include: the decreased distance between forest patches and increased patch cohesion,decreased mean patch size of urban lands and increased green patches in urban land to split large and linked urban patches.These measures may mitigate carbon losses during urban sprawl in urban areas.