World Resource Institute

Information Needs for Climate Risk Management

By Molly Hellmuth, Director of the Climate and Society publication series; International Research Institute of Climate and Society, Columbia University

Question Six: What types of information are needed for adaptation decision making?

The ability to manage climate risk is a measure of development. Proactive policymaking and planning means taking action now to prevent and/or reduce future anticipated impacts, and can include no regrets strategies such as technologies and practices for improved efficiency. However, an exclusive focus on ‘no regrets’ strategies can lead to the false idea that there are no trade-offs. The author argues that information needs required to inform decision making for responsive, proactive and robust policymaking and planning are overlapping, particularly as all three rely on the identification of vulnerabilities, hazards and risk as foundations for identifying strategic decisions and alternatives. She then seeks to answer key questions related to information needs.

Better protection from, and resilience to, climate variability is a clear measure of development. Climate variability and extremes, such as floods, droughts and storms, severely affect livelihoods, economic performance and key assets.  The impacts of such events can have a disproportionate effect on developing economies and people, and can be a major set-back to economic growth and development[1] [2],[3]. In some economies reliant on climate-sensitive natural resources, such as agriculture, economic growth actually tracks annual rainfall variability[4]. Climate change threatens to exacerbate these challenges and increase the burden on decision makers as they try and make sense of what to do differently given the perceived enormity of the challenge and the paucity of relevant and reliable information upon which to act.

Adaptation has lagged behind mitigation efforts in part because the information needs for adaptation are problem and context specific. Climate change models developed initially to understand the implications of global development scenarios (i.e. energy consumption patterns) on our climate were appropriated early on to inform place-based sector impacts and adaptation strategies. While useful for understanding broad scale implications, the practical limitation to this approach is that the set of climate information developed to answer the question; given different development scenarios, how might man impact climate at the end of the century is not the same set of information that is needed to answer; how can water shortages be avoided in (e.g.) Niamey today and over the next 20-30 years.

The starting point for adaptation is to get a better understanding of what information is valuable to better manage climate risks today and tomorrow in a particular locale, given a particular problem. Of course, real societal benefits from this information can only be realized if the information is actionable (that is only if there exists in parallel the infrastructure, communication technology, active institutional relationships and policies and the capacity to produce, interpret and act on the information at scale).

This paper focuses solely on the information needs of decision-makers for improved climate risk management for responsive, proactive and robust policy making and planning. Despite different planning horizons and/or strategies, the information needs for all three types of decisions is often overlapping; particularly as all three types of policy making and planning share the common objectives of identifying and characterizing vulnerabilities, hazards and climate-related risks.

Responsive policymaking and planning

Recent flooding in Pakistan during the summer of 2010 displaced millions, killed nearly 2000 people, and set back numerous development goals. The Pakistan flood, while devastating to many people, does not necessarily stand out in the region in terms of the number of people affected or the number of casualties for a climate-related disaster. What does stand out is the occurrence of such a low probability event (a 1000 year return period in the upper catchment) which was without precedent in the known historical record, and was completely unanticipated[5]. As a result, preparedness for flooding was low and consequences were high. In this era of global change, decision makers will need to respond more quickly to such surprises.

In the case of disasters, response measures are those which are taken immediately prior to and following a disaster and are directed towards saving life, protecting property and dealing with the immediate damage caused by the disaster[6]. However, disaster response is most efficient and effective if contingency plans and policies are already in place to inform decision makers; thus, disaster response necessarily includes aspects of preparedness and advance planning. Contingency and emergency response plans involve delineating response needs and priority actions given different scenarios of risks, vulnerabilities and impacts.  Response measures should also be forward looking, i.e. promoting actions which will both minimize or mitigate impact and speed the recovery process.

Proactive policymaking and planning 

The ability to manage climate risk is a measure of development, and currently challenges developed and developing countries alike. Due to global changes such as population growth, land use and economic growth, the challenge of managing climate risk is increasing. The economic impact of climate-related disasters is increasing, and studies on the potential economic impact of climate change[7],[8] expect that trend to continue; while at the same time effectively arguing that proactive climate risk management saves lives and money in the long run. Not surprisingly, over the past few decades there has been a gradual shift in paradigm in the humanitarian community from one focused solely on response to a more balanced approach which includes more cost-effective preparedness and prevention.

Proactive policymaking and planning means taking action now to prevent and/or reduce future anticipated impacts. This is in line with the precautionary principle, which implies that there is a social responsibility to protect the public from exposure to harm when there is a plausible risk (such as in the case of climate change). No regrets strategies are a type of proactive policy making, in that they are strategies taken today aimed at maximizing positive and minimizing negative outcomes for communities and societies in climate-sensitive areas such as agriculture, food security, water resources and health. The 'no regrets' aspect of climate risk management means taking climate-related decisions or actions that make sense in development terms, whether or not a specific climate threat actually materializes in the future.

No regrets strategies are often technologies and practices geared towards improved efficiency; for example, developing markets, integrating seasonal forecasts into resource allocation decisions or repairing infrastructure. They are also the types of tools that reduce exposure to climate vulnerabilities and enable the opportunistic exploitation of favorable climate conditions, such as improved early warning systems and risk-transfer instruments such as index insurance. However, an exclusive focus on no regrets strategies can lead to the false idea that there are no trade-offs and that attention does not need to be paid to the incremental challenges of climate change.

Robust policymaking and planning

In Europe, large glaciers are projected to lose between 30% and 70% of their volume by mid-century (Schneeberger et al., 2003; Paul et al., 2004), which will significantly affect spring and summer water discharge levels. Water users and planners will have to be prepared for changes over the season and years in both the amount and availability of the water resource. Initial increases as the glacier melts, particularly in the spring may lead to increased flooding, and there may be less water available in the long term as the glaciers recede.

In situations where the range of possible climate changes that could occur becomes very broad (or very uncertain), then the decision-making framework needs to be changed, as Matalas and Fiering have suggested [9], so that the robustness of adaptation decisions over a wide range of climates is more important (i.e. has lower economic regrets) than making a decision that is optimal for one or a small number of climate states. Robust policy and planning strategies are designed to perform well across a wider range of plausible future scenarios, requiring flexible planning and management systems.

Information needs for responsive, proactive, robust policymaking/planning

Information needs required to inform decision making for responsive, proactive and robust policymaking and planning are overlapping, particularly as all three rely on the identification of vulnerabilities, hazards and risk as foundations for identifying strategic decisions and alternatives. Though the main objective of these types of policymaking and planning varies, of principle concern is enabling decision makers to make better informed decisions about how to manage resources and risks for improved livelihood and development outcomes.

Though responsive policymaking and planning is not confined to real-time information (it includes some aspects of advance preparedness and planning), rapid assessment of the unfolding situation using information gathered in real time is of essence to timely and effective decision making. This is distinct from the information needs for proactive and robust policymaking and planning which rely on a broader set of information. Robust policymaking and planning could fit under the umbrella of a proactive approach; the distinction here has to do with the strategy chosen after the information is interpreted and analyzed. For example, if the scenarios indicate a wide range of outcomes, a more flexible (robust) response strategy may be necessary.

In an emergency response situation, decision makers must prioritize information needs quickly, initiating a rapid assessment and information gathering– geared towards preventing loss of life, identifying strategic shelter locations and pathways for moving supplies and people. Information needs vary significantly by hazard and location and are too numerous to cite here[10]. Useful hazard information includes identifying the location and extent (what type of hazard and where), intensity and severity characteristics (e.g., flood levels; speed of winds and volume/rate of rainfall during hurricanes; magnitude and intensity of an earthquake), duration (from a few seconds or minutes in the case of flash flood to months or even years in the case of drought), and degree of predictability.

The above information is also useful for preparedness and prevention, including additional information and analysis which identifies past, present and potential hazards and their effects, such as the frequency and probability of occurrence (in both the short and the long term). Information about the speed of onset of a hazard event is principally relevant to disaster preparedness and early warning systems but may also have a bearing on planning decisions (e.g., planning secure evacuation routes). Planners should also be aware of secondary hazards resulting from a hazard event, hazards outside the area of concern that could affect it and how hazard events occur, including not only natural physical processes but also the impact of human activities that create or exacerbate hazards [11].

There are many methodologies available to identify vulnerabilities - primarily focused on gathering information on the political, socioeconomic, ecological, natural resource context of a country. For example, the World Food Programme conducts Comprehensive Food Security and Vulnerability Analysis (CFSVA) which organizes information on food supplies, markets, livelihoods, coping strategies, nutrition, health, and education to identify vulnerabilities. CFSVAs are conducted under normal conditions, rather than during a crisis as WFP hopes to facilitate rapid needs assessments when disaster strikes by identifying the location of the most vulnerable people, as well as the source of their vulnerability.

Climate is a resource which can be used to improve decision making within key development sectors, thereby reducing the impacts of adverse climate, or increasing benefits from favorable climate. A wide array of development decisions, from responsive to robust, can be better informed by climate information and services, including historical, real time and forecasted climate data and information [12].

As a starting point, to characterize the nature of the climate risks at a particular location, we rely on the past. Historical climate observations are central to many development applications, as they provide the basis for planning and optimizing investments (i.e. feasibility, design, location, performance). Analysis of climate observations allows planners to better understand trends, derive climate statistics of interest, and place current observations into historical context. Developing this baseline of climate risks is second nature in developed countries; unfortunately in many developing countries reliable and accessible climate information is in short supply.

Real time climate observations serve as a useful proxy for climate sensitive variables, such as soil moisture and habitat for disease vectors, as well as aid in early warning systems for flood, drought and disease. Operational systems for forecasting these types of impacts derive at least some of their predictability from real time observations. The use of real time data and information to monitor and predict malaria outbreaks is a new innovation that is being introduced in Southern Africa to help manage epidemic malaria.

Where proven to have skill, forecast products at different time scales (weather, medium range, seasonal) may contribute to operational management, hazard management and longer term planning when integrated into appropriate decision making frameworks. For example, at operational scales for water resources management we rely heavily on historic and current hydrologic information (water levels, flows, etc) and weather forecasts, and in some innovative cases, seasonal forecasts.

In particular, the skill of seasonal climate forecasting has been strengthened over the years. Seasonal forecasts have been proven useful for planning agricultural activities, and as a starting point for early warning and response planning. Of course, such forecasts indicate a tilt in the odds towards a particular outcome; they will never give a perfect prediction; one that turns out to be 100% right. The challenge is to incorporate such probabilistic information, with its explicit uncertainties, into decision making.

For longer term planning of infrastructure, policy and investment, we can no longer rely solely on the past observations due to climate change. At the intersection between year-to-year climate variability and climate change lies decadal variability (over one or more decades, usually involving predictions over the next 10-30 years). This timescale has immediate relevance to strategic planning, and is consequently the subject of much ongoing research. Over relatively long timescales (i.e. 30 years or more) and large spatial scales (i.e. hemispheric, global), today's climate change models broadly agree, both with each other and with physical theory, about what is likely to happen in aggregate, at least with regard to anthropogenic climate change.

  • How much/what types of information are sufficient before an adaptation intervention can be advanced?

The uncertainty surrounding future climate can result in confusion, paralysis or the adoption of business as usual approaches. However, managers routinely take into consideration uncertainty associated with relatively high impact socio-economic drivers of change. For example, economic planners factor into national plans assumptions about demographic changes such as population growth and urbanization, despite high levels of uncertainty. Climate uncertainty presents an additional challenge, but need not be a stumbling block if it can be demystified and treated as an additional factor impacting on health, safety and welfare.

In fact, at longer term planning time scales, ignoring climate change can result in inefficient and suboptimal investments and decisions. Planners have historically managed climate risks with differing degrees of success dependent in part upon the quality and scope of the climate information available to them. For example, designing water reservoirs based on short records of river flow have resulted in inefficient constructions and inequitable operating rules when the expected river flow did not occur. In places with good climate records, expectations about how climate is likely to change have been developed by creating simulations of future climate that are consistent with past variations. This information can be helpful to managers and policy makers looking at longer term investments and strategies.

Methodologies have been developed and tested to evaluate the potential economic, environmental, social risks and implications of climate change, as well as the implications on investment alternatives. For example, decision makers can weigh the pros, cons, costs and benefits of different investment strategies in light of climate change predictions (including those 2-3 decades out). One of the biggest 'climate' risks faced by decision makers is that of making mistakes based on false expectations about the climate. These mistakes can result in unreliable, costly, and underperformance of investments (such as water example described above). Economists are developing strategies and tools for weighing the risks, costs, and benefits of different courses of action in order to address and improve development outcomes in the face of future climate change uncertainty [13].

Decision makers may choose to act on uncertain information if they deem the costs and/or consequences of inaction too high. This thinking is also complimented by shifts in adaptation and development thinking towards no regrets strategies. These types of strategies make common sense- that is they are generally strategies which improve efficiencies and make economic sense, but they may not be optimal strategies given a particular outcome.

  • What stakeholders should be involved in determining these information needs? How can decision makers strike a balance between an inclusive, participatory approach to defining information needs which will lend legitimacy to response activities while reacting quickly to the situation at hand?

In situations where response time is of essence, striking a balance between inclusive approaches and quick decision making can be difficult. In 2010, when the earthquake hit Haiti, the Haitian government was effectively paralyzed. UN organizations stepped in to fill this void, essentially taking charge of emergency response activities. As the rainy season approach, there was concern that the situation could be exacerbated as many people living in tents would now be exposed to flooding, disease (such as cholera) and potential landslides. A network of internal and external organization provided information to humanitarians to help them identify risks and areas suitable for providing shelter.

Yet bypassing existing national government risk structure - which seemed to be a necessity in the days after the earthquake - had the unfortunate side effect of further marginalizing the ability of the Haitian's government to take the lead. In the months after the initial shock, UNDP and WMO redoubled efforts to build the capacity of the national meteorological department (CNM) and the ministry of agriculture to provide climate services and manage climate risks.  

More broadly for decision making, participatory approaches engaging vulnerable groups, both poor and non-poor, should be included in information gathering processes to determine their concerns, including perceptions of risk, behavioral response and priorities in strengthening resilience [14]. Other stakeholders with relevant knowledge and expertise should also be consulted, including civil society organizations, civil servants in relevant line ministries and departments (e.g., social welfare, agriculture, transport, health) in national and local government, specialist - related public agencies, the private sector and academic and research institutions.  Trade-offs will inevitably have to be made amongst competing needs and users in this increasingly complex landscape.

  • Do information needs vary depending on whether the risk is a surprise versus a long-term threat? 

Schneider and Turner [15] describe how use of a strict definition of surprise logically entails that we cannot anticipate the event, but that if surprise is defined as a condition in which perceived reality departs qualitatively from assessment of expectations then the events  can be considered knowable in one manner or another. They highlight several ways that surprises can be better anticipated- broadly involving integration and synthesis techniques (searching for connections across problem domains, disciplines, and perspectives), increased focus on a larger fraction of the research effort on "outlier" outcomes and support for more work at the edges (and across edges) of conceptual and problem areas. In addition, research discourse should welcome unconventional views.

It's not so much that information needs vary as they are constrained; thus, the mind set and approach of how risks and vulnerabilities are characterized, derived, anticipated and defined must change to bridge some of the knowledge gaps. For example, contingency and emergency response plans typically have relied on historical climate information to identify potential future hazards (i.e. floods and droughts). However, in the case of the 2010 Pakistani floods, the magnitude of the flood was completely unanticipated, and there was not an existing contingency plan to pull off the shelf, which led to dire consequences. Under the new regime of climate change and climatic uncertainty, it will be necessary to take into consideration climate change projections and anticipate and plan for potential surprises. Even looking further into the past, perhaps at paleontological records, could identify previously unrecorded hazards.

Finally, many hazard or problem arenas are intrinsically subject to surprise and despite anticipatory measures, some surprises will inevitably occur. However, by increasing the resilience and adaptability of at risk communities, the impacts can be lessened.  

  • Do information needs vary if there is less certainty about the risk?

In developing countries, in particular, uncertainty due to historically lower investments in data gives rise to sub-optimal conditions for decision-making. There is often a trade-off between "optimal" and robust solutions, and the appropriate option in a given context is typically determined by the level of risk. Poor characterization of these risks and uncertainties often leads to inefficient solutions [16].

The approaches to gathering information, and the types of information used to make decisions, necessarily vary when there is less certainty about the risk. Numerous approaches to assessing risks have evolved, including the development of quantitative risk analysis methods such as probabilistic risk assessment, and the more recent "democratization" of the risk analysis process that provides stakeholders a greater voice in determining relevant uncertainties and risks. Such an "analytic-deliberative" approach is similar to the methods used to engage stakeholders in integrated water resources management.



[2]UN and World Bank.2010. Natural Hazards, UnNatural Disasters.


[4] Grey, D. and C. Saddoff, 2005. Water for growth and development: a theme document for the 4th World

Water Forum. Washington, World Bank.


[6] Carter, W. N. (1991) Disaster management : a disaster manager's handbook Manila, Philippines : Asian Development Bank


[8] Stern review: The economics of climate change

[9] N.C. Matalas and M.B. Fiering. 1977. “Water-Resource Systems Planning” in National Academy of Sciences (ed). Climate, Climatic Change, and Water Supply.

[10] A good resource is:

[11] Taken from:

[12] Annex 4:

[13] Callaway, J. M., Louw, D.B., Nkomo, J. C., Hellmuth, M.E. and Sparks, D.A. (2006) The Berg River Dynamic Spatial Equilibrium Model: A New Tool for Assessing the Benefits and Costs of Alternatives for Coping with Water Demand Growth, Climate Variability, and Climate Change in the Western Cape. AIACC Working Paper No. 31. International START Secretariat, Washington, DC