Relationships between climate, topography, water use and productivity in two key Mediterranean forest types with different water-use strategies

Abstract

Climate and topography have both strong effects on forest water and carbon cycles. However, little is known about their combined effects on the long-term water use and productivity of forests that have different water-use strategies. Here, we used structural equation modelling (SEM) to test direct and indirect influences of climate and topography on the long-term (mean over 2000–2014) evapotranspiration (ET) and gross ecosystem productivity (GEP), as assessed from satellite-based models, across two major co-occurring Mediterranean forest types with different water-use strategies (oak woodlands and pine forests). The estimated GEP and ET were higher by c. 6% and 15%, respectively, in the oak woodlands (Quercus calliprinos) than in pine forests (Pinus halepensis). As a result, the water use efficiency was higher in the pine forests (by 9%), consistent with P. halepensis conservative behaviour. Using the SEM, we found that the mean annual surface skin temperatures had the largest influence on the productivity and ET, with a similar net adverse effect across both forest types. In contrast, the mean annual precipitation was not related to GEP across both forest types but positively affected the ET in the oak woodlands. Slope and aspect had both significant but secondary influence on the forests fluxes, with higher GEP and ET found on the steeper slopes across the oak woodlands and higher ET found on the steeper slopes across the pine forests, associated with north-facing aspects. Applying the SEMs for the pine and oak forests, we predicted reductions of 16% and 31% in the productivity of both forests for projected increases in temperatures of 1°C and 2°C, respectively. Our results suggest that projected warming may have a strong impact on the productivity of Mediterranean forests, severely decreasing the CO2 uptake of the trees, independent of their water-use strategy.