The accelerating impact of climate change on sea level rise (SLR) has intensified the examination of its effects on coastal regions. This study focuses on Extreme Sea Level Rise (ESLR) and its potential impacts on Europe and North Africa up to 2050, in particular for agriculture. Utilising Joint Research Centre (JRC) Global Extreme Sea Level projections and Copernicus GLO30 Digital Terrain Models (DTM), we mapped areas vulnerable to ESLR under Representative Concentration Pathway (RCP) scenarios 4.5 and 8.5. Through a topological approach, we generated spatially explicit maps of at-risk regions in the Mediterranean basin and northern coastal EU, overlaying them with data from FAO on crop locations, yields, and values (GAEZ). This method allowed us to estimate the magnitude of ESLR's impact on local agricultural systems. Findings reveal that ESLR can severely affect coastal agriculture, suggesting significant potential agricultural losses, impacting food security and economic stability. This research underscores the urgent need for adaptive strategies, including saline agriculture, to mitigate ESLR impacts.
Article
Rising Tides, Sinking Crops: Assessing the Impact of Extreme Sea Level Rise on Coastal Agriculture in Europe and North Africa
https://doi.org/10.21203/rs.3.rs-4950906/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
In recent years, the study of Sea Level Rise (SLR) has gained an unrivalled priority driven by the accelerating impact of climate change on our planet's oceans 1. The term SLR broadly refers to the gradual increase in the average level of planetary open water bodies. Relative Sea Level Rise (RSLR) and Extreme Sea Level Rise (ESLR) are two critical concepts to this field, each representing a unique declination of the phenomenon 2. While they are often discussed in tandem, their distinctions and interactions hold the key to understanding the future of coastal environments and the broader implications for climate adaptation strategies 3.
Relative Sea Level Rise (RSLR), often colloquially referred to as SLR, describes the change in sea level in relation to adjacent land surfaces. This concept encapsulates both the absolute rise in global sea levels driven by the thermal expansion of seawater and the melting of ice sheets, and local changes in land elevation 4. Extreme Sea Level Rise (ESLR) 5, according to the definition provided by the Intergovernmental Panel on Climate Change (IPCC), represents the upper limit of the projected sea level rise throughout the 21st century 6. This upper limit is determined by amalgamating the highest estimates from all contributing factors with the additional effect of extreme events such as storms, surges, and high tides, which are predicted to become more severe and frequent for the upcoming future. These extreme events, capable of causing catastrophic flooding, have become a focal point for researchers and policymakers, as they pose immediate and tangible threats to coastal communities and ecosystems 7. The estimation for RSLR for the next ~ 100 years ranges from 0.2 to over 2 metres in the worst-case scenario 2; 8. The occurrence of an ESLR event can more than quadruple this value, thereby rising the sea level up to a maximum of 8 metres 9.
Global warming, primarily driven by the emission of greenhouse gases (GHGs), is the main cause of SLR. Increased GHG concentrations in the atmosphere lead to higher global temperatures, which in turn cause the thermal expansion of ocean waters. As water warms, it expands and occupies more volume, contributing significantly to SLR 8; 9. Additionally, elevated temperatures result in accelerated melting of polar ice caps and glaciers, particularly in regions like Greenland and Antarctica, adding substantial volumes of water to the oceans. Besides these effects, global warming also enhances the frequency and intensity of extreme weather events, including storms and high tides. Stronger storms lead to more severe storm surges, while higher baseline sea levels amplify the effects of high tides, both of which increase the risk of coastal flooding and erosion 10.
Overall, SLR - in both its Extreme and Relative forms - is poised to exert substantial impacts on global coastal regions, with the extent of these effects contingent upon various factors, including SLR rates, local infrastructure susceptibility, geomorphological characteristics, land use patterns, population growth trends, and community adaptive capacities 1; 2; 5; 6; 11. While SLR undeniably influences diverse sectors, certain anthropogenic activities and sectors exhibit heightened vulnerability 11; 12. Notably, the ramifications of SLR extend to: (i) Infrastructures, where SLR may compromise critical elements like roads, bridges, airports, and ports, potentially disrupting transportation networks and instigating economic and social repercussions 13 14. (ii) Urban settlements, as SLR has the potential to devalue coastal properties, escalate insurance costs, and, in extreme cases, render entire neighbourhoods or cities uninhabitable due to recurrent flooding or permanent inundation 15 16. (iii) Agriculture, particularly in low-lying coastal areas utilised for farming, faces vulnerability to flooding and saltwater intrusion 12 17. This dual impact can result in damaged crops, reduced yields, and consequential effects on food security and economic stability in affected regions 18. (iv) Tourism-dependent coastal areas, such as beaches and resorts, confront potential disruptions from SLR, including damage or destruction of infrastructure and coastal erosion 19, (v) Key infrastructure, including power plants, oil refineries, ports, and railways may experience damage and disruptions in production and distribution due to SLR. This not only carries economic and social implications but also contributes to further climate change by disrupting energy systems 14 13. If no adaptation measures are taken, annual flooding by 2100 could affect 0.2–4.6% of the global population due to a rise in global mean sea level of 25–123 cm 1; 2. This scenario is expected to lead to relevant annual economic losses 8 ranging from 0.3–9.3% of global gross domestic product 20. To address SLR, especially in its extreme form, many communities are considering (and in some cases already initiating) the implementation of adaptation strategies such as building sea walls, elevating buildings, and creating natural and/or semi-natural barriers 20 such as embankments, dykes, mangroves barriers, or oyster reefs 18. However, reducing GHG is equally required to slow down SLR and protect coastal communities and ecosystems against rates of sea level rise that are far beyond adaptation capacities.
Here, we focus on Extreme Sea Level Rise (ESLR) due to its potential to cause significant casualties and economic loss 21. The aim of this research is to map areas vulnerable to ESLR in Europe and the Mediterranean basin to provide an initial assessment of the economic impact of extreme events, thereby informing policy localization and shaping effective mitigation strategies.22. Given the fact that the frequency and intensity of ESLR is projected to increase because of climate change, our work considers different scenarios and vulnerability levels achieving a reliable and comprehensive risk assessment map 3. In the second part of the work an economic analysis is conducted to provide an initial assessment of the foreseeable losses due to an ESLR event. While most of the literature put their attention on residential areas quantifying the direct and indirect damage to houses, infrastructure, and human lives, we focused our lens on cropland and agricultural losses 23. Our estimates shows that urban areas are the one with the highest potential damage per square metre in case of a sea or river flooding event, however they represent only 5% of the total areas prone to ESLR. Cropland on the contrary represents almost 60% of the areas vulnerable to ESLR. In addition, the consequences of an extreme event in agricultural areas can extend well beyond the immediate period, with latent effects manifesting over subsequent years. Soil salinization is a primary concern, as saltwater intrusion during ESLR events leads to salt deposition in the soil, coastal erosion, and alterations in hydrological patterns that influence the balance between saltwater and freshwater17. This would lower crop yield and production for years, exposing communities to food insecurity. Furthermore, the persistent salinization and altered water resources can lead to long-term degradation of agricultural land, diminishing its viability and resilience 24 ; 25. The economic impact, coupled with the threat to local food systems, could escalate into wider social and economic challenges, particularly for communities heavily reliant on agriculture for their livelihoods and sustenance. This paper aims to dissect the nuances of ESLR, differentiating it from the broader concept of RSLR, and to provide a spatial assessment of potential ESLR impact on coastal and inland agricultural areas under future climate change constraints.
Estimate ESLR Vulnerability Areas Extent
The areas prone to ESLR in Europe and along the coast of North Africa are modelled making use of the Joint Research Centre (JRC) Global Extreme Sea Level projections and Copernicus GLO30 DTM 26 ; 27 ; 28. The ESLR projections incorporate various factors, including RSLR, storm surge and astronomical tides. All these factors are considered in the Baseline Scenario, developed on 1980–2015 observations, and in 4 climatic projections that consider different emission patterns and timeframes. The analysis considers 2 different climate change trends based on distinct Representative Concentration Pathway (RCP) trajectories, RCP 4.5, and RCP 8.5. The first is related to a moderate increase in global temperatures where the climate effect of human activities is dampened by effective mitigation actions 21 .The RCP 8.5 is a more severe scenario in which emissions continue to rise throughout the 21st century. RCP 8.5, generally used as the basis for worst-case climate change scenarios, was initially based on what proved to be an overestimation of projected coal outputs. 29. The 2 timeframes for which the effect of human activities is measured are 2050 and 2100, thus resulting in considering a total of 5 different scenarios. For each scenario 3 different risk probabilities are considered. These probabilities refer to the 5th, 50th and 95th percentile of the ESLR projections using Monte Carlo iterations (from the seminal work of Vousdoukas et al.) 21, which represent the foundation of our analysis 21. These values span a conservative estimate (5th percentile), indicating a lower ESLR; progress through a more probable projection (50th percentile), albeit with increased severity; and extend to an even more severe estimate (95th percentile), evoking a worst-case scenario. We assume the uncertainty as a proxy for vulnerability to generate spatial maps indicating the likelihood of specific regions being impacted by ESLR in the future 20 ; 8. In this work we assume "vulnerability" as the combination of geographical suitable conditions (i.e. low DEM – Digital Elevation Model - coupled with the span of probability of ESLR given the available projections).
The maps of the impacted areas are generated by subtracting the Digital Elevation Model (DEM) with the ESLR projections with 100-year return period. ESLR projections are presented as points on the coastline. Each point contains the sea level in metres for an extreme event that has the 1% possibility to occur each year. The projections refer to 5 climate scenarios and 3-severity levels for a total of 15 possible combinations 30. The spatial association of terrestrial locations with the nearest ESLR estimates along the coast, accomplished through Thiessen tessellation, identifies all raster inland elevation pixels associated with a specific ESLR sample value. Subsequently, subtracting elevation values from projected ESLR estimates allows for the determination of the geographical extent of regions likely to be impacted by future rising oceans 31. This preliminary spatial map is masked with the geographical boundaries of persistent water bodies, i.e., rivers, lakes, and coastline) to not take into consideration areas that are just submerged. In the same way ESLR impacted areas that don't have pixels exhibiting spatial contiguity with the coastline or with permanent water bodies directly connected to it are identified as isolated and eliminated 32. The exclusion of isolates was crucial for achieving a clearer and more robust representation of terrestrial areas prone to submersion under ESLR forecasts for IPCC 30 scenarios RCP 8.5 and 4.5 up to 2050 and 2100. Without this filtering step, in some regions, extensive areas lying below sea level but with no hydrological spatial contiguity with open waters would have been incorrectly included 33.
Historical Insights and Future Projections
The first result stemming from our projections (Fig. 1) is that a higher exposure to ESLR is not linearly linked to a larger area of impacted land. This is due to the morphology and hydrology of the coastline. For example, in the west coast of Ireland ESLR projections are very severe but due to very high and steep cliffs the areas vulnerable to an extreme event are smaller 34. On the contrary there are areas in which the magnitude of the projected extreme phenomenon is not among the most severe, but the consequences can be disastrous, such as in the Nile delta. Figure 1a outlines that there are 5 macro-areas showing a significant vulnerability to ESLR events. The first and most notable one groups the coasts of (i) Belgium, Netherlands, Germany, and Denmark facing the North Sea. This region has a long history of coastal flooding besides the already mentioned and famous North Sea Flood of 1953 35 other events of similar magnitude have occurred over the centuries causing the spread of death and devastation. Notable among these are the St. Lucia's flood of 1287, the St. Marcellus flood of 1362, the 1530 St. Felix's flood, and the Christmas Flood of 1717 36, each illustrating the lethal potential of North Sea storm surges. These events affected a broad swath of Northwest Europe, collectively resulting in approximately 14,000 deaths 35. Many of these storm surges have also had disastrous consequences across the English Channel, identifying the (ii) United Kingdom as another high-risk area. Following this are the (iii) Po Valley, the (iv) western coast of France, and the (v) Nile Delta 37.
These regions are experiencing an increased vulnerability to ESLR due to a confluence of environmental and anthropogenic factors. In the North Sea region and United Kingdom, enhanced storm surges frequently exacerbate sea level rise, compounding the risks to coastal infrastructure and habitats. On the contrary, in the Po Valley and the Nile Delta 37, subsidence due to natural and human-induced processes has resulted in significantly lower land elevations relative to sea level, heightening their vulnerability to flooding. Additionally, these areas suffer from varying degrees of inadequate coastal management, which fails to mitigate the effects of rising sea levels effectively. These factors combine to increase the frequency and severity of flooding 38, posing substantial risks to ecological systems, economic stability, and human populations in these regions 33 ; 36 ; 39.
Considering the five climate change scenarios previously mentioned, we observe that the area under threat of an ESLR event increases with the severity of the climate projection and the time frame considered. Thus - as reasonably expectable - the RCP8.5 scenario is much more severe than the RCP4.5, and similarly 30, projections for 2100 are worse than those for 2050 and the Baseline scenario 29. In the Baseline scenario, the estimated total area under the threat of ESLR along the coasts of Europe and North Africa ranges from 72,077 km² to 81,857 km², with a median of 74,895 km². These values remain relatively stable, with minor increases of 9% and 10% for the RCP4.5 and RCP8.5 scenarios, respectively, averaged across the three risk levels in the 2050 projection timeframe. However, more significant increases are observed in the projections for 2100. Specifically, the exposed area increases by 16% from the Baseline for the RCP4.5 scenario and by 29% for the RCP8.5 scenario 38. Although the Baseline scenario is preferable, we note a minor increase in 2050 without significant differences between the two RCP scenarios across the three risk levels. On the other hand, in the 2100 timeframe, we observe the most notable differences between the RCP8.5 and RCP4.5 scenarios, especially for the most severe events that are unlikely but can have the most destructive effects, leading to an increase of 42% in areas of low vulnerability. Given the escalation of these risks, it is imperative to integrate robust, forward-thinking coastal management strategies that incorporate both mitigation and adaptation measures tailored to these vulnerabilities, ensuring the resilience of affected communities against impending SLR, and associated extreme events 8.
From Land Cover to Crop Losses
Once the areas affected by ESLR have been estimated, further analysis to explore the economic and social impacts were performed. To this end, the ESLR areas are overlaid with Corine Global Land Cover 40, and intersections are computed.
In Fig. 2, we observe a significant prevalence of vulnerable cropland, which accounts for 58% of the total, in comparison to natural (37%) and urban (5%) areas. Additionally, cropland exhibits the least variability across different scenarios, as these areas are typically situated closer to the sea. Consequently, croplands face a greater risk of being impacted by ESLR events, even those of minor magnitude. In contrast, cities show a higher vulnerability to extraordinary events 24, which, although less likely, can have a more severe impact 41. This observation is consistent with strategic urban planning perspectives that position urban areas away from the coast or in locations protected by natural or artificial barriers 42. Therefore, while cities are generally less at risk under ordinary conditions, an extreme climate event could potentially cause substantial damage to infrastructure and pose significant risks to human lives. The heightened vulnerability of cropland to ESLR poses a serious threat to European food security, because of the magnitude of its impact on coastal agricultural lands 43 ; 41. Many of Europe's fertile regions, such as the river deltas in Italy, the Netherlands, and Egypt, are located near seacoasts. As ESLR continues to intensify with climate change, the potential for reduced agricultural output and compromised food supply chains could lead to increased food insecurity across the continent, highlighting the urgent need for resilient agricultural practices and enhanced coastal defences 44.
Assessing the Agricultural Impact of ESLR on Coastal Lands
Delving deeper into the economic analysis, we quantified the impacts of an ESLR event on agricultural production across various climate change scenarios. Utilising the GAEZ 2015 dataset 45, which provides productivity data for 26 crops with a resolution of approximately 8.5 km², we estimated the potential loss in agricultural production that could result from such an event 46. By overlaying ESLR projections onto the GAEZ dataset, we calculated the percentage of each cell vulnerable to ESLR.
We assumed no production in the year of the event and a uniform crop distribution within each cell, which enabled us to estimate the overall loss of agricultural productivity. The accompanying graph illustrates the average loss of agricultural productivity for each nation. Notably, the Netherlands exhibits the highest percentage loss, ranging from 33.6% under the Baseline Scenario to 39.2% in the worst-case scenario. Other countries like Libya, Portugal, Italy, France, Germany, and Albania show losses ranging between 1% and 5% 47.
Our projections suggests that the economic impact of ESLR on agriculture could be severe, particularly in regions where agricultural lands are near the coast. Specifically, the situation in Egypt illustrates how even modest exposure to ESLR can lead to disastrous impacts, given the region's morphology and land use. The Nile Delta, a densely populated and highly modified area, concentrates nearly all the country's agricultural and other productive activities. An extreme ESLR event in this region would have catastrophic consequences, as reflected in our analysis. To estimate the economic impact of the phenomenon, the productivity data for each cell (measured in 1000 tonnes) is multiplied by the price per tonne as recorded by FAOSTAT for each nation 48 . This analysis yields intriguing results, revealing nations that may not appear highly impacted at first glance. Although Egypt remains the foremost in terms of total agricultural losses, The Netherlands, Turkey, France, and Germany also emerge as significantly affected, alongside the United Kingdom and Belgium, primarily due to the specific crops likely to be impacted by ESLR. The Netherlands is potentially significantly affected. However, the country has a long history of managing saltwater intrusion. Major parts of the agricultural lands are situated below mean seal level. River Rhine water is used to push back and reduce saltwater intrusion. Therefore, the projections in this paper will need fine-tuning with more accurate DEM local data for the local situation in The Netherlands. Turkey presents a notable case 49 , with certain areas showing vulnerability to ESLR, such as the district of Bafra. Situated in a low-lying region near the Black Sea coast and traversed by the Kızılırmak River, Bafra is minimally affected by storm surges from the Black Sea 48 . However, projections indicate that some areas are still at risk of flooding. Given that these lands are among Turkey’s most fertile and are utilised for growing high-quality tobacco, the economic repercussions could be substantial. When considering the total value of production vulnerable to ESLR across all crops and nations 50 , the figures amount to $18 million in the best-case scenario and rise to $26 million in the worst-case scenario.
These findings underscore the urgent need for proactive measures to safeguard agricultural productivity in the face of increasing risks from Extreme Sea Level Rise 11. The experience of the Netherlands illustrates that it requires 25 to 50 years to get agreement and implement large scale coastal protection schemes 12. The disparities in potential loss across different nations highlight the importance of region-specific strategies that consider local agricultural practices51, crop types, and the geographic vulnerabilities of each area. For nations like the Netherlands, where significant losses are projected even in less severe scenarios, the implementation of robust coastal defences and the development of salt-tolerant crop varieties could mitigate some of the negative impacts 48. Additionally, enhancing early warning systems and improving regional planning can help to reduce the economic burden on nations with high-risk zones like the Nile Delta and Bafra. Furthermore, recent research revealed that certain species known for biofuel potential cannot solely enable highly adaptive mechanisms to salinity stress (very high on land affected by ESLR), but even thrive in saline soils (i.e. Schrenkiella parvula of the Brassicaceae family) 52, hence being an indication that agricultural strategies aiming at adapting to climate change adaptation may stimulate to reconsider current agricultural pattern, favouring biofuel crops on saline soils, so to reserve non-saline arable lands for cash crops.
Given the economic stakes involved, international cooperation and funding for research into resilient agricultural practices 15 and climate adaptation technologies will be critical 44. These collaborative efforts should aim not only to prevent immediate losses but also to ensure the long-term sustainability 53 of food production systems globally, protecting them against future ESLR events and other climate-related challenges 43.
ESLR Vulnerable area identification
In this study, we utilised the ESLR projections developed by Vousdoukas et al. 2018 21 to delineate areas vulnerable to extreme sea-level rise events across Europe and North Africa. We constructed a raster map with a resolution of 30 metres to highlight these regions at risk. This model incorporates the Copernicus GLO-30 Digital Elevation Model (DEM) 26, a global elevation dataset with an approximate resolution of 30 metres. Individual tiles, in .tif format, were procured from Amazon Web Services (AWS) using Python scripts tailored to the required geographic extent. These tiles were subsequently merged into a unified file and reprojected into the ETRS 1989 LAEA (EPSG: 3035) coordinate system for consistency and analytical precision 27 ; 28.
To integrate ESLR scenarios with a 100-year return period, provided by the Joint Research Centre (JRC) in .csv format, we converted these files into .shp format. To spatially distribute the point data, a lattice of Thiessen polygons (or Voronoi diagram) was constructed. This methodological approach ensures that each area within the model is associated with the nearest coastal projection point, facilitating a continuous spatial analysis 55. The Thiessen polygons were generated using ArcGIS Pro and subsequently rasterized to align with the DEM grid in the same ETRS 1989 LAEA projection. The rationale to identify the ESLR vulnerable areas is:
Given L and S, the two sets representing points on the land and in the sea, respectively, and having defined the functions: \(\:DEM(x,y)\), which identifies the elevation above sea level for each pair of coordinates, and \(\:ESLR\left({p}_{i}\right)\), which associates each point on the coast, \(\:{p}_{i}\), with the relative value of the height of the extreme sea level event, we defined the risk function R function of \(\:DEM(x,y)\) and \(\:ESLR(x,y)\) the Voronoi transformation of \(\:ESLR\left({p}_{i}\right)\). This function identifies vulnerable areas, assigning them value 1, while to non vulnerable areas and sea the value 0 is assigned. To do so the Arcpy Python library was used to compute the differences between the DEM and the ESLR projections. Cells yielding a positive difference were assigned a value of NaN, indicating areas not affected by ESLR, while those with a negative value were marked with a 1, recognizing areas likely to be impacted by ESLR. However, this operation may inadvertently include false positive results, such as depressed areas below sea level, with no spatial connection with open waters whatsoever, and/or situated hundreds of kilometres from the coast. To refine these predictions, isolated areas - regions projected to be impacted by sea level rise but not directly/indirectly connected to the sea - were excluded from the analysis. This was achieved by running the Distance Accumulation tool on the ESLR preliminary map using as source data the European Environment Agency (EEA) coastline shapefile. Lastly, we masked the results to correspond more accurately with existing water bodies, utilising the G1WBM Water Body Map resampled at 30m resolution 2.
This procedure was run for all the 15 combinations of climate change scenarios, timeframes and vulnerability levels producing as output 15 single band tif raster. Each raster has pixels with value 1 that represents the regions under the threat of an ESLR event while the others are assigned to No data. In this way we produce spatially explicit maps to address the consequences of an ESLR event that happens with probability 1% each year.
Land Cover Assessment
Vulnerability mapping offers significant insights for environmental and socio-economic analysis, particularly through the exploration of land cover within areas susceptible to Extreme Sea Level Rise (ESLR). Our initial approach involves a detailed examination of land cover characteristics 42. We employed a methodology that overlays the projected geographic extent of future ESLR projections onto current land-use maps, enabling the assessment of potential impacts across different land-use categories. Specifically, we utilised the Corine Global Land Cover 39dataset, which we reprojected into EPSG:3035, featuring a resolution of approximately 100 metres 23. To streamline the analysis, we consolidated the 20 classes representing natural elements - excluding those designated as 'Built Up' (code 40) and 'Croplands' (code 50) - into a single 'Natural' class, resulting in three primary categories: Built Up, Cropland, and Natural 10.
Subsequently, assuming land-use patterns and associated values would remain constant in the future, the reclassified Corine layer is overlaid with the 15 distinct ESLR projection scenarios. This integration assigns land cover values to the corresponding ESLR layers, facilitating the classification of each pixel based on its vulnerability to sea level rise. Then the Arcpy 'Tabulate Area' function is used to calculate the area covered by the pixels within each class. The results of this spatial analysis were meticulously compiled and stored in a tabular format 29. This refined approach not only enhances our understanding of the potential spatial distribution and intensity of ESLR impacts but also supports robust decision-making for mitigation and adaptation strategies in vulnerable coastal regions highlighting the importance of adopting it also in the agricultural sector.
Agricultural Production Losses and Economic Impacts
Given that 58% of the land affected by extreme sea level rise (ESLR) is agricultural, we proceeded to assess potential land-capability and crop value losses due to such events. For this analysis, we utilised the Global Agro-Ecological Zones (GAEZ) 2015 dataset. This dataset offers global, gridded data (at 5-arcminute resolution) on irrigated and rainfed crop areas, production, and yield across 26 different crops and crop categories, based on national statistics. Firstly, the .tif files from GAEZ 2015 were reprojected into the ETRS 1989 LAEA (EPSG: 3035) coordinate system. Using the Python GDAL library, we quantified the ESLR vulnerable cells (~ 30m) within each GAEZ cell (~ 8.6km). Then we calculate the complement of the percentage for each GAEZ cell vulnerable to ESLR. This results in a raster ranging from 0 to 1 where cells with value 1 indicate no ESLR impact, whereas those with value 0 would be completely affected by an ESLR event. This derived raster was then multiplied by the GAEZ productivity raster to estimate agricultural productivity losses for each of the 15 climate change scenarios and across the 26 crops 47. This calculation assumes a uniform distribution of cultivated areas within each cell and a constant crop mix over time. These assumptions enabled us to estimate, with reasonable accuracy, the total agricultural production for each nation and the corresponding losses attributable to ESLR events 43.
To estimate the economic impact of an ESLR event we multiplied the raster of the agricultural productivity (in 1000 tonnes per pixel) for each of the 15 scenarios by the producer price per tonne of each crop by nation. The price of the 26 crops or group of crops are downloaded by FAOSTAT, and the crop categories are created following the GAEZ 2015 metadata. Then the zonal statistics are calculated to aggregate the data at national level (NUTS-0). Other descriptive statistics are calculated to complete the analysis and the results are displayed in maps. Interesting results come out in poorer regions that with a good coastal land management could lower the risks linked to an ESLR event 18.
Future perspectives and key limitations
This study has crucially delineated the regions vulnerable to Extreme Sea Level Rise (ESLR), fostering a deeper understanding of the potential risks and enabling the development of targeted strategies for coastal land management and mitigation 12. Through meticulous analysis using cutting-edge geographic and economic models, our findings highlight the profound impacts of ESLR on coastal regions, with a particular focus on agricultural lands, which encompass 58% of the areas at risk. Our research underscores the critical need for efficient mitigation strategies to reduce exposure and enhance resilience, particularly in agricultural zones that sustain significant economic and social value. The integration of ESLR projections with the GAEZ dataset has provided a novel perspective on the potential economic impacts, illustrating not only the immediate threats but also the long-term challenges to food security and regional economic stability. These impacts are especially pronounced in regions like the Nile Delta, where even modest ESLR events can have catastrophic consequences due to the area's dense population and agricultural reliance 13.
Furthermore, the methodological advancements in this study offer a robust framework for future research and policymaking. By combining high-resolution DEM data with detailed crop and economic data, we have enabled more precise mapping of ESLR vulnerability and its economic implications. This approach not only enhances our understanding of the spatial distribution of risks but also supports the formulation of region-specific adaptation strategies, which are essential in the face of increasing ESLR incidents. However, the study is not without limitations 18. The resolution of the global Digital Elevation Model (DEM) employed, while among the highest currently available, may inadequately represent smaller-scale geographic features that influence vulnerability assessments, particularly anthropogenic structures such as embankments, dykes, and water gates designed to prevent seawater intrusion. The GLO 30 DEM exhibits limitations in detecting coastal defensive structures, exemplified by the Delta Works in the Netherlands, a pivotal system constructed to protect the nation's coasts. Therefore, our analysis likely overestimates the effects of Extreme Sea Level Rise (ESLR) in highly developed countries with sophisticated coastal management systems, such as the Netherlands, Germany, and Denmark. Conversely, in Mediterranean countries like Egypt, Italy, and Spain, there is a notable deficiency in large-scale coastal protection infrastructure and a lack of awareness regarding their importance. This absence of protective measures and understanding may result in more accurate vulnerability assessments for these regions. Additionally, the static nature of the crop distributions and economic valuations used in our models does not account for potential adaptations or changes in agricultural practices over time, which could alter the outcomes under different climate scenarios. Considering these findings and limitations, it is imperative that future research continues to refine the spatial and economic models of ESLR impact 19. Enhanced modelling precision, coupled with dynamic economic and agricultural data, will be crucial in developing more effective adaptation and mitigation strategies. A higher resolution DEM that would also consider artificial dykes and sea walls would return more faithful and current predictions (although, to the best of our knowledge, this is not available for the scale of analysis implemented in this work). These strategies must be tailored not only to the physical and economic landscapes but also to the socio-cultural contexts of the affected regions, ensuring that the most vulnerable communities are equipped to withstand the challenges posed by climate change. By providing a comprehensive assessment of the risks and a detailed blueprint for mitigation, this study contributes a critical piece to the puzzle of climate resilience 54. It calls for an integrated approach that combines scientific inquiry with practical policymaking 55, leveraging international cooperation and innovation to safeguard our most vulnerable coastal regions against the impending challenges of Extreme Sea Level Rise 6 ; 21.
Acknowledgement
We would like to acknowledge that the data and results generated in this study are available upon request, free of charge, to all non-profit scientists and academics. Please contact the corresponding author for access. This work was supported by the SALAD project, funded by the ERA‐NET Cofund FOSC program, which has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 862555 (https://www.saline-agriculture.com/en), and by the EUniWell Well-Being Research Incubator Programme of the EUniWell Consortium (European Universities for Wellbeing, https://www.euniwell.eu/).
- Becker, M., Karpytchev, M. & Hu, A. Increased exposure of coastal cities to sea-level rise due to internal climate variability. Nat Clim Chang 13, 367–374 (2023).
- Horton, B. P. et al. Estimating global mean sea-level rise and its uncertainties by 2100 and 2300 from an expert survey. NPJ Clim Atmos Sci 3, 18 (2020).
- Taherkhani, M. et al. Sea-level rise exponentially increases coastal flood frequency. Sci Rep 10, 6466 (2020).
- Becker, M., Karpytchev, M. & Papa, F. Hotspots of Relative Sea Level Rise in the Tropics. in Tropical Extremes 203–262 (Elsevier, 2019). doi:10.1016/B978-0-12-809248-4.00007-8.
- Rashid, Md. M., Wahl, T., Chambers, D. P., Calafat, F. M. & Sweet, W. V. An extreme sea level indicator for the contiguous United States coastline. Sci Data 6, 326 (2019).
- Tebaldi, C. et al. Extreme sea levels at different global warming levels. Nat Clim Chang 11, 746–751 (2021).
- Dang, A. T. N., Reid, M. & Kumar, L. Assessing potential impacts of sea level rise on mangrove ecosystems in the Mekong Delta, Vietnam. Reg Environ Change 22, 70 (2022).
- Vousdoukas, M. I. et al. Economic motivation for raising coastal flood defenses in Europe. Nat Commun 11, 2119 (2020).
- Alfieri, L. et al. Global projections of river flood risk in a warmer world. Earths Future 5, 171–182 (2017).
- Gregory, J. M. et al. Concepts and Terminology for Sea Level: Mean, Variability and Change, Both Local and Global. Surv Geophys 40, 1251–1289 (2019).
- de Ruig, L. T. et al. An economic evaluation of adaptation pathways in coastal mega cities: An illustration for Los Angeles. Science of The Total Environment 678, 647–659 (2019).
- Wassmann, R., Hien, N. X., Hoanh, C. T. & Tuong, T. P. Sea Level Rise Affecting the Vietnamese Mekong Delta: Water Elevation in the Flood Season and Implications for Rice Production. Clim Change 66, 89–107 (2004).
- Nazarnia, H., Nazarnia, M., Sarmasti, H. & Wills, W. O. A Systematic Review of Civil and Environmental Infrastructures for Coastal Adaptation to Sea Level Rise. Civil Engineering Journal 6, 1375–1399 (2020).
- Dawson, D., Shaw, J. & Roland Gehrels, W. Sea-level rise impacts on transport infrastructure: The notorious case of the coastal railway line at Dawlish, England. J Transp Geogr 51, 97–109 (2016).
- Du, S. et al. Hard or soft flood adaptation? Advantages of a hybrid strategy for Shanghai. Global Environmental Change 61, 102037 (2020).
- Ann Conyers, Z., Grant, R. & Roy, S. Sen. Sea Level Rise in Miami Beach: Vulnerability and Real Estate Exposure. The Professional Geographer 71, 278–291 (2019).
- Corwin, D. L. Climate change impacts on soil salinity in agricultural areas. Eur J Soil Sci 72, 842–862 (2021).
- Katarzyna Negacz, Žiga Malek, Arjen de Vos & Pier Vellinga. Saline soils worldwide: Identifying the most promising areas for saline agriculture. J Arid Environ 203, (2022).
- Arabadzhyan, A. et al. Climate change, coastal tourism, and impact chains – a literature review. Current Issues in Tourism 24, 2233–2268 (2021).
- Hinkel, J., Lincke, D., Vafeidis, A. T., Perrette, M., Nicholls, R. J., Tol, R. S., ... & Levermann, A. (2014). Coastal flood damage and adaptation costs under 21st century sea-level rise. Proceedings of the National Academy of Sciences, 111(9), 3292-3297.
- Vousdoukas, M. I. et al. Global probabilistic projections of extreme sea levels show intensification of coastal flood hazard. Nat Commun 9, 2360 (2018).
- Monioudi, I. Ν. et al. Climate change impacts on critical international transportation assets of Caribbean Small Island Developing States (SIDS): the case of Jamaica and Saint Lucia. Reg Environ Change 18, 2211–2225 (2018).
- Calafat, F. M., Wahl, T., Tadesse, M. G. & Sparrow, S. N. Trends in Europe storm surge extremes match the rate of sea-level rise. Nature 603, 841–845 (2022).
- Ataie-Ashtiani, B., Werner, A. D., Simmons, C. T., Morgan, L. K. & Lu, C. How important is the impact of land-surface inundation on seawater intrusion caused by sea-level rise? Hydrogeol J 21, 1673–1677 (2013).
- Eini, M., Kaboli, H. S., Rashidian, M. & Hedayat, H. Hazard and vulnerability in urban flood risk mapping: Machine learning techniques and considering the role of urban districts. International Journal of Disaster Risk Reduction 50, 101687 (2020).
- Copernicus. GLO-30 DEM. (2021).
- Cuellar, A. C., Cenci, L., Santella, C. & Albinet, C. Evaluating the Copernicus Dem Dataset Potential for the Identification of (Flash) Flood-Prone Areas by Using a Geomorphological Approach. in IGARSS 2022 - 2022 IEEE International Geoscience and Remote Sensing Symposium 5997–6000 (IEEE, 2022). doi:10.1109/IGARSS46834.2022.9884948.
- Meadows, M., Jones, S. & Reinke, K. Vertical accuracy assessment of freely available global DEMs (FABDEM, Copernicus DEM, NASADEM, AW3D30 and SRTM) in flood-prone environments. Int J Digit Earth 17, (2024).
- Riahi, K. et al. RCP 8.5—A scenario of comparatively high greenhouse gas emissions. Clim Change 109, 33–57 (2011).
- Kim, J., Choi, J., Choi, C. & Park, S. Impacts of changes in climate and land use/land cover under IPCC RCP scenarios on streamflow in the Hoeya River Basin, Korea. Science of The Total Environment 452–453, 181–195 (2013).
- Poulter, B. & Halpin, P. N. Raster modelling of coastal flooding from sea‐level rise. International Journal of Geographical Information Science 22, 167–182 (2008).
- Tromble, E. et al. Aspects of Coupled Hydrologic-Hydrodynamic Modeling for Coastal Flood Inundation. in Estuarine and Coastal Modeling (2009) 724–743 (American Society of Civil Engineers, Reston, VA, 2010). doi:10.1061/41121(388)42.
- Mehta, D., Dhabuwala, J., Yadav, S. M., Kumar, V. & Azamathulla, H. M. Improving flood forecasting in Narmada river basin using hierarchical clustering and hydrological modelling. Results in Engineering 20, 101571 (2023).
- BRÁZDIL, R., KUNDZEWICZ, Z. W. & BENITO, G. Historical hydrology for studying flood risk in Europe. Hydrological Sciences Journal 51, 739–764 (2006).
- Gerritsen, H. What happened in 1953? The Big Flood in the Netherlands in retrospect. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 363, 1271–1291 (2005).
- Blöschl, G. et al. Current European flood-rich period exceptional compared with past 500 years. Nature 583, 560–566 (2020).
- Kellner, A., Brink, G. J. & El Khawaga, H. Depositional history of the western Nile Delta, Egypt: Late Rupelian to Pleistocene. Am Assoc Pet Geol Bull 102, 1841–1865 (2018).
- Vousdoukas, M. I. et al. Developments in large-scale coastal flood hazard mapping. Natural Hazards and Earth System Sciences 16, 1841–1853 (2016).
- KEMP, A. C. & HORTON, B. P. Contribution of relative sea‐level rise to historical hurricane flooding in New York City. J Quat Sci 28, 537–541 (2013).
- Büttner, G. CORINE Land Cover and Land Cover Change Products. in 55–74 (2014). doi:10.1007/978-94-007-7969-3_5.
- Brath, A., Montanari, A. & Moretti, G. Assessing the effect on flood frequency of land use change via hydrological simulation (with uncertainty). J Hydrol (Amst) 324, 141–153 (2006).
- Wolff, C., Nikoletopoulos, T., Hinkel, J. & Vafeidis, A. T. Future urban development exacerbates coastal exposure in the Mediterranean. Sci Rep 10, 14420 (2020).
- Naef, F., Scherrer, S. & Weiler, M. A process based assessment of the potential to reduce flood runoff by land use change. J Hydrol (Amst) 267, 74–79 (2002).
- Dar, M. H. et al. Transforming rice cultivation in flood prone coastal Odisha to ensure food and economic security. Food Secur 9, 711–722 (2017).
- Global Agro-Ecological Zone V4 – Model Documentation. (FAO, 2021). doi:10.4060/cb4744en.
- Grogan, D., Frolking, S., Wisser, D., Prusevich, A. & Glidden, S. Global gridded crop harvested area, production, yield, and monthly physical area data circa 2015. Sci Data 9, 15 (2022).
- Hunter, J. Estimating sea-level extremes under conditions of uncertain sea-level rise. Clim Change 99, 331–350 (2010).
- Reed, C. et al. The impact of flooding on food security across Africa. Proceedings of the National Academy of Sciences 119, (2022).
- Aksoy, H., Demirel, H. & Seker, D. Z. Exploring possible impacts of sea level rise: the case of Izmir, Turkey. International Journal of Global Warming 13, 398 (2017).
- Hall, J. W., Sayers, P. B., Walkden, M. J. A. & Panzeri, M. Impacts of climate change on coastal flood risk in England and Wales: 2030–2100. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 364, 1027–1049 (2006).
- Raihan, A. A review of the global climate change impacts, adaptation strategies, and mitigation options in the socio-economic and environmental sectors. Journal of Environmental Science and Economics 2, 36–58 (2023).
- Sun, Y., Oh, DH., Duan, L. et al. Divergence in the ABA gene regulatory network underlies differential growth control. Nat. Plants 8, 549–560 (2022). https://doi.org/10.1038/s41477-022-01139-5
- Wahl, T. et al. Understanding extreme sea levels for broad-scale coastal impact and adaptation analysis. Nat Commun 8, 16075 (2017).
- Yunus, A. et al. Uncertainties in Tidally Adjusted Estimates of Sea Level Rise Flooding (Bathtub Model) for the Greater London. Remote Sens (Basel) 8, 366 (2016).
- Wrathall, D. J. et al. Meeting the looming policy challenge of sea-level change and human migration. Nat Clim Chang 9, 898–901 (2019).
There is NO Competing Interest.