Analysis of potential nature-based solutions for the Mun River Basin, Thailand

The changes in average climatic conditions and extreme events have the potential to disrupt human and ecological systems (Vaghefi et al. 2013). The frequency and intensity of extreme rainfall events and the temperature extremes, for all RCP (Representative Concentration Pathway) scenarios, are projected to rise in the mid of 21st century (IPCC 2014). In this study, we used Shared Socio-economic Pathway (SSP)5-8.5. Various studies have focused on the rise in long-term temperature and variability in precipitation in different regions of the world, as well as their environmental consequences (Christy et al. 2009; Gebrechorkos et al. 2019; Alahacoon & Edirisinghe 2021). The extent of climate change impacts and the link between the consequences varies at different levels and by region with increasing temperatures leading to temperature extremes, heatwaves, hydrological changes, floods, and droughts (Arnell et al. 2019; Zisopoulou & Panagoulia 2021).

Thailand is not an outlier when it comes to the possible effects of anthropogenic climate change on the natural world. Climate change is anticipated to exacerbate the pre-existing risks of droughts in the dry season and floods in the wet season in such a scenario (Hydro & Agro Informatics Institute 2012). For a few climate factors in Thailand, the impact has been expected to be severe in some cases. According to Khadka et al. (2022) under the high emission scenario, SSP5-8.5 increases in maximum and minimum temperatures will occur in northeast Thailand by 1.45 °C (0.8–1.9 °C) and 1.54 °C (1.1–1.9 °C), respectively. In addition, annual rainfall will become temporally more uneven with increases (2–8%) during the rainy season (June–October) and decreases of (6–11%) during the pre-rainy season (March–May).

Recent research has highlighted the effects of climate change on hydrology and extreme events (Ligaray et al. 2015; Sharma & Babel 2017; Shrestha & Lohpaisankrit 2017) in various river basins in Thailand. Both the mean annual discharge and the annual maximum daily flow of the Chao Phraya River Basin are expected to increase in the future (Kure & Tebakari 2012). While research conducted by Hoang et al. (2016) on the Mekong River predicted that the hydrological cycle would become more intense in the future, with an increase in both annual (5–16%) and seasonal flows. Shrestha & Lohpaisankrit (2017) predicted an increase in yearly flood severity under both emission scenarios, with an additional 60 km² of land flooded under the 1 in 100 chance event every year in the Yang River Basin, Thailand.

In particular, South Asia's geography makes it susceptible to natural disasters and the most flood vulnerable regions in the world (Shah et al. 2020). Flooding is the most frequent natural disaster in Thailand resulting in the loss of life and damage (Prabnakorn et al. 2019b). Between 1984 and 2014, Thailand suffered 66 floods, which affected 48.7 million people, costing approximately USD $45 billion worth of damages (Prabnakorn et al. 2019b). In 2011, Thailand faced the worst flooding in half a century (Kittipongvises et al. 2020) and 65 of Thailand's 77 provinces were declared disaster zones, which impacted over 10 million people. The World Bank estimated it to be the fourth most costly natural disaster in the world from 1995 to 2011 (Kittipongvises et al. 2020).

Droughts will also become more severe, last longer, and occur more frequently as a result of climate change (Dai 2011; IPCC 2011). Lee et al. (2016) found that the frequency and extent of worldwide agricultural droughts are expected to rise in the future, particularly in the northern hemisphere. Droughts will be exacerbated by increased temporal variability, even if precipitation increases in some regions. According to FAO (2011), droughts are very common in northeastern region of Thailand while approximately 90% of rice in the northeast is farmed in a rainfed system, making it vulnerable to climate change and variability.

Ahmed et al. (2022), Dorst et al. (2019), Ruangpan et al. (2019), Hewett et al. (2020) and Mills et al. (2020) suggested that NBS aim to help societies address a variety of environmental, social, and economic challenges in sustainable ways. They are actions inspired by, supported by, or copied from nature; both using and enhancing existing solutions to challenges, as well as exploring more novel solutions.

Based on new green technologies (Huang et al. 2020), nature-based solutions (NBS) can be applied to a range of ecosystems (Bridgewater 2018), promoting ecosystem services and enhancing drought and flood resilience to climate change, while improving human well-being (Ahmed et al. 2022). They can be both structural (green-blue infrastructure, e.g., wetlands) and non-structural measures (e.g., holistic, improving the local knowledge through stakeholder engagement) (Ruangpan et al. 2019; Hewett et al. 2020). Albert et al. (2019) defined NBS as actions that (i) alleviate a well-defined social challenge, (ii) utilise ecosystem processes, and (iii) are embedded within viable governance or business models. Arguably, they are more sustainable than traditional grey infrastructure (Nelson et al. 2020). Dorst et al. (2019) found several similarities between NBS, Ecosystem-based Adaption (EbA), and Green Infrastructure; however, the three varied on what qualified as ‘nature’. Nevertheless, as Nelson et al. (2020) stated the term ‘solutions’ should not be oversold, and that in fact though NBS provides a solution, this ‘solution’ is ultimately a long-term process that requires dedicated efforts from all those involved.

Within the same field of research, the concept of ‘NBS’ is very similar to other green strategies in the flood literature. For example: Low Impact Developments (LIDs), Sustainable Drainage Systems (SuDS), Water Sensitive Urban Design (WSUD), Best Management Practices (BMPs), Blue-Green Infrastructure (BGI), EbA, Ecosystem-based Disaster Risk Reduction (Eco-DRR) (Ruangpan et al. 2019; Ahmed et al. 2022), Catchment Systems Engineering (Hewett et al. 2020), and Natural Flood Management (Cooper et al. 2021). Though NBS are well used for adaptation purposes in the northern hemisphere, they have only recently been adopted in the Southern hemisphere.

Most recently, Hekrle (2022) provided a systematic review of 153 scientific publications, focusing on data collection techniques and perspectives of well-being used when eliciting preferences towards multiple ecosystem services provided by NBS. The results found that most NBS research has been conducted in urban areas (not agricultural), with most of the studies using questionnaire surveys as a main technique of data collection. The study concluded that there is a need for future research into how NBS implementation influences individuals, communities, and social well-being, the benefits of which should be included in practical policy decisions.

Though NBS studies are increasing, Lechner et al. (2020) indicated that there were few use case studies focusing on applications within Southeast Asia. Kumar et al. (2021) found that between 1965 and 2021 only 19.3% of NBS research were within Asia with only two using Thailand as a case study. Ahmed et al. (2022) analysed 20 articles from 2000 to 2021 and found only two provided examples in Thailand, while other studies were targeting China, Hong Kong, Malaysia, Brazil, and Africa. The identified NBS studies related to Thailand include Horstman et al. (2014), Majidi et al. (2019), and Vojinovic et al. (2021), however where the former of these apply NBS to a coastal region, and the latter two question the effectiveness of NBS within urban catchments. Moreover, there appears to be very limited literature on applying and evaluating NBS for flood or drought hazards within agricultural regions ((Gómez Martín et al. 2021).

Most studies investigating NBS effectiveness are limited to empirical studies, thus ignoring factors that cannot be studied empirically, i.e., long-term climate change (Gómez Martín et al. 2021) or land use change (LUC) (Brown et al. 2018). Consequently, as stated by Gómez Martín et al. (2021), scenario modelling approaches that consider temporal projection are crucial to understanding the limitations of NBS. Furthermore, (Hewett et al. 2020; Cooper et al. 2021) found that the scientific evidence and published data surrounding the effectiveness of NBS for managing and mitigating flood risk is limited. This is a major barrier to their inclusion in catchment-scale management. Thus, there is a clear need for further research to quantify the effectiveness of NBS in managing flood risk. Croeser et al. (2021) argued that current NBS assessments tend to either give highly aggregated results or are tailored to only one specific ecosystem service. Instead, Croeser et al. (2021) demonstrated how Multi-Criteria Decision Analysis (MCDA) can be used to select a number of NBS to address multiple challenges, which advanced the practice of NBS selection, however, the study is solely urban based within European case studies.

Mubeen et al. (2021) aimed to build on spatial analysis and proposed a methodology for the allocation of large-scale NBS using suitability mapping. The methodology was implemented using ESRI ArcMap software to map the suitability for four types of NBS interventions: floodplain restoration, detention basins, retention ponds, and river widening. Flood maps were used to determine the volume of water to be stored for flood risk reduction, the suitability maps produced indicate the potential for selection and allocation of large-scale NBS. Similarly, Alves et al. (2022) developed the NEEDS for ACTION framework, which combines Multi-Criteria Decision Analysis-Geographical Information Systems (MCDA-GIS) and participatory approaches for applying NBS in the semiarid region of Brazil. The study suggested that one of the main challenges for applying NBS on the local scale is to systematically consider the local conditions, such as LUC, vulnerability and exposure, and their associated uncertainties (Alves et al. 2022).

Another example is seen in Prabnakorn et al. (2021). The study analysed the adverse impacts of basin-scale floods and droughts on rice cultivation to provide feasible solutions to mitigate the disasters within the Mun River Basin. The study demonstrated that while the total storage capacity of in situ and ongoing projects is sufficient to tackle both hazards, it can only be achieved if the implementations are effectively utilised. Based on this, the authors proposed that small farm ponds, a subsurface floodwater harvesting system, and oxbow lake reconnections could provide additional solutions for the region (Prabnakorn et al. 2021). Though the study is highly useful in combining solutions for flood and drought management, the potential future implications of NBS were not assessed, and the study took a sole agronomist direction. In addition, Koncagul (2018) found that the ponds could capture over 3 billion m³ (almost 30%) of the wet season flows to be harvested and recharge the shallow aquifers of Chao Phraya River Basin. Thus, reducing the magnitude of flooding and offsetting groundwater decline, providing a solution for droughts.

In response to these gaps, the main objective of this article is to develop an MCDA-GIS framework to quantitatively assess single and combinations of NBS within the Mun River Basin (Thailand). The objective is not only for flood risk mitigation, but also providing a systematic review of current NBS for both floods and drought risks at the studied basin. This study draws upon the ideas developed by (Croeser et al. 2021) and proposes an innovative method to analyse the long-term effectiveness of different NBS strategies by integrating them into MCDA-GIS techniques to determine the potential change in flood hazard. For that, we incorporated a number of NBS into plausible future LUC scenarios to analyse the long-term effectiveness of NBS strategies and quantitatively represent how NBS could reduce flood hazards within the region. In addition to this, we propose a methodology for assessing the effectiveness of potential NBS for drought.

The next section of this article shows the developed methodology, with a combination of a detailed literature review of NBS case studies, the MCDA-GIS approach with LUC modelling scenarios, and the validation process. Thereafter, results and discussions describe the effectiveness of NBS in relation to the reduction of flooding hazards. Limitations and future research are then described, including the proposal of a framework for the assessment of NBS for drought. The article is finalised with the conclusions.

Criteria for selecting the papers within the Google Scholar search engine included ‘Nature-Based Solutions Flooding’, ‘Nature-Based Solution Flood Adaptation’, ‘Nature-Based Solution Drought’, ‘Nature-Based Solution Drought Adaptation’, and ‘Nature-Based Solutions and Climate Change’. These studies, published in English, covered worldwide case studies, with a total of thirteen finding case studies from Southeast Asia, of these, eight are specific to Thailand. The Evidence Tool: Nature-based Solutions Evidence Tool (<naturebasedsolutionsevidence.info>) was also considered for screening articles published. A range of mitigation techniques were established for drought, flood, or a combination of both hazards, the most frequently occurring were re/afforestation, changing farming techniques (includes concepts such as changing crops, cover cropping, agroforestry, notill, and organic farming) and the creation of wetlands. The summary of results obtained from the literature review, including: (a) Articles applying NBS for the different continents and countries, (b) Number of publications applying NBS types for flood and drought mitigation and the year they were published can be found within the Supplementary material, Appendix 1.

Four land use (LU) scenarios were selected based on the findings of Penny et al. (2021) and section 2.2 of this paper: Observed (OBs), Business-as-Usual (BAU), Re/Afforestation (FOR), Agricultural Change (CROP), and Wetland Creation (WET), as well as their combinations: NBS1 (combination of FOR and WET), NBS2 (Combination of FOR and CROP), and NBS3 (Combination of FOR, WET and CROP) (i.e., see more details in Figure 4). The BAU scenario is considered as the ‘worst case’ scenario, which expresses when the land use trends remain unchanged but under near-future (2021–2050) climate, and the latter looked at how potential single or combination of NBS could help reduce flood risk. FOR, CROP, and WET scenarios illustrate the application of each NBS in the GIS environment. Each map was produced for the Mun River Basin with the adjustment of the multiple criteria according to the land use scenarios predicted by Penny et al. (2021).

Through evapotranspiration, inception and transpiration, forests play a key role in controlling water, mitigating flood risk and delaying flood peaks, both temporally and spatially (Cooper et al. 2021). They also play a vital role in improving catchment function (Hewett et al. 2020). The selection of the FOR scenario was based on the fact that Thailand's forest cover has fluctuated in the past five decades, from 53.5% in 1961 down to 27.3% in 1990, returning to 31.6% in 2015 (V4MF 2016). In 2014, the government launched the Master Plan for Forest Resources Protection and Sustainable Management to increase forest cover to 40% within 10 years, with the aim to ‘resolve the problems of forest destruction, trespassing of public land, and sustainable management of natural resources’ (V4MF 2016). Consequently, the FOR scenario increased forest restoration (Penny et al. 2021), supporting green growth laid down by the (Office of the National Economic & Social Development Board 2017).

For the CROP scenario, it was considered that flood and drought conditions affect rice growth and its production (Prabnakorn et al. 2021), within the Mun River Basin 90% of the rice fields are rainfed (Prabnakorn et al. 2019a), however, over the past 30 years the average annual precipitation was insufficient for the dry season and also in some areas the wet season too (Prabnakorn et al. 2021). Previous studies have found that cultivation incentives such as changing crop choice can help flood mitigation by reducing runoff (Zandersen et al. 2021) and decrease drought vulnerability by growing species less water intensive, thus more water resilient and tolerant to drought stresses (Fedele et al. 2018). In the study conducted by Penny et al. (2021) stakeholders within the Mun River Basin were asked to rank agricultural crops in terms of their risk to drought, results found that 47% believed paddy rice to be at most risk from drought. Furthermore, the study went on to argue that 45% of the Mun's soil is more suitable for growing field crops and Perennial and Orchard styled crops than paddy fields. Consequently, one future scenario will look at NBS in terms of changing agriculture.

Finally, wetland restoration and/or creation, which literature has shown multiple successful applications in flood mitigation, were considered. The future scenario developed by Penny et al. (2021) looked at the restoration of the watershed/wetland area of the Mun to fall in line with the 2018 Water Resources Act and National Water Resources management plan, whereby 2037 the total Marsh and Swamp area would increase by 1% to cover the target restoration area of watershed/wetlands. Floodplains via wetlands provide key ecosystem services (Jakubínský et al. 2021): through water storage, they are effective buffers in reducing hydrological risks such as drought, wildfires and floods (Belle et al. 2018). They are arguably much more efficient than grey infrastructure such as dams (Sahani et al. 2019) and small water storage ponds in terms of economy, flood, and drought risk reduction, and environmental conservation (Grygoruk et al. 2013; Sahani et al. 2019; Acreman et al. 2021). For these reasons, the WET scenario was also included in the analysis.

Along with these four initial scenarios, three more potential solutions were developed: NBS1 (combination of FOR and WET), NBS2 (Combination of FOR and CROP), and NBS3 (Combination of FOR, WET and CROP). These additional scenarios meant that all possible solutions of land use derived NBS were acknowledged/recognised. In previous studies within NBS Thailand, both Majidi et al. (2019) and Vojinovic et al. (2021) concluded that a combination of NBS would be the most effective for flood and drought reduction.

Using a methodology primarily derived for the ENRICH Project (e.g., Enhancing Resilience for future Hydro-meteorological extremes in the Mun River Basin in the northeast of Thailand), we used MCDA-GIS to spatially and quantitatively analyse and assess flood hazard (Penny et al., 2022). Raster maps for slope (ALOS-JAXA), elevation (ALOS-JAXA), land use (source Land Development Department), soil types (source Harmonised World Soil Database), drainage density (distance from Mun River), and mean rainfall (Khadka et al. 2022).

Near-future (2021–2050) climate, projected by Khadka et al. (2022) was used in this study. The mean rainfall was normalised and based on eight climate models participating in HighResMIPs (Haarsma et al. 2016) of CMIP6 (Table 1). The study by Khadka et al. (2022) assessed the changes in the climate for the near-future (2021–2050) with respect to the baseline period of 1981–2010. The future projections are available for SSP 5-8.5 which represents the high emission scenario (O'Neill et al. 2017).

The modelling approach was validated using previous flood events within the studied region between 2004 and 2022. A total of 136 flood locations were plotted on the flood hazard map (observed scenario – OB scenario) (Figure 5). These locations were reported for flooding by national newspaper articles. The validation was performed by comparing areas with and without flooding by extracting the values of the cells from the observed period using the ‘Sample tool’ within ArcMap 10.6.1 (ESRI), based on a methodology developed by Alves et al. (2021).

The OB scenario was validated with the 136 flood points retrieved from newspaper articles (Figure 5). 84 (62%) points were found within ‘High and Very High’ Flood hazard areas. 38 points (28%) are classified within the moderate hazard level, and only 14 points (10%) represent areas of Low flood hazard with no points found classifying Very Low. This can be explained by the fact that minor flooding (i.e., ‘Low’ and ‘Very Low’) would not receive much attention from national newspaper coverage. In summary, the sample analysis methodology showed that 122 points were classified between ‘Moderate and Very High’ flood hazard, which validates the mapping in approximately 90% of the dataset. Nevertheless, as Alves et al. (2021) stated, the other 14 points representing areas with ‘very low’ and ‘low’ classifications of hazard also consist of flooding events, and can create flood impacts, especially with extreme precipitation.

If LUC trends continue (BAU) an area of 23.6% will see an increase in flood hazard, however with the addition of a single NBS (CROP-WET-FOR) it was found that areas of very low hazard increased and very high flood hazard decreased, this was especially for forestation scenario where increases and decreases of around 2% occurred, medium and high hazard areas also decrease (Figure 6(a)). Compared to the BAU scenario, areas covering 7.6, 5.1, and 11.6% find a decrease in flood hazard for CROP, WET, FOR, respectively, this identifies that forestation as a single solution alone is the most significant when reducing flood hazard (Figure 7(b)). This concurs with the findings from Babel et al. (2021). In fact, the WET scenario finds a similar increase in hazard as BAU compared to the observed period, highlighting that wetlands alone will not make much difference to the flood hazard. Nevertheless, within the WET scenario area close to the river saw a decrease in hazards, especially in Si Sa Ket and Nakhon Ratchasima (Figure 8). Comparing FOR with BAU increases in low and very low hazard was observed in Buri Ram and Nakhon Ratchasmis locations where increased forest growth occurred (Penny et al. 2021). Similar decreases in hazard are seen in CROP scenario though for locations; Buri Ram and Surin (Figure 8).

When combing multiple NBS together (NBS1, NBS2, and NBS3) flood hazard can further be reduced (Figure 7(a) and 7(b)). NBS1, NBS2, NBS3 follow similar trends to the ones described above with decreasing Very High hazards in areas along the river tributaries and in the East; Nakon Ratchsima, Si Sa ket, Rio et, and increasing low hazards seen the south catchments; Nakron Ratchims, Buri Ram, and Surin (Figure 8).

When comparing NBS1 and NBS2 combinations of Afforestation and Wetlands vs. Afforestation and Crop change, respectively, – NBS2 provides the better alternative (Figures 6(a), 7(a), 7(b), and 8). The root cause is probably due to that changing crops covers a wider surface area than the potential wetland increase. In the northeastern region there has already been a shift from rice to less water demanding field crops like sugarcane, and cassava thus the crop types have been matched to the water available (Barnaud et al. 2006). (Chausson et al. 2020) argued that while wetlands have been demonstrated as cost-effective for improving water quality and reducing flooding from heavy rain in urban areas there is very little evidence to suggest that they are effective nor suitable to rural and peri-urban areas, especially in lower-income nations.

Nevertheless, NBS3 offers the best alternative with 12 and 11% hazard area covered for very low and very high hazards, respectively, compared to the respective 9 and 14% observed during the BAU scenario (Figures 7(a), 7(b) and 8). This equates to an area of 15% displaying a decreasing flood hazard. Results agree with previous studies on NBS in Thailand – Majidi et al. (2019), Vojinovic et al. (2021) and Babel et al. (2021) – that a combination of NBS is the most effective. Our findings are also in agreement with (Acreman et al. 2021) and (Prabnakorn et al. 2021) that the restoration of forests and wetlands will reduce flood and drought damage, and their conservation can prevent future increases in hazard and risk.

Cities worldwide are being asked to rethink and redesign towards natural hazard mitigation, especially floods and droughts. In this sense, several countries are implementing NBS, however, many times, this is being made without the clear understanding of their effectiveness for the local situation (Snep et al. 2020). With the adoption of NBS often being postponed owing to that their short-term economic benefits (a major concern among decision-makers) are limited (Huang et al. 2020).

Though we have only considered flood hazards, future studies could take this work further by quantitatively modelling how drought hazards would be affected if NBS were implemented. If drought hazard was assessed using MCDA-GIS, as recently described by Cordão et al. (2020), our methodology to review NBS using future LUC scenarios, could also be applied to drought hazard. In addition, following up the previously published work of Prabnakorn (2020) which assessed the effectiveness of certain NBS regarding drought, the water requirements for the current land use can be calculated. We highlight the importance of considering agricultural crops, and also the analysis of the difference compared to the potential water demand under future climatic scenarios.

The present work can be taken further with the use of hydrological-hydraulic model, for example, ‘Bluespot’ analysis helps to model flood risk by modelling flood inundation within landscape sinks/depressions that are filled during rainstorms (Balstrøm 2022). Bluespot in simple terms is a landscape depression, and thus gives flood depth. Previous studies that have used Bluespot analysis have solely investigated urban flooding (Baby et al. 2021; Pallathadka et al. 2021; Saeed et al. 2021; Thrysøe et al. 2021). If used in this case, an agricultural case study, using hot spot areas of key interest would enable the research to see the flood depth reduction due to NBS.

Future studies could also calculate the potential water storage from NBS in order to determine the water deficit or surplus. The comparison between the different water storages can indicate the optimal solution for the Mun River Basin, based on the current local conditions, such as land uses and water storage. In addition to the above, for both flood and drought hazards the MCDA approach used in this study could be further enhanced by including more parameters, such as the runoff coefficient (Shadmehri Toosi et al. 2019, 2020), the Topographic Wetness Index (Tang et al. 2018; Feizizadeh et al. 2021), groundwater depth (Nigusse & Adhanom 2019), or the Soil Water Potential could be used as an irrigation index under different types of soil and climatic conditions (Kumar et al. 2019). Moreover, if this methodology is applied, we recommend that strategies for participatory planning should involve focus group(s) of local stakeholders within the region.

There are different methods for dealing with uncertainties in modelling approaches. Uncertainty is inherent in virtually all information in real-life decision situations (Danielson & Ekenberg 2019), within model simulations is an important aspect of research especially when model outputs are used to support water management decisions (Refsgaard et al. 2007). The method AHP (Analytical Hierarchy Process) is the most common MCDM method followed by ANP (Analytic Network Process) (De Brito et al. 2018). Both use pairwise ratio scoring methods that were previously used in other flood mapping studies (Duan et al. 2009; Elsheikh et al. 2015; Rahmati et al. 2016; Ghosh & Kar 2018; Tang et al. 2018; Cordão et al. 2020). Some other authors used individual weighting schemes rather than equal weights (as used in this study). Other MCDA method options include SMART (proportional scoring), CAR (Cardinal ranking), and P-SWING (proportional scoring and cardinal ranking) (Danielson & Ekenberg 2019). Tools using CAR and SMART provide similar accuracy to AHP but require less input and mental effort from decision-makers (De Brito et al. 2018). In this sense, the next steps of this research will also evaluate multiple weightage distributions, and not only equal weights, for flood hazard indicators.

As previously discussed, NBS can provide benefits beyond flood and drought hazard reduction. In this sense, we suggest the analysis of multiple benefits with the application of NBS. For this, other studies such as Ashley et al. (2020) may support the analysis of the multiple benefits over time, and O'Donell et al. (2018) and Morgan & Fenner (2019) with the spatial representation of single and multiple benefits along with GIS methodologies. Besides this, we suggest the analysis of spatial scale interactions within natural and urban environments, with public participation and the consideration of limitations that may overmine the effectiveness of NBS.

Finally, we also acknowledge the importance of understanding the limitations of proposals and how it is a key factor for reducing the ‘maladaptation’ of solutions. This is cited by Schipper (2020), in which poorly designed adaptation strategies can result in maladaptation, where exposure and sensitivity to climate change impacts are instead increased as a result of action taken. Maladaptation refers to the ‘process whereby people become even more likely to be negatively affected by climate change’. Considering land use scenarios and predicted change may support the reduction of maladaptation, but we highlight the need for considering the short and long-term positive (and negative) impacts of NBS, as well as addressing the impacts for vulnerability and exposure of communities at flood and drought risk.

In this article, we provide an overview of different continents and countries that reported the outcomes of NBS for floods and drought reduction; however, some areas remain less analysed, which is the case of Thailand. Also, it was shown that most approaches are developed for urban areas, whereas case studies with rural and agricultural regions continue limited. In this sense, this article contributes to the field by presenting a methodology for NBS application for the Mun River Basin in Thailand. The methodology was developed to provide an integrated MCDA-GIS approach for applying NBS, by considering the different land uses of the Mun River Basin, which is a predominantly agricultural region, with regular periods of floods and droughts. Potential solutions already in play within Thailand included: furrows, small farm ponds, subsurface floodwater harvesting systems, oxbow lake reconnections and Living Weirs, however, these small scale NBS alone cannot provide solutions to the region. Results of this article show that combined NBS (of Wetlands, Re/Afforestation and Changing Farming Techniques) are more effective than single strategies (Figures 6(a), 7, and 8). When alone, forestation is the most successful NBS in reducing flood hazard – but a combination of all three provides the best solution for NBS within the Mun River Basin. Arguably, NBS within the Mun should not be considered as a single action to protect or restore nature but as a process that engages with the local stakeholders to merge natural and human systems.

This work was conducted under the project title ‘Integrated Management of Flood and Drought in the Mun River Basin, Thailand’ funded by the UK's Natural Environment Research Council (NERC) under the NERC COP26 Adaptation and Resilience Project Scoping Call. Data were taken from ‘Enhancing Resilience to Future Hydro-meteorological Extremes in the Mun River Basin in northeastern Thailand-ENRICH’ project. Funding agencies of the project of ENRICH-1 include the following: Thailand Science Research and Innovation (TSRI), and the NERC (NE/S002901/1) under the Newton Fund. Also, the first author was supported by the UK EPSRC Water Informatics Science and Engineering (WISE) CDT, grant no. EP/L016214/1 and the fourth and fifth authors’ work on NBS has been supported by the EU H2020 project RECONECT – Regenarating ECOsystems with Nature-based solutions for hydro-meteorological risk rEduCTion (grant agreement ID 776866).

Data cannot be made publicly available; readers should contact the corresponding author for details.

The authors declare there is no conflict.