Efficient Estimation of Biomass from Residual Agroforestry

Abstract

Cost-effective sampling methods for the estimation of variables of interest that are time-consuming are a major concern. Ranked set sampling (RSS) is a sampling method that assumes that a set of sampling units drawn from the population can be ranked by other means without the actual measurement of the variable of interest. We used data on vegetation dynamics from satellite remote sensing as a means in which to rapidly rank sampling units across various land covers and to estimate their residual agroforestry biomass contribution for a small cogeneration facility located in the center of a study area in central Italy. A remote sensing map used as an auxiliary variable in RSS enabled us to cut down the photo-interpretation of the residual biomass present in sampling units from 745 to 139, increase the relative precision of the estimate over common simple random sampling, and avoid individual subjective bias being introduced. The photo-interpretation of the sampling units resulted in a 1.12 Mg ha−1 year−1 mean annual density of residual biomass supply, although unevenly distributed among land cover classes; this led to an estimate of a yearly supply of 132 Gg over the whole 2276 km2 wide study area. Further applications of this study might include the spatial quantification of biomass supply-related ecosystem services.