研究目的
To compare three different thinning intensities in terms of the on-site C stock after 13 years and to develop models of biomass and SOC in Pinus halepensis forest, based on low density ALS in southern Spain.
研究成果
The study confirmed that heavy thinning gave the highest efficiency with regard to C sequestration. The use of low density ALS data and kNN algorithms provided accurate estimates of on-site C stocks in Aleppo pine forests, useful for silvicultural planning and future monitoring.
研究不足
The use of low density ALS data may lead to an underestimation of tree height and consequently errors in biomass estimation. The time delay between ALS data acquisition and field data collection (5–8 years) was not considered a significant source of error due to the low growth rate of the pine forest under study.
1：Experimental Design and Method Selection:
The study used a factorial randomized block design with three thinning intensity treatments (unthinned-control, moderate thinning, and heavy thinning) and three replicate blocks. ALS data were used to estimate forest metrics and C stocks.
2：Sample Selection and Data Sources:
83 plots were established within the forested areas of 'Los Cuadros' in Murcia, Spain. Field data included diameter at breast height, stand density, basal area, and height of trees.
3：List of Experimental Equipment and Materials:
ALS data were provided by the PNOA, using an airborne Leica ALS60 discrete return sensor. Field measurements were taken using a calliper and Vertex III hypsometer.
4：Experimental Procedures and Operational Workflow:
ALS metrics were computed for each plot after normalizing the data by subtraction of the DEM. Non-parametric kNN models were developed to estimate Wt and SOC.
5：Data Analysis Methods:
The accuracy of the Knn predictions was estimated by internal validation, external validations, and cross-validation. The best models were selected based on R2 and RMSE values.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
