Vegetation Phenological Metrics and Accumulated Antecedent Precipitation (AAP) in dryland pastures.

Abstract

Vegetation dynamics in dryland systems is highly dependent on soil moisture availability. Arid and semi-arid ecosystems are under the pressure of climate change and are facing overgrazing and logging, leading to increased degradation and desertification. The drylands of Mendoza, Argentina, are fragile ecosystems devoted to cattle breeding on native bushes and rangelands. Livestock farming relies on the productivity of natural resources, closely related to the monthly, annual, and seasonal rainfall, which is a critical driver of vegetation productivity and dynamics. However, the limited availability of precipitation data from gauging stations prevents a detailed analysis of the relationship between rainfall and vegetation. Therefore, satellite-estimated rainfall becomes a valuable information source to overcome this constraint. This study aims to analyze the relationship between the antecedent accumulated precipitation (AAP) and the vegetation dynamics in terms of phenological metrics (Length of Growing Season – LGS; Peak of Growing Season – PGS; Amplitude of Growing Season – AGS) for four vegetation types in Southeast Mendoza, Argentina (Bush steppe with low land cover; Open Bush; Forest of Prosopis Flexuosa; and Psammophilous Grassland). Vegetation parameters were derived using the software TIMESAT from Savitzky-Golay smoothing NDVI series of MODIS-Terra (MOD13Q1 V6.1) over 20 years (June 2000 to May 2020) and then correlated to AAP estimated by satellite using GPM (Global Precipitation Measurement) considering three time periods: Spring (accumulated precipitation of September to December), Spring plus Summer (September to February) and the duration of the Growing Season of each vegetation type. All vegetation types showed a similar response and behavior regarding the AAP and vegetation dynamics metrics. The LGSs are similar, from 187 days for Psammophilous grassland to 198 days for Forest of Prosopis. However, there are differences at the start of the season. The PGSs (peak of NDVI) and the AGS show higher correlations to the spring and summer precipitation, while the LGS correlates to spring and accumulated precipitation during the growing season. This information can help manage cattle grazing, avoid overgrazing, and manage production sustainably. Tracking vegetation responses to rainfall in space and time is of utmost importance for managing the limited resources,