Testing drought sensitivity of different land use types via a low parameter isotope-aided ecohydrological model approach in a lowland headwater catchment, Germany

Stable water isotopes are naturally occurring conservative tracers that act as a fingerprint of water sources and ecohydrological fluxes. Previous studies have shown that some of those fluxes, like evapotranspiration and infiltration, are influenced by vegetation. Thus, land use will play an increasingly important role in water partitioning considering projected climate change-induced shifts of patterns in precipitation and increased atmospheric water demand. The sensitivity of different land use types to drought conditions and their influence on water partitioning varies, and still lacks understanding.

We used stable water isotopes to follow the pathway of precipitation into soil at a lowland headwater catchment with multiple land use types (forest, grassland, arable and agroforestry sites) and integrated our data into a one-dimensional, tracer-aided, plot scale model. The model requires precipitation, potential evapotranspiration and leaf area index as input data and the results were calibrated to real time soil moisture and isotope data. The dataset was collected in the long-term experimental Demnitzer Millcreek Catchment (DMC), Germany, over the growing season of 2021 and includes hydroclimatic conditions as well as isotopes in precipitation, soil water and groundwater. The 2021 conditions, though relatively average in terms of wetness, were affected by a dry spring, an exceptionally large summer storm event (~60 mm) as well as “memory effect” of previous intense drought years.

The implementation of the isotope calculations into the model showed that such a simple, low-parameterisation approach with easily accessible input data can be used to estimate the water balance and track isotopic transformations under plot sites with various land use conditions. The most rapid turnover of water was found under arable land use which resulted in short-term crop vulnerability to drought and slow but more rapid recovery and replenishment of moisture deficits. Forest soils showed slower water turnover with lower soil moisture, mainly reflecting higher interception losses and higher transpiration rates. This, together with access to deeper water, means drought stresses build more slowly at forest sites but can last much longer as storage recovery is slow (>1 year) due to high evapotranspiration. Via adapting the model input data, we further simulated drought conditions to assess the “water footprint” of alternative land use under drought stress.

Our study illustrated the potential of stable water isotope data for simplified ecohydrological modelling approaches to quantify water partitioning. The different effects of land use types on ecohydrological fluxes were successfully simulated and their drought resilience was estimated. For the DMC and similar lowland catchments with similar soil types (sand at forest, loam at grassland and crops) and land cover in Central Europe with the modelled drought conditions, forest sites will initially be more resilient but more vulnerable to lasting droughts, while grassland and arable sites tend to recover more quickly, but can be rapidly stressed by short-term severe events. The modelling provides an experimental framework for assessing the differential effects of droughts of varying longevity and severity on alternative land use strategies.