Earth's climate is changing due to anthropogenic greenhouse gas emissions, leading to increased frequency and severity of extreme weather events like droughts and floods. The impacts of these events are not solely determined by climate changes but also by the sensitivity of the affected biophysical environment. While emission reduction strategies are crucial, local-scale adaptation actions are necessary to reduce this environmental sensitivity. Nature-based Solutions (NbS), such as ecosystem management and restoration, are proposed as effective adaptation strategies to buffer societies from the impacts of climate-related extremes. These solutions aim to reduce or offset the impacts of anthropogenic climate change on the physical climate system, particularly regarding extreme events. Examples include restoring or protecting riverbanks, wetlands, and catchments to improve water security and flood protection. Despite the growing interest in NbS, quantitative evidence demonstrating their ability to modulate the biophysical impacts of anthropogenic climate change on extreme events remains scarce. Most quantitative assessments focus on climate change mitigation (carbon sequestration) or global temperature effects, lacking the regional and local specificity needed for effective decision-making. This study aims to address this gap by quantitatively assessing the impact of NbS in moderating the biophysical impacts of anthropogenic climate change on a specific drought event.

Existing literature highlights the potential of NbS in climate change adaptation, but quantitative evidence on their ability to modulate the impacts of anthropogenic climate change on extreme events is limited. While many studies explore the mitigation potential of NbS (carbon sequestration), fewer studies quantitatively assess their impact on extreme weather events' biophysical impacts. Reviews of ecosystem services and disaster risk reduction mention NbS but mostly focus on impacts without isolating the contribution of NbS to the effects of anthropogenic climate change. Furthermore, research often lacks a focus on locally relevant NbS options in the Global South. Quantifying the potential of NbS is challenging due to the lack of counterfactual scenarios (a world without human influence). Each extreme event is unique, resulting from both natural and anthropogenic factors, making complete attribution to ACC impossible. However, ACC alters the characteristics of extreme events, making quantitative assessment possible through event attribution science. This field has advanced significantly, with numerous studies quantifying ACC's role in specific extreme weather events. While some studies extend event attribution to hydrological systems and societal impacts, fewer studies integrate NbS into this framework. This study builds upon this existing research by employing event attribution methods to analyze the role of NbS in a specific drought.

This study utilized a joint event attribution framework to assess the impact of anthropogenic climate change (ACC) and invasive alien tree (IAT) management on streamflow during the 2015-2017 Cape Town "Day Zero" drought. The study focused on two headwater catchments: the Upper Berg (78 km²) and Du Toits (46 km²), critical sources of water for Cape Town. The framework involved linked climate and hydrological model simulations. Daily streamflow was simulated for the drought period (2015-2017) using climate change attribution inputs (daily rainfall and reference evapotranspiration) from three climate model experiments: (i) Hadley Centre Regional Model (HadRM3P) nested in the Hadley Centre Global Atmospheric Model (HadAM3P-N96) from the weather@home modelling system (W@home); (ii) ECHAM5.4 from the Climate of the 20th Century Plus (C20C+) Detection and Attribution project (C20C); and (iii) a multi-model ensemble from the fifth phase of the Coupled Model Intercomparison Project (CMIP5). Four climate-IAT states were simulated: (i) Natural Current (NC) – a natural climate without ACC and current IAT levels; (ii) Actual Current (AC) – the actual climate with current IAT levels; (iii) Actual Cleared (ACL) – the actual climate with all IATs cleared; and (iv) Actual Invaded (AI) – the actual climate with full catchment IAT invasion. A locally validated, physically based hydrological model (MIKE SHE) coupled with a channel routing model (MIKE HYDRO) was used. Each climate-IAT state was represented by an ensemble of 145 hydrological model simulations. The analysis used two main metrics: (i) QR%, the percentage of reference state (NC) streamflow realised for each comparator state; and (ii) QR% point difference, the percentage point difference between the QR% of the AC state and the other states. A bootstrap percentile confidence interval methodology was used to assess confidence levels in estimates.

The multi-model synthesis showed that ACC reduced drought-period streamflow by -17% (95%CI: -22, -12%) in the Upper Berg and -22% (95%CI: -29, -15%) in the Du Toits. Clearing IATs before the drought would have ameliorated these reductions, with a 9-percentage point gain (95%CI: +3, +15%) in the Du Toits. In the Upper Berg, the effect of clearing was less substantial (+1%, CI: -4%, +6%), likely because of prior clearing efforts. If IATs had spread to full catchment coverage, the impact of ACC on streamflow would have been substantially worse (-31% and -43% in the Upper Berg and Du Toits respectively). Even with complete clearing, the ACC signal wasn't fully removed, indicating that NbS alone cannot fully offset ACC impacts. Analysis of rainfall and reference evapotranspiration showed a strong ACC influence on rainfall and a weaker, less significant effect on reference evapotranspiration. This supports previous findings that reduced rainfall was the dominant driver of the 2015-2017 drought. The study demonstrated the effectiveness of managing IATs in mitigating ACC impacts on streamflow, particularly in heavily invaded catchments. The benefits of clearing are greater in catchments with higher levels of invasion while maintaining low invasion levels is crucial in already cleared catchments.

The findings demonstrate the effectiveness of invasive alien tree clearing as an NbS in mitigating the impacts of anthropogenic climate change on drought streamflow, particularly in water-stressed environments experiencing woody encroachment. The study highlights the importance of considering the scale and extent of invasion when implementing NbS strategies. The results are relevant to water-limited regions globally facing similar challenges. While clearing invasive alien trees can significantly reduce the impact of climate change on drought streamflow, it cannot completely offset the effects of ACC, emphasizing the need for integrated adaptation approaches. The dominance of reduced rainfall as the main driver of the drought is consistent with regional climate projections and highlights the importance of addressing precipitation changes in adaptation planning.

This study provides quantitative evidence that NbS, specifically invasive alien tree clearing, can be an effective adaptation strategy to mitigate the impact of anthropogenic climate change on drought streamflow. The findings highlight the importance of integrated adaptation strategies and the context-specific nature of NbS. Future research should explore the evolving contribution of NbS to offsetting ACC impacts in a warmer and drier climate and investigate the conditions under which human influence on climate might exceed the potential of NbS to reduce hydrological drought impact. The results emphasize the value of strategically targeting heavily invaded catchments while maintaining low invasion levels in already cleared areas to optimize water security in a changing climate.

The study focused on two specific catchments in South Africa, limiting the generalizability of findings to other regions. The hydrological model, while validated, has inherent uncertainties. The study used climate model simulations, introducing uncertainties related to model limitations and future climate projections. While the study incorporates a multi-model ensemble, there might be inherent biases in individual models. Lastly, the study focused on the impact of invasive alien trees, neglecting other potential NbS or adaptation strategies.