Afforesting the UK: Potential Hydrological Impacts

Ambitious targets to expand forested land area have increased over the last decade as governments, businesses, and individuals seek to use woodland as carbon sinks. Currently, it is unknown how proposed afforestation rates will influence catchment water resources and hydrological processes. Both the temporal and spatial scale of proposed afforestation are unprecedented on contemporary timescales and we lack the systematic and quantified understanding of its impact on streamflow at catchment scales. Furthermore, the efficacy of afforestation as a form of natural flood management has yet to be tested across multiple catchments (> 30 km²).

 

The UK Government has pledged to use afforestation as a major component of its approach to reach net zero carbon emissions by 2050. In this project, we investigate the influence of afforestation upon streamflow dynamics in twelve catchments across the British Isles. We aim to determine how woodland planting extent and location influences catchment streamflow response and sensitivity, and which catchment attributes account for these changes. To do this, we use physics-based land surface model JULES (Joint UK Land Environment Simulator) at a 1 km resolution to understand the potential hydrological changes to theoretical afforestation scenarios.

 

Land cover afforestation scenarios were created according to proximity to existing land cover, drainage basin structure and afforestation rate (up to 288 potential land cover scenarios per catchment). The period of 2000-2010, a flood-rich period, was used to simulate and compare how each afforestation scenario would influence catchment flow exceedance levels and streamflow regime using the CHESS-met dataset.

 

Results show increasing afforestation has a clear impact upon streamflow dynamics. A strong negative correlation between increasing afforestation and median and low flows exists but is weaker for higher flows. Some afforestation scenarios could increase the highest flows in the period. Quantile regression on the results of our simulations shows a median change of -1.0 ± 0.21 mm yr^-1 (-0.26 ± 0.10%) for the median flow exceedance per percentage point of broadleaf woodland planted across all catchments. Planting according to Shreve order, or contributing area, led to statistically significant differences in streamflow dynamics. Climatic catchment attributes correlated strongly with catchment median flow sensitivity to afforestation.

 

These results help us to understand how afforestation may influence catchment response to external climatic forcing.  We hope it provides evidence to policymakers wishing to understand the implications of afforestation on water resources and the foundation to understand its future catchment-scale impacts on streamflow.