Effect of compound events on oak tree vitality in a climate change hotspot: analysis of time series in a traditional Spanish dehesa

The Mediterranean region has been identified as a hotspot of climate change characterized by a large tree mortality. Especially holm (Quercus ilex L.) and cork oak trees (Quercus suber L.) in high-value and nature-based agroforestry systems (in Spain known as dehesa) have multiple positive effects, e.g., on the microclimate, carbon storage, erosion prevention, increase of soil water content and soil nutrient concentration. Many studies dealing with the oak decline (also called seca) reported the infestation by root pathogens, in particular the soil-born pathogen Phytophthora cinnamomi, as the main driver. However, rapidly, the focus shifted to the interaction of the pathogen and single abiotic conditions like drought.

We assume that compound events (co-occurring warm spells and soil drought) have a larger correlation with vegetation indices than single climatic drivers. We analyse time series of two vegetation indices, namely the Normalized Difference Vegetation Index (NDVI) and the kernel Normalized Difference Vegetation Index (kNDVI) as an indicator for greenness and vitality. In particular, we focus on the trend of both indices over about two decades (2003-2021) in eight different plots in our study area, on a dehesa in Huelva province, Andalusia. Subsequently, we correlate them with the decomposed signal of compound events.

Based on precipitation and temperature data, we calculated two drought indices, namely the standardized precipitation index (SPI) and the standardized precipitation evapotranspiration index (SPEI). We then used these indices together with temperature to calculate so-called compound events, a co-occurrence of extreme values in multiple environmental drivers. To assess the status of the vegetation, we calculated the NDVI and its newly proposed kernel variant kNDVI from MODIS (MYD13Q1) and Landsat (4-5, 7,8) data in eight different plots in our study area. The kNDVI is a non-linear generalization of the NDVI and showed good behaviour in the Mediterranean and correlates stronger with the gross primary productivity (GPP) than the original NDVI. To extract physically meaningful information, we decomposed the time series signals with the Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) method by Torres et al. (2011) into seasonality, trend, and a remainder part. CEEMDAN is suitable for non-linear and non-stationary time series. To analyse the relationships between vegetation indices and possible climatic drivers, we subsequently calculate lagged cross-correlations (i.e., correlation between different time series) between the Intrinsic Mode Functions (IMFs) of the signal expressing the trend and different seasonalities.

We extracted different positive and significant (p < 0.01) NDVI trend signals from the MODIS time series. The seasonal component corresponded to the expected annual cycle. Based on these first results, we will correlate the NDVI and kNDVI trend signals with the calculated compound events to observe their role in the oak tree mortality.