Soil carbon stock research has gained prominence in environmental studies amidst climate change concerns,especially given that soil is one of the largest terrestrial carbon reserves.Accurate predictions necessitate co...Soil carbon stock research has gained prominence in environmental studies amidst climate change concerns,especially given that soil is one of the largest terrestrial carbon reserves.Accurate predictions necessitate comprehensive soil profile measurements,which are resource-intensive to obtain.To address this,depth functions are employed to derive continuous estimates,aligning with standardized depths.However,global datasets employing depth functions in raster format have not been widely utilized,which could lower financial costs and improve accuracy in data-scarce regions.Furthermore,research into aggregating depth functions for realistic carbon stock estimations remains limited,offering opportunities to streamline cost and time.The aim of this study was to apply equal-area splines to estimate soil carbon stocks,utilizing SoilGrids and iSDAsoil datasets in a 317-km^(2) Quaternary catchment(30°48′E,29°18′S)in KwaZulu-Natal,South Africa.Both datasets were resampled to a 250-m resolution,and the splines were interpolated to a depth of 50 cm per pixel.Various aggregation methods were employed in calculation,including the cumulative sum(definite integral),discrete sum(sum of 1-cm spline predictions),and the mean carbon stock(mean to 50 cm).Quantitative evaluation was performed with 310 external soil samples.SoilGrids showed higher predictions(100–546 kg m^(-2))than iSDAsoil(66.9–225 kg m^(-2))for the cumulative sum.The discrete sum also exhibited higher prediction values for SoilGrids(293–789 kg m^(-2))compared to iSDAsoil(228–557 kg m^(-2)).SoilGrids aggregated with the discrete sum closely matched previous studies,estimating total carbon stock for the catchment at 7126 t,albeit with spatial inconsistencies.However,when evaluating with an external dataset,the results were not satisfactory for any method according to Lin's concordance correlation coefficient(CCC,correlation of a 1:1 line),with all models obtaining a CCC below 0.01.Similarly,all models had a root mean squared error larger than 59 kg m^(-2).It was concluded that SoilGrids and iSDAsoil were spatially inaccurate in the catchment but can still provide information about the total carbon stock.This method could be improved by obtaining more soil samples for the datasets,incorporating local data into the spline,making the method more computationally efficient,and accounting for discrete horizon boundaries.展开更多
文摘Soil carbon stock research has gained prominence in environmental studies amidst climate change concerns,especially given that soil is one of the largest terrestrial carbon reserves.Accurate predictions necessitate comprehensive soil profile measurements,which are resource-intensive to obtain.To address this,depth functions are employed to derive continuous estimates,aligning with standardized depths.However,global datasets employing depth functions in raster format have not been widely utilized,which could lower financial costs and improve accuracy in data-scarce regions.Furthermore,research into aggregating depth functions for realistic carbon stock estimations remains limited,offering opportunities to streamline cost and time.The aim of this study was to apply equal-area splines to estimate soil carbon stocks,utilizing SoilGrids and iSDAsoil datasets in a 317-km^(2) Quaternary catchment(30°48′E,29°18′S)in KwaZulu-Natal,South Africa.Both datasets were resampled to a 250-m resolution,and the splines were interpolated to a depth of 50 cm per pixel.Various aggregation methods were employed in calculation,including the cumulative sum(definite integral),discrete sum(sum of 1-cm spline predictions),and the mean carbon stock(mean to 50 cm).Quantitative evaluation was performed with 310 external soil samples.SoilGrids showed higher predictions(100–546 kg m^(-2))than iSDAsoil(66.9–225 kg m^(-2))for the cumulative sum.The discrete sum also exhibited higher prediction values for SoilGrids(293–789 kg m^(-2))compared to iSDAsoil(228–557 kg m^(-2)).SoilGrids aggregated with the discrete sum closely matched previous studies,estimating total carbon stock for the catchment at 7126 t,albeit with spatial inconsistencies.However,when evaluating with an external dataset,the results were not satisfactory for any method according to Lin's concordance correlation coefficient(CCC,correlation of a 1:1 line),with all models obtaining a CCC below 0.01.Similarly,all models had a root mean squared error larger than 59 kg m^(-2).It was concluded that SoilGrids and iSDAsoil were spatially inaccurate in the catchment but can still provide information about the total carbon stock.This method could be improved by obtaining more soil samples for the datasets,incorporating local data into the spline,making the method more computationally efficient,and accounting for discrete horizon boundaries.
