By Vinij Tansakul
Over the past decade, aquaculture across the Southeast Asia-Pacific region repeatedly absorbed damage from extreme climate events, such as saltwater intrusion in the Mekong Delta, severe droughts that pushed pond water levels to abnormal lows, and sudden heavy rainfall that shifted water quality so rapidly that stocked animals became stressed and died end masse. What stands out across many of these cases is that farmers were not unaware of the risks. What they lacked was a systematic tool to assess, prioritize, and plan climate risk at the level of their farm, in concrete and actionable terms.
Survey research across the Mekong Region shows that most shrimp farmers assess climate risk primarily through the perception-based approach of a lived experience, rather than through a systematic, indicator-based assessment. While this reflects valuable local knowledge, it suffers from this critical limitation: it cannot be compared across locations, time periods, or farming systems, which is precisely the kind of information needed to plan effective adaptation investment, whether at the level of an individual farm or at the level of policy.
This article presents the concept, structure, and components of a Farm-Level Climate Risk Assessment Toolkit, built on internationally recognized frameworks and adapted for practical use. The goal is to give farmers, extension officers, and policymakers across the Southeast Asia-Pacific region a means to translate abstract climate risk into farm-level information they can act on.
The Industry’s Challenge: Gap Between Risk Perception and Systematic Risk Assessment
Climate risk to aquaculture farms is more complex than commonly understood, because it does not arise from climate alone. It emerges from the interaction of three components: hazard, exposure, and vulnerability of the production system. In practice, this perception gap manifests itself in three distinct ways.
First, many smallholder farmers in the region still assess risk primarily through experience. Survey research on shrimp farms in the Mekong Delta found that the risk factors farmers cite most frequently are rapid temperature shifts, abnormally heavy rainfall, drought, and saltwater intrusion; yet their awareness is rarely converted into quantitative indicators that can be tracked or compared systematically over time.
Second, different farming systems carry markedly different levels of vulnerability even when faced with identical hazard. Comparative research on extensive versus semi-intensive and intensive shrimp farming systems in the Mekong Delta found that the impacts of environmental shocks increased significantly with the intensity of farm management. This confirms that any risk assessment tool that works must be context-adaptable to each production system and not be applied as a single uniform set of indicators across every farm.
Third, risk is not distributed evenly across space or time. Hydrological research on saltwater intrusion in the Mekong Delta shows that the severity of this phenomenon fluctuates sharply from year to year, particularly during El Niño events, creating high uncertainty for seasonal farming and cropping decisions. Farmers therefore need a tool that can pinpoint seasonal and location-specific risk hotspots, not merely a general warning that “the climate is changing.”
These three gaps are precisely why aquaculture in the region needs a risk assessment tool that is systematic, measurable, and deployable at the level of the individual farm, not merely an academic framework that only functions at national or regional scale.
A Scientific Framework: From Vulnerability to Risk, Following the IPCC Approach
The Evolution of Climate Risk Frameworks
Knowledge of assessing climate risk to fisheries and aquaculture has evolved considerably. The original framework proposed by the Intergovernmental Panel on Climate Change (IPCC) in 2001 defined “vulnerability” as the sum of three components: exposure, sensitivity, and adaptive capacity. In its Fifth Assessment Report (AR5) in 2014, the IPCC revised this framework to be more comprehensive, defining “risk” as arising from the interaction of three principal components.
(1) Hazard: a climate-related event or trend that could cause harm, such as heatwave, abnormally heavy rainfall, storm, or saltwater intrusion.
(2) Exposure: the extent to which a production system (farm, stock, infrastructure) is situated in a location or at a time where it could be affected by that hazard.
(3) Vulnerability: under this revised framework, vulnerability is narrowed to comprise a system’s sensitivity to the hazard counterbalanced by its capacity to adapt to or recover from the resulting impact.
This shift in framing carries real practical value. It draws a clear line between what is “uncontrollable” (the climate hazard itself) and what is “manageable at the farm level” (the reduction of exposure through site or seasonal choices and the strengthening of adaptive capacity). This distinction is the core design principle that any farm-level risk assessment tool must build in, so that users can clearly separate the two and plan their response accordingly.
Application to Aquaculture
A national-scale climate risk assessment for aquaculture conducted in Oman offers a concrete example of applying the IPCC AR5 framework in practice. That study broke risk down into four practical components: (1) species-specific thermal sensitivity, (2) exposure to flooding and storm surge, (3) hazard from low-oxygen conditions, and (4) vulnerability to disease outbreaks. The study found that shrimp culture carried the highest risk, driven by high pond exposure to flooding and storm surge combined with high vulnerability to disease.
This approach reflects a key design principle carried into farm-level tools: risk must be broken down into measurable sub-components specific to each farming system and species, rather than be assessed as a single, broad-brush risk level that offers no clear basis for practical recommendations.
Components of a Farm-Level Climate Risk Assessment Toolkit
An effective farm-level climate risk assessment toolkit should comprise five core modules, linked as a continuous process running from hazard identification through to actionable output.
Module 1: Hazard Identification
The first step is to compile climate hazard data relevant to the farm’s location, classified into sudden-onset hazards (storms, flash floods, heatwaves) and slow-onset hazards (saltwater intrusion, sea-level rise, shifting rainfall patterns). This module should draw on both historical climate data and medium- to long-term climate projections for specific locations.
In the context of the Mekong Delta and its coastal aquaculture farms, research identifies saltwater intrusion as the most significant hazard, with a severity that depends on the upstream freshwater discharge and that intensifies markedly during El Niño events. In Thailand’s northern inland freshwater aquaculture areas, a survey of tilapia pond farms found that the principal risks were sudden water temperature shifts and extended dry spells.
Module 2: Exposure Mapping
This module assesses the degree to which a farm is “exposed” to the hazards identified, based on spatial and temporal factors: distance from the coastline or river mouth, farm elevation relative to sea level, proximity to natural buffer zones such as mangroves or wave breaks, and culture season during periods of peak risk. Exposure assessment matters because it is a factor farmers can partly manage themselves, through pond siting, dyke design, and adjustment to the culture calendar to avoid high-risk windows.
Module 3: Sensitivity Assessment
This module examines the specific characteristics of a production system that make it more or less affected by hazard. Key indicators include the species cultured and its environmental tolerance range (optimal salinity or temperature bands), the culture system (earthen pond, cage, closed system), the stocking density, and the farm infrastructure such as its water-level control capacity. For example, white leg shrimp tolerate a wider salinity range than giant freshwater prawn, meaning the two systems have markedly different sensitivities to saltwater intrusion as a hazard.
Module 4: Adaptive Capacity Assessment
This module is the critical piece most often overlooked in conventional risk assessment, because adaptive capacity depends not only on a farm’s physical attributes but also on the farmer’s economic and social capital. A widely applied framework here is the Five Capitals Framework.
| Capital Type | Example Indicators in an Aquaculture Farm Context |
|---|---|
| Human Capital | Knowledge of climate risk management, farming experience, access to weather forecast information |
| Natural Capital | Water source quality, presence of natural buffer areas such as mangroves |
| Physical Capital | Farm infrastructure, aeration systems, backup aeration, backup power, road access to the farm |
| Social Capital | Farmer group membership, information-sharing networks, access to extension services |
| Financial Capital | Cash reserves, access to credit, crop/aquaculture insurance coverage |
This framework was applied to construct a Livelihood Vulnerability Index (LVI-IPCC) in a case study of shrimp farming communities in Tra Vinh province, Vietnam. Spanning all five capitals, the index used as many as 42 sub-indicators demonstrating thereby that a genuinely comprehensive assessment of adaptive capacity requires socio-economic indicators alongside physical ones.
Module 5: Risk Rating and Actionable Output
Once data from the four modules above has been gathered, the final step is to combine the risk levels according to this principle:
Risk rating = Hazard risk level + Exposure risk level + Vulnerability risk level
Where vulnerability here is calculated as sensitivity offset by adaptive capacity (that is, the higher the adaptive capacity, the lower the overall vulnerability at a given level of sensitivity). Results are typically presented as a Risk Matrix that classifies risk into low, low-to-moderate, moderate, high, and very high categories, allowing users to prioritize response measures concretely.
| Exposure Level | Low Vulnerability | Moderate Vulnerability | High Vulnerability |
|---|---|---|---|
| Low Exposure | Low Risk | Low-Moderate Risk | Moderate Risk |
| Moderate Exposure | Low-Moderate Risk | Moderate Risk | High Risk |
| High Exposure | Moderate Risk | High Risk | Very High Risk |
The Food and Agriculture Organization (FAO) developed its Climate Risk Toolbox (CRTB) to screen climate risk for agricultural projects and investments worldwide. Its design principles follow a similar level-aggregation logic and explicitly emphasize that assessment output must lead to concrete adaptation measures, not merely to a risk rating with no actionable recommendation attached.
Regional Evidence: Lessons Shaping the Toolkit’s Design
Developing a farm-level risk assessment toolkit for the Southeast Asia-Pacific region requires a grounding in empirical lessons from the field. Key findings include the following.
High Year-to-Year Volatility of Saltwater Intrusion in the Mekong Delta
Hydrological research shows that salinity levels in the southern Mekong Delta have varied substantially from year to year, particularly during the El Niño events of 2010, 2015-2016, and 2019-2020, when severe drought coincided with saltwater intrusion. The principal driver of intrusion severity is the low volume of freshwater discharge from upstream, shaped by both climate variability and upstream reservoir management. This finding indicates that an effective risk assessment tool must connect to near-real-time water-level and salinity data, not rely solely on long-term averages.
Farm Intensity Significantly Affects Vulnerability
Research on shrimp production impacted by environmental events in Vietnam’s Mekong Delta found that intensively and extensively managed farms were affected by environmental shocks in different ways and to different degrees. This confirms that “farming system” is a key variable that must be built into a tool’s sensitivity module.
Risk Perception Directly Shapes Adaptive Behavior
Research on shrimp farmer decision-making in the Mekong Delta found that knowing about the advent of irregular weather was a significant driver on farmers’ decisions to harvest early to reduce losses. Conversely, farmers who knew about drought warnings were less likely to harvest early because drought is a familiar condition in the region with established, routine coping mechanisms already in place. This underscores why the adaptive capacity module includes indicators of farmers’ accumulated experience with each specific hazard type.
Engineering Measures Carry Varying Economic Returns
A benefit-cost analysis of measures to reduce flood risk for shrimp farms in Thailand found that measures such as raising pond dykes or installing backup drainage systems carried different economic returns depending on a location’s baseline risk level. This confirms that a farm-level risk assessment tool should not stop at scoring risk but should consider an assessment of the cost-effectiveness of feasible response measures at each risk level.
The Economic Cost of Inaction: Concrete Cases for Investing in Risk Assessment Tools
To make the case for investing in risk assessment tools more tangible, it is worth examining the real economic damage from past events; this is damage that, in most cases, could have been substantially reduced with more effective risk assessment and advance warning systems in place.
The 2011 Mega Flood
The 2011 Mega Flood was an expensive lesson about the lack of spatial early-warning systems. Thailand’s 2011 floods caused an estimated 1.43 trillion baht (roughly US$46.5 billion) in national economic losses, according to the World Bank estimates. It was one of the costliest natural disasters recorded globally. Losses to the affected agricultural and industrial sectors alone were estimated at 600 billion baht.
Within the aquaculture sector, the Bang Pakong River Basin in Chachoengsao Province (the country’s largest inland shrimp-farming area) was especially hard hit when large numbers of shrimp escaped from ponds during the flooding. Since then, the area has continued to experience recurrent flash flooding almost every year.
A follow-up study mapped flood vulnerability for shrimp farms in the Bang Pakong Basin using a GIS-based multicriteria evaluation with clear quantitative thresholds. It found that the areas identified as most vulnerable on the resulting map matched precisely the areas that were affected by the 2011 flood. Shrimp farm areas became highly vulnerable once 10-day cumulative rainfall exceeded 250-300 millimeters. This is a powerful illustration of the following point: had a spatial risk assessment tool with this kind of quantitative warning threshold been in place before 2011, farmers in high-risk areas could have received location-specific warnings in advance and could have planned accordingly.
The 2010-2016 Shrimp Disease Crisis
Although Early Mortality Syndrome or Acute Hepatopancreatic Necrosis Disease (EMS also known as AHPND) was primarily caused by a bacterial pathogen rather than a direct climate factor, it remains an instructive example of the cumulative impacts of “systemic vulnerability” in the absence of an early risk surveillance tool.
An economic cost analysis of shrimp disease in Asia estimated that cumulative losses to Thailand’s shrimp industry between 2010 and 2016 exceeded US$11.58 billion. Export losses reached US$4.2 billion and more than 100,000 jobs were lost in the sector.
In the first half of 2013 alone, a year marked by severe drought coinciding with the disease outbreak, industry operators estimated revenue losses of 16 billion baht from falling shrimp shipments, a roughly 30 percent year-on-year decline. While EMS’s primary cause was not climatic, this case illustrates that when a production system faces compounding stress from multiple factors simultaneously, climatic and biological alike, economic damage can escalate rapidly. This is precisely why a risk assessment tool’s sensitivity module should consider multiple risk dimensions together rather than in isolation.
A Regional Comparison
To illustrate risk at the scale of an individual farm, research on the impact of Cyclone Bulbul on shrimp farms in Bangladesh, a coastal context broadly comparable to Southeast Asia, found that affected farms suffered average losses of roughly US$4,633 per farm from pond flooding and dyke damage, with 14 percent of farms totally destroyed, 57 percent heavily damaged, and 29 percent moderately damaged. Farm-level figures of this kind are especially useful for communicating risk with smallholder farmers, since they translate abstract national-level damage totals into something tangible at the household scale.
| Case | Location/Period | Damage Value | Key Design Implication |
|---|---|---|---|
| Mega Flood | Thailand, nationwide (2011) | 1.43 trillion baht (national) | Need for quantitative early-warning thresholds (e.g. 10-day cumulative rainfall) |
| Recurrent Flooding, Bang Pakong Basin | Chachoengsao Province (2011-present) | Recurs almost annually | Spatial vulnerability maps accurately predicted high-risk areas |
| EMS/AHPND Disease Crisis | Thailand, nationwide (2010-2016) | US$11.58 billion (cumulative) | Compounding multi-factor stress multiplies economic damage |
| Cyclone Bulbul | Bangladesh, farm level (2019) | US$4,633/farm | Farm-level figures communicate risk concretely to farmers |
These figures paint a clear picture: the cost of not having a systematic risk assessment tool is far higher than the cost of developing one. While no study has yet calculated a benefit-cost ratio for an investment in climate risk assessment specific to aquaculture in Southeast Asia, comparative research on coastal flood defense in Europe found that efficient investment in raising flood dikes could avoid as much as 83 percent of the flood damage, with an average benefit-cost ratio of 8.3 to 14.9 times the investment. While that evidence comes from outside both the aquaculture sector and the region, it offers a useful comparative benchmark: preventive investment generally delivers substantially better returns than absorbing damage after the fact.
Connections to Nature-based Solutions (NbS)
A farm-level climate risk assessment toolkit functions as the “starting point” of the Nature-based Solutions adaptation process, rather than as a solution. That is, the tool identifies which farms should be prioritized for which nature-based interventions. For example: (1) farms with high exposure risk to flooding and storm surge should be prioritized for mangrove restoration or natural wave-break rehabilitation; (2) farms with high sensitivity to saltwater intrusion but low adaptive capacity may be well suited to integrated culture systems, such as rice-shrimp rotation, which help diversify economic risk; and (3) farms with low social capital (lacking farmer groups or information networks) should be supported in forming farmer associations and information-sharing systems before an infrastructure investment is made.
Through this approach, a risk assessment tool becomes more than a measurement instrument. It becomes a strategic prioritization mechanism that allows the limited resources of development organizations, government agencies, and programs to be directed toward the areas and measures offering the greatest adaptation return. This aligns with the principle that FAO emphasizes in developing its global climate risk screening tools: assessment results must translate into prioritizing the areas and measures with the highest vulnerability first.
Practical Recommendations
For Smallholder Farmers: Begin by systematically recording unusual climate events affecting the farm (date, duration, damage incurred), even in simplified form. Farm-level historical records are the single most important raw material for hazard-identification and sensitivity modules. This is data that extension agencies typically can only access when farmers record it themselves.
For Extension Agencies and Researchers: Develop risk assessment tools in an easy-to-use format, such as a field questionnaire which an extension officer can complete together with a farmer in 30-45 minutes per farm. The tool should be designed to be used offline at first, so it can function in areas without internet access, and then sync to a central database once signal is available. This is essential for reaching smallholder farmers in remote areas.
For Policymakers and Development Organizations: Use farm-level risk assessments aggregated across multiple locations to build regional risk maps that inform policy decisions. For example, to identify pilot sites for mangrove restoration programs or to design climate insurance mechanisms priced according to each location’s actual risk level.
A Point for Consideration: A risk assessment tool is not a perfectly accurate forecast for the future. It is a prioritization instrument built on the best data currently available. Users should therefore revisit and update assessments periodically, particularly when new climate data becomes available or when a farm undergoes a significant change in its production structure.
Conclusion
A farm-level climate risk assessment toolkit is more than an academic framework. It is a bridge between macro-scale climatological knowledge and practical, on-the-ground observations farmers make. By breaking risk down into assessable components (hazard, exposure, sensitivity, and adaptive capacity), the toolkit enables both individual farmers and supporting institutions to prioritize adaptation investment rationally, rather than spread limited resources evenly across all locations regardless of their actual risk level.
In a Southeast Asia-Pacific region facing intensifying climate variability, investing in the development and dissemination of a practical, accessible farm-level risk assessment toolkit (one explicitly linked to NbS measures) is not merely a nice-to-have. It is essential knowledge infrastructure for building a resilient and sustainable aquaculture sector over the long term.
AQUADAPT 