부산대학교 도서관

The predicted increase in the frequency and magnitude of climate extremes threatens the stable delivery of crucial ecosystem services, such as water purification, climate regulation and the delivery of wood and products. Within that context, it is important to assess and monitor the stability of ecosystems with respect to environmental anomalies at large spatial scales. Time series of vegetation properties derived from satellite imagery, such as the Normalized Difference Vegetation Index (NDVI) provide an interesting asset to quantify large scale stability metrics. Yet, three interfering factors may hamper the reliable quantification of vegetation stability using remote sensing time series: (i) noise, (ii) spatial heterogeneity of climate anomalies and (iii) temporal changes in vegetation response. To characterise or account for these interfering factors, three frameworks or methods were illustrated: (i) a Monte Carlo simulation analysis to characterise noise, (ii) an ARx model to account for the spatial heterogeneity of climate anomalies and (iii) a running window analysis to characterise the temporal variability in vegetation response.