Abstract
Climate change is increasing the vulnerability of Mediterranean coniferous plantations. Here, we integrate a Landsat time series with a physically-based distributed hydrological model (Watershed Integrated Management in Mediterranean Environments—WiMMed) to examine spatially-explicit relationships between the mortality processes of Pinus pinaster plantations and the hydrological regime, using different spectral indices of vegetation and machine learning algorithms. The Normalized Burn Ratio (NBR) and Moisture Stress Index (MSI) show the highest correlations with defoliation rates. Random Forest was the most accurate model (R2 = 0.79; RMSE = 0.059), showing a high model performance and prediction. Support vector machines and neural networks also demonstrated a high performance (R2 > 0.7). The main hydrological variables selected by the model to explain defoliation were potential evapotranspiration, winter precipitation and maximum summer temperature (lower Out-of-bag error). These results show the importance of hydrological variables involved in evaporation processes, and on the change in the spatial distribution of seasonal rainfall upon the defoliation processes of P. pinaster. These results underpin the importance of integrating temporal remote sensing data and hydrological models to analyze the drivers of forest defoliation and mortality processes in the Mediterranean climate.
Highlights
Mediterranean Basin forests have undergone a severe and accelerated process of historical degradation [1]
Based on the experience of previous results [21,34] which investigated the use of Landsat for forest defoliation assessment in the same area, the objective of this study was to integrate information derived from long time-series of Landsat images and the hydrological model Watershed integrated Management (“Watershed Integrated Management in Mediterranean Environments (WiMMed)”) to assess the defoliation processes of Pinus pinaster Aiton plantations in Southern Spain
We investigated the cause-effect relationship between the defoliation of Pinus pinaster forests in southern Spain using vegetation indices derived from Landsat satellite data and hydrological variables at the site scale, obtained with the hydrological model WiMMed, using three machine learning methods: Random Forest (RF), Support Vector Machines (SVMs) and artificial neural networks (ANN)
Summary
Mediterranean Basin forests have undergone a severe and accelerated process of historical degradation [1]. Field measurement of these processes is expensive, labor-intensive, and often limited in temporal scope and spatial scale [11] For this purpose, most studies have used hydrological models to delineate the soil-hydrological cycle [12]; there is a limited number of studies where hydrological models were used in the context of forest decline analysis, in forest plantations [13]. Most studies have used hydrological models to delineate the soil-hydrological cycle [12]; there is a limited number of studies where hydrological models were used in the context of forest decline analysis, in forest plantations [13] One of these models used in this study, the Watershed Integrated Management in Mediterranean Environments (WiMMed) model, is a physically-based, distributed hydrological model, which has been used to obtain a spatial interpolation of the meteorological variables and the physical modeling of the water in the soil. This model was developed for surface hydrology studies in Mediterranean climates [14]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
