Adaptation to climate change in water, sanitation and hygiene: assessing risks, appraising options in Africa

LEAD ORGANIZATION

ODI – Overseas Development Institute, UK

ORGANIZATIONS INVOLVED

OPML; Richard Carter & Associates

The risks to delivery of WASH results posed by climate change in Africa, drawing on three country case studies and respective DFID portfolio assessments

This report presents the findings of research into the risks to delivery of WASH results posed by climate change in Africa, drawing on rapid case study reviews of WASH programming in Malawi, Sierra Leone and Tanzania. A separate Case Study Report provides further detail on country background and findings.

Water is predicted to be the main channel through which the impacts of climate change will be felt by people, ecosystems and economies. However, predicting impacts on the availability and quality of freshwater resources, and more so water-dependent services and sanitation, remains difficult. While there is a high level of confidence in the processes linking emissions to warming, much less is known about how warming will manifest itself at the local level through changes in rainfall, runoff, groundwater recharge and climate extremes. This reflects challenges with the downscaling of climate models, but also the significance of intervening factors such as changes in land cover which may have a greater influence on local systems and services than climate change. In general, the level of confidence in climate change projections decreases as their potential utility for making decisions on how to adapt increases (OECD, 2013).

At the same time, the rapid increase in awareness and concern about climate change, and the need to identify concrete adaptation responses, risks driving demand for certainty beyond what the science community can realistically achieve (Conway, 2011). What is clear, however, is that existing levels of climate variability, together with other pressures on resources and services, already cause major problems. These will undoubtedly get worse as climate change intensifies and other pressures increase, with the result that hard-won public health and poverty alleviation gains could be lost (Howard et al, 2010; Calow et al, 2011).

Against this background, Section 1 of this report begins by looking at the WASH landscape, the gains made, but also some of the bottlenecks that hold back progress. In view of their importance and the sums of money involved, it is remarkable how little is known about the performance of services, why they so often fail, and the resources they depend on. For example, donor-supported programmes across Sub-Saharan Africa (SSA) prioritise both service extension and rehabilitation, but do so with little if any evidence on performance or the causes of failure.

Given the uncertainties with rainfall projections, but also the known risks associated with existing variability, there are strong arguments for a vulnerability rather than an impact-led approach to risk assessment and planning. This implies a stronger focus on ensuring the reliability and protection of drinking water sources and simple changes to latrine design to reduce the risks of flooding under current climate variability as a first step towards adaptation through relatively low cost changes in design or practice. Where long term investment decisions are involved, e.g. with dams, treatment works and piped networks, a greater range of variability should be considered to avoid costly mistakes. In terms of economic appraisal, longer time frames and the need to balance current costs with more distant benefits raise questions around the appropriate discount rates to apply.

How can this be achieved, and what needs to change? While there is a proliferation of toolkits and guidance on climate change adaptation, there is remarkably little on WASH, or on the practical substance of adaptation more widely (Fankhauser & Burton, 2011). Moreover, what there is has focused almost exclusively on system vulnerability and technical change. The technical emphasis mirrors a wider trend in adaptation policy: specifically the preference for concrete and more readily identifiable (and measurable) things, and the reduction of adaptation policy to lists of analytical, planning and delivery processes that need technical know-how to make them work (Lockwood, 2013). As Section 1 makes clear, however, it is far from clear that a lack of technical capacity (or the toolkits that support it) are the most pressing constraints. Rather, we see problems with the ability to deliver basic public services, resilient or not. Hence the argument, summarised in Section 1, to reframe the problem: from climate change and WASH to WASH governance in an era of climate change.

Having looked at the context for adaptation decision-making in WASH, Section 2 of the report looks at the risks to delivery of WASH results in three countries: Malawi, Sierra Leone and Tanzania. In each country, the aim was to canvass opinion on a risk screening approach that could be applied by programme designers and implementers to identify and mitigate risks. In light of the discussion above the approach considered climate change in context – as one of a number of threats affecting WASH outcomes - and focused also on WASH institutions and governance. The approach is relatively simple and straightforward to apply and is based on a two-step process. Step 1 is a national level assessment of risks to WASH that can be set out as a ‘traffic light’ scorecard, using documented indicators of vulnerability and expert judgment. Step 2 attempts to determine the extent to which a WASH programme addresses key risks and vulnerabilities, again using a simple scoring system, documented evidence and expert judgment. This helps identify adaptation options or measures. Guidance on the approach and how to apply it is presented in Appendix B. While the scorecard approach is not intended for country comparisons, and the focus of the country visits was on the screening approach rather than results and option identification, a number of common issues emerge from the country consultations.

Firstly, while there is a general awareness of climate risk in all three countries, there are opportunities to translate this into practical measures that could increase the resilience of WASH programmes. In terms of longer term climate change, uncertainties in climate projections present perhaps an entirely rational barrier to prioritisation, at least for simple systems with a design life of 10-20 years (Conway, 2011). However, this is more difficult to justify given the widely perceived impact current climate variability already has on WASH results. While the causes of service failure or under-performance can be difficult to unravel, floods undoubtedly cause sanitation systems to overflow, result in damage to infrastructure and create widespread health problems. Existing seasonality of rainfall affects the performance of springs and shallow wells tapping smaller groundwater systems with low storage, leading to water rationing and use of unsafe sources. And environmental degradation exacerbated by intense rainfall events clearly impacts on infrastructure and poses a longer term threat to the resource base. In all three countries, simple steps could be taken to mitigate some of these risks, identified in the country workshops and Step 2 of the screening process. Addressing issues such as catchment protection, water resources management and the lack of basic knowledge on resource conditions and trends will take longer, but is essential as climate change accelerates and competition for water grows (Howard et al, 2010; Calow et al, 2011).

Secondly, the risk screening process suggests that existing political and institutional bottlenecks act as a serious brake on service delivery and sustainability. An effective central state remains important for adaptation because of its direct role in allocating resources and setting incentives. Effective local government is necessary to deliver services, or oversee their provision by others. Yet capacity constraints continue to block pathways to better WASH outcomes. These include the ability to supervise construction and enforce standards, and the ability (and incentive) to build and use a knowledge base on local climate, water resource conditions and pressures, and environmental conditions more generally. These bottlenecks are being addressed in partnership with government in each of the country programmes. Overcoming them is central to the delivery of more resilient water and sanitation services – arguably more so than technical change alone.

Section 3 of the report looks at the use of cost-benefit analysis (CBA) as a means of appraising the adaptation options prioritised through the risk screening process. The value of CBA lies in its ability to narrow the scope for ‘pure judgement’, providing a more secure and transparent basis for investment decision-making. However, robust CBA requires reasonable data on what would happen to WASH interventions and outcomes ‘with’ and ‘without’ adaptation. Since there are few hard data linking climate to WASH outcomes, and the current study was based on very limited time in-country, the examples provided in this report are indicative. The main aim is to show how CBA could be used as an appraisal tool, alongside risk screening, to identify a broad set of adaptation options and then go about prioritising them.

The approach outlined in Section 3 focuses on potentially low regret interventions that could be expected to offer significant economic benefits under a range of different climate futures (IPPC, 2012). These include changes to water point construction, and simple changes to the design of sanitation systems. These are termed Best Practice (options) under Existing and Increasing Climate Variability (BPEICV), and are compared with Business as Usual (BAU) baselines in which low regrets measures are not implemented. This provides us with the ‘with’ vs. ‘without’ comparison needed to identify differences in costs and benefits. Also considered are a number of softer adaptation options, including better catchment management, flood risk mapping and hydro-meteorological data collection, where benefit estimation and attribution become more difficult.

The low regrets measures tend to increase net benefits relative to the BAU case, with benefits estimated using Disability Adjusted Life Years (DALYs) averted and projected time savings. Global studies have shown that these comprise the largest proportion of benefits of WASH interventions. The simplest CBA applications described in Section 3 take the form of a ‘discrete option analyses’ in which the impact of a single type of intervention within a WASH programme is examined. In the Tanzania case, the best practice intervention involves improved, drought-resistant construction of boreholes for rural water supply. This incurs additional upfront costs in the form of deeper drilling and the supervision of contractors, but results in an uninterrupted stream of benefits over a 10 year period, compared with a baseline in which drought results in the failure of the source, loss of benefits and extra rehabilitation costs. In this illustration, the benefit-cost ratio (BCR) rises from roughly 1.4 to 1.7 with the intervention – a modest increase. Discrete option analysis is also used to show how the performance of alternative rural sanitation designs can be evaluated. In Malawi, raising and lining rural latrines in a (hypothetical) flood-prone area incurs additional costs but is assumed to prevent most flood-induced collapses. The result is an increase in the BCR from 2.1 to 2.9. In each case, assumptions (e.g. around flood frequency and severity) can be changed to see how sensitive the results are to key variables.

More challenging programme-level CBAs are illustrated for Malawi and Sierra Leone that include several (linked) interventions. In the Sierra Leone case, for example, CBA is applied to flood risk mapping and improved (flood resistant) latrine design in the informal settlements of Freetown, assuming major floods occur every five years. In the Malawi example, the most detailed of the three country illustrations, data on the functionality of rural water supplies over time and likely causes of failure are used to identify benefits that could be attributed to investments in (climate-sensitive) water resources assessment, monitoring and catchment protection that might increase the functionality of water points. In both cases, BCRs increase with the interventions compared with the no intervention baseline, in the Malawi case from around 2.8 to 3.1 over a 20 year period, and in Sierra Leone from 2.3 to 2.7. Again, key variables such as flood and drought frequency, the costs of infrastructure damage and the benefits arising from uninterrupted access to services can be changed as better data become available. Although these worked examples are tentative, the findings suggest that adaptation pays: across all three countries and in each CBA illustration, investment in additional adaptation (the difference between best practice and business as usual) incurs extra upfront costs, but leads to greater future benefits over modest (10-20 year) time horizons.

In terms of next steps, it is important to highlight the limitations of the research presented and the gaps that need to be filled. Firstly, the risk screening approach outlined in Section 2 attempts to be both simple enough to be applied quickly, and detailed enough to provide useful insight into programme design. This is a difficult balance, and the end result may be too generic for some. An obvious way forward would be to provide a programme breakdown between rural water supply, rural sanitation, urban water supply and urban sanitation along the lines of WSP’s Country Status Overview (WSP, 2011). This could replace or supplement Step 2 – the programme level assessment.

Secondly, the economic analysis could be strengthened and extended in a number of ways, since this report provides only a preliminary set of examples. Sticking to relatively simple options such as water point construction and latrine design, the aim would be to flesh out some damage/failure functions based on previous climate events, assembling available data and case histories on what happened to WASH infrastructure and services during and after flood or drought episodes (for example). Using a stronger mix of primary and secondary data than we were able to gather here, future investment scenarios and sensitivity tests could then be prepared, using different frequencies of extremes, or increasing the damage function to represent more intense extremes in the future. This could also inform decisions around whether to invest heavily in upfront climate proofing, or accept damage and cost ‘spikes’ associated with the periodic repair or rehabilitation of less robust, lower cost infrastructure.

End-to-end assessment of national-level WASH policies and major donor portfolios in three countries, using many of the AGWA principles as they existed in 2013.

Finally, Section 4 of the report summarises the main conclusions of the project and identifies some next steps in terms of the application of research findings and further research needs. A main conclusion is that clear opportunities exist to increase the resilience of WASH, and that adaptation should start with the measures that tackle existing climate risks that countries such as Malawi, Tanzania and Sierra Leone already face. A key argument is that many of these measures, such as improved siting, design and construction of water points, or changes in latrine design, are relatively simple, if capacity exists to implement. The preliminary CBA conducted on this project suggests that such measures are likely to bring positive returns, even over short time periods. Perhaps more importantly, the report illustrates how CBA can be used to compare the costs and benefits of different adaptation measures, and how sensitivity analysis can be applied to see how results change under different scenarios and assumptions. From a policy angle, the distinction drawn between ‘Business as Usual’ programme design and ‘Best Practice under Existing and Future Climate Variability’ may also be helpful in defining - and putting a monetary value on – additionality. That is, the demand from developing countries for adaptation to be supported over and above Official Development Assistance (ODA), and the need for developed countries to ensure that any additional finance is used to reduce vulnerability to climate change specifically.

PROJECT WEBSITE OR RESOURCES

This was a study carried out by a team at ODI for DFID in 2013-2014, and is available online: http://www.odi.org/publications/8154-climate-change-adaptation-wash-water-sanitation

KEY WORDS

Water & sanitation, Africa, evaluations and assessments, adaptation, Malawi, Sierra Leone, Tanzania, risk screening, cost-benefit analysis

COUNTRIES OR REGIONS INVOLVED

Sierra Leone, Tanzania, Malawi

KEY STAKEHOLDERS

DFID