Generating a Baseline Map of Surface Fuel Loading Using Stratified Random Sampling Inventory Data through Cokriging and Multiple Linear Regression Methods

Surface fuel loading is a key factor in controlling wildfires and planning sustainable forest management. Spatially explicit maps of surface can highlight the risks fire. Geospatial information critical enabling careful use deliberate fire setting also helps to minimize possibility heat conduction over lands. In contrast lidar sensing and/or optical based methods, an approach integrating in-situ inventory data, geospatial interpolation techniques, multiple linear regression methods provides alternative load estimation mapping mountainous forests. Using stratified random sampling cokriging analysis, data 120 plots distributed four kinds types were collected order develop total model (lntSFL-BioTopo model) fine (lnfSFL-BioTopo for generating tSFL fSFL maps. Results showed that combination topographic parameters such as slope, aspect, their cross products pine stand, non-pine conifer broadleaf conifer–broadleaf mixed stand was able appropriately describe changes loads with diverse terrain morphology. Based on cross-validation method, study site had RMSE 3.476 tons/ha 3.384 tons/ha, respectively. average all plots, relative error 38% (PRMSE). The reciprocal bias both SFL-BioTopo models tended be exponential growth function amount load, indicating accuracy proposed method likely improved further study. modeling, natural logarithm transformation prevented outcome negative estimates thus estimation. results, this paper defined minimum unit (MSU) area collecting fuels using model. Allocating MSUs at boundary center plot prediction mapping.