REMOTE SENSING FOR MONITORING AND MAPPING LAND PRODUCTIVITY IN ITALY

摘要: We present a remote sensing-based methodology for rapid assessment of status and trends of Land Productivity (LP) at national and sub-national scales. This methodology aims at supporting environmental national and international policies to achieve the Land Degradation Neutrality target in the framework of the UN Agenda 2030 and the Sustainable Development Goals. The work was performed in Italy using the NASA-MODIS Normalized Difference Vegetation Index (NDVI) as proxy indicator of LP status and trend. LP status was identified by mean and standard deviation of 2000-2015 yearly LP values. LP trends of the yearly time series were computed using Mann-Kendall (MK) and Contextual MK (CMK) tests. The amount of land with valid increasing and decreasing trends is estimated assuming the 95% significance level of trends in the areas with good pixel reliability. The area of increasing and decreasing LP are estimated for the national territory and for different land covers. The widespread observed positive LP variations were correlated to the progressive renaturalization of lands subsequent to the decrease of agricultural activities and increasing precipitation trends in the winter season. LP decrease affected very limited areas and hot spots were correlated to changes of seasonal precipitation and human activities. The municipalities most affected by LP decline or increase are identified for future monitoring and validation activities.