| Forest growth and afforestation act as the major carbon sinks of terrestrial ecosystems.Large afforestation project contributes to maintain the carbon balance of forests and alteres local energy budgets,which has the potential to offer feedback on local and regional climates.China has the largest planted forest area in the world.Understanding afforestation and its forest biomass pattern as well as its biophysical effects can help to reasonably distribute and sustainably utilize forest resources and evaluate China's carbon storage of the planted forests.Currently,the evaluation of forest biomass is mainly based on tranditional field measurement or single remote sensing data source?low,medium and high resolution?-based short-term inversion methods.Moreover,the combination of physical models or biochemical models is usually not suitable for popularization due to the models'complexities,thus,the labor-saving and cost-effective methods to map yearly forest aboveground biomass remain rare.Additionally,the afforestation?planted forest?distribution is related to carbon accounting and maintains a close relationship with forest biomass.The distribution of afforestation?planted forests?comes from the forest resource survey report compiled by the State Forestry Administration,however,its data availability or timeliness need to be tested.How to develop continuous and consistent afforestation data suitable for the regional research relates to the rationality of forest biomass and carbon assessment,and afforestaion reasonably responding to climate change mechanism?land surface temperature?.An optical,SAR and national forest inventory?NFI?based forest aboveground biomass mapping algorithm was developed and validated in northern Guangdong province first in this work.Based on the algorithm,further improvements were made to construct a new method by integrating optical,SAR,spaceborne Lidar and NFI to map wall-to-wall forest AGB in the entire Guangdong province,followed by a mapping performance comparative analysis of the two algorithms.Ultimately,the relationship between afforestation and forest AGB,the surface biophysical effects including albedo and evaportranspiration of plantation forests?afforestation and reforestation?,and the influence of afforestation on surface temperature were clarified in this work.The major research contents and conclusions were summarized as follows:We estimated forest AGB from 1990 to 2011 in northern Guangdong,China,based on a spatially explicit dataset derived from six years of national forest inventory?NFI?plots,Landsat time series imagery?1986–2011?and Advanced Land Observing Satellite?ALOS?Phased Array L-band Synthetic Aperture Radars?PALSAR?25 m mosaic data?2007–2010?.Optical-SAR model was developed for modeling and validating in combination with spectral variables and derivatives by using the random forests algorithm.The root mean square error?RMSE?of plot-level validation was between 6.44 and 39.49?t/ha?,the normalized root-mean-square error?NRMSE?was between 7.49%and 19.01%and the mean absolute error?MAE?was between5.06 and 23.84 t/ha.The highest coefficient of determination R2 of 0.8 and the lowest NRMSE of7.49%were reported in 2006.A clear increasing trend of the mean AGB from the lowest value of 13.58 t/ha to the highest value of 66.25 t/ha was witnessed between 1988 and 2000,while after2000 there was a fluctuating ascending change,with a peak mean AGB of 67.13 t/ha in 2004.By integrating AGB change with forest disturbance histories mapped by VCT algorithm,the trend in disturbed area closely corresponded with the trend in AGB decrease.To determine the driving forces of these changes,the correlation analysis and exploratory factor analysis?EFA?method were used to find a factor rotation that maximizes this variance and represents the dominant factors of nine climate elements and nine human activities elements affecting the AGB dynamics.Overall,human activities contributed more to short-term AGB dynamics than climate data.Harvesting and human-induced fire in combination with rocky desertification and global warming made a strong contribution to AGB changes.We mapped wall-to-wall AGB data?from 1986–2016?in the entire Guangdong province by combining the forest inventories and multisource remotely sensed data,including the Ice,Cloud,and Land Elevation Satellite data and Landsat dense time series imagery,and L-band Synthetic Aperture Radar?PALSAR?mosaic data in Guangdong,China on the basis of the optical-SAR model Durning the modeling,we used random forest?RF?and stochastic gradient boosting?SGB?algorithms to determine the optimal variables of the statistical models for mapping and validation of the AGB purpose.Our results showed that the Geoscience Laser Altimeter System?GLAS?-based AGB correlated well(R2adj=0.89,n=277,p<0.001,RMSE=21.24 t/ha)with those obtained using the field-based method that used an RF-based approach,although inevitably,there is a saturation problem.The combined remotely sensed optical and radar imagery and ancillary data sets for mapping AGB using the RF algorithm yielded a stronger(R2adj=0.86,n=558,p<0.001,RMSE=11.35 t/ha)linear correlation with those produced using the GLAS waveform data than that produced using the SGB algorithm.The accuracy of RF-GLAS-based model was superior to the above optical-SAR model.Additionally,the total amount of AGB had increased from 1986 to 2016 by 55.9%.The same increasing trend was observed for total AGB in both mid-subtropical?from 42%to 62%?and south-subtropical?from 38%to 57%?evergreen broadleaved forests,whereas a decreasing trend was witnessed in the tropical forest,particularly after 2010.There was an upward trend of total AGB among the four economic zones of Guangdong;the mountainous area had the highest AGB value distribution,accounting for58%–70%,followed by the Pearl River Delta region?20%–30%?,the western coast of Guangdong?3%–9%?,and the eastern coast of Guangdong?2%–7%?.By integrating phenological variables and climatic variables,the annual,seasonal and spatio-temporal distribution of forest aboveground AGB were obtained.By comparing RF-GLAS model,optical-SAR model and NFI data,the results found that the evaluation results of RF-GLAS model had the highest AGB value,which can be explained by the different definition of forest from the remote sensing and NFI perspectives.Combining with different spectrum and texture measurements of PALSAR data,and Landsat phenology variables is for accurate identification of forest and non-forest,as well as afforestation and planted forests,and understanding the carbon sink effect of afforestation and the biophysical effect on climate.We identified forest and non-forest preliminarily through different machine learning algorithms including decision tree?C50?,support vector machine?SVM?,random forests?RF?and stochastic gradient boosting?SGB?based on PALSAR mosaic data.SGB classification was the most effective one among the four methods.Next,combining Landsat phenology variables?cumulative NDVI maximum?was used to further identify forest and non-forest,finally,the SGB-NDVI classification model based on the stochastic gradient algorithm and the maximum NDVI dry season data was generated.Then,in order to describe the biological physical change mechanism of the planted forests?afforestation and reforestation?and their effects on the land surface temperature.After the aggregation of the planted forests,natural forests,cropland and grassland in Guangdong province,grid pixels were selected to help identify the difference in annual land surface temperature?daytime/nighttime?,evapotranspiration and black sky albedo data between planted forests and the natural forests,or cropland,grassland,and implement the planted forests for analyzing land surface temperature feedback.Results showed that the classification validation results of 2007,2010 and 2016,with the overall accuracy up to85%and the Kappa coefficient up to 0.58.By comparison with field survey data and existing land cover product,we found that the overall accuracy of this study was better than that of Japan PALSAR/FNF products,and remains unchanged or slightly less accurate than those of global land cover 30 m product and vegetation change tracking?VCT?model product in a small area of verification in Guangdong province.Despite some prediction errors,SGB-NDVI classification model can effectively identify the distribution of afforestation and reforestation events,and generate a reasonable distribution map of afforestation.The trend of afforestation changes closely conformed to the trend of AGB.Both afforestation area and AGB value were the highest in1991-1995,and both of them were the lowest between 2006 and 2010.The AGB trend of the afforestation region of every five years indicated that afforestation and forest AGB peaked at p122r043 covering area between 1991-1995 and 2001-2005,and reached a low value between1996–2000 and 2006–2010.The p121r043 covering area reached a low value between 2006 and2010,peaked in 1991–1995,and then showed certain fluctuations.Overall,these variations were closely related to the changes of the afforestation strategies and urbanization.Moreover,the remote sensing based forest age was obviously linearly related to forest AGB increase.Furthermore,the effects of the planted forests on land surface temperature was explored based on the MODIS land surface data and the biophysical mechanism of the planted forests,the results showed that the humid climate in Guangdong Province made the surface temperature cooling,and the daytime and nighttime temperature had a cooling effect,especially in summer.Findings from this study will provide a basis for exploring the performance of forest AGB change detection in subtropical and tropical regions,which can help policymakers and the remote-sensing community understand forest biomass carbon stocks associated with afforestation and deforestation under the impacts of forest disturbance and climate change,and a new direction was opened for the exploration of large area biomass mapping by using multi-source data.Afforestation has a feedback impact on the climate change in Guangdong province and provides instructions for future afforestation projects in southern China or other places,also contributes to provide scientific reference for the balance of global climate. |
PDF Full Text Request