Multi-year monitoring land surface phenology in relation to climatic variables using MODIS-NDVI time-series in Mediterranean forest, Northeast Tunisia

The Mediterranean region is one of the most vulnerable regions to climate change. The majority of climate models forecast a rise in temperatures and less rainfall, which have been observed in recent decades. These changes will affect several vegetation properties, especially phenological dynamics and traits, by increasing drought intensity and recurrence. In this climate change context, the present study aims to assess the evolution of vegetation state and its relation with the climate dynamics in the Mediterranean forest region of northeast Tunisia using Land Surface Phenology (LSP) metrics and the vegetation index (NDVI) analysis from 2000 to 2017. To conduct this work, we used precipitation and temperature data from the two closest weather stations and 16-day NDVI composite images from the MODIS satellite source, with 250-m spatial resolution. Three phenological metrics- start of season (SOS), end of season (EOS), and length of season (LOS) - were obtained and compared for different vegetation types. The LSP variation in response to climatic metrics was also analyzed. The results showed that the LSP in our study area changed significantly during the 2000-2017 period, which includes an average 7.8 days delay in the SOS, an average advance in the EOS by 5 days, and LOS shortened by an average 12.8 days. Autumn (Pr_9) and spring (Pr_3 and P3_4) precipitations, as well as maximum temperature (Tx9+10), represent the best climate parameters to explain the changes in LSP. Both the NDVI and SPEI showed a significant high correlation (p < 0.001) on longer time scales. LSP and NDVI proved useful tools for monitoring the vegetation state according to climate for better planning territory purposes.

Automated summary

The study found significant changes in vegetation state in the Mediterranean region, including shifts in the start and end of season, as well as season length. Precipitation and maximum temperature were identified as key climate parameters explaining these changes. NDVI and SPEI showed high correlation on longer time scales.