Weathering climate change: some simple rules to guide adaptation decisions

Abstract

This paper discusses some of the elements that may characterise an efficient strategy to adapt to a changing climate. Such a strategy will have to reflect the long time horizon of, and the prevailing uncertainties about, climate change. An intuitively appealing approach therefore seems to be to enhance the flexibility and resilience of systems to react to and cope with climate shocks and extremes, as well as to improve information. In addition, in the case of quasi-irreversible investments with a long lifetime (e.g. infrastructure investments, development of coastal zones), precautionary adjustments may be called for to increase the robustness of structures, or to increase the rate of depreciation to allow for earlier replacement. Many of these measures may already have to be considered now, and could be worthwhile in their own right, independent of climate change considerations.

Introduction

It is common to distinguish between two basic responses to climate change: mitigation and adaptation. Trade-offs between the costs of mitigation, the costs of adaptation and the impacts of the enhanced greenhouse effect not covered by either mitigation or adaptation guide the choice between policy strategies for climate change. Both mitigation and adaptation pose significant analytical and policy challenges, yet the respective discussions have evolved at a different pace so far. The study of mitigation measures is well under way and the analysis is continuously refined (cf. Hourcade et al., 1996, Watson et al., 1996). Understanding is likely to increase further now that measures are starting to be implemented.¹ Adaptation options, in contrast, have been the subject of much fewer studies.

Adaptation has mainly been studied in the context of impact analysis, where some understanding of society's adaptive potential is needed to better understand the consequences of unabated climate change (see Smith et al., 1996 and Tol et al., 1998 for surveys). Knowledge about climate change impacts and vulnerability is accumulating, although progress is hampered by the complexity of the problem and the lack of empirical material. Pearce et al. (1996) and Watson et al. (1996) survey the impacts and vulnerability literature which typically assumes little or no adaptation. The adaptation-specific literature has identified an increasing number of adaptation options (see Office of Technology Assessment, 1993 Smit, 1993, also Watson et al., 1996) and methods for their assessment are being developed (e.g. Carter et al., 1994, Smith and Quan Chu, 1994, Fankhauser, 1996, US Country Studies Program, 1996, Mendelsohn and Bennett, 1997, Smith et al., 1997, Yohe and Neumann, 1997). While policy issues have been touched upon by some authors (e.g. Magalheas, 1996, Callaway et al., 1997, Downing et al., 1997, Frederick, 1997, Frederick et al., 1997, Major and Frederick, 1997, Smith, 1997), a thorough discussion of strategic policy issues has been lacking so far, leaving the field to unproven assertions and commonplace statements.

In this paper, we try to substantiate or repudiate some of these statements. Starting from first principles, the paper develops some basic rules of how adaptation could be designed to be efficient—so that the negative effects of climate change are minimised (and positive effects maximised). A wide variety of adaptation options has been put forward, with mostly differing characteristics. The paper starts with a classification of available options, and discusses how the different types of measures interact (Section 2).

One often heard statement is that, with a significant change in climate not expected for at least another two decades, there is no need for immediate adaptation. In Section 3, we discuss this point of view and analyse the optimal timing of adaptation measures. We argue that for long-lived investments, and investments sensitive to rapidly changing climate parameters (primarily extreme weather), climate change needs to be accounted for already in today's investment decisions, especially since weather extremes could be noticed much earlier than changes in mean climate (e.g. Katz and Brown, 1992). Weather-sensitive, long-lived investments particularly comprise infrastructure for water management. Forest plantations, (rail)roads and buildings are also long-lived, and may be vulnerable to changing weather conditions. Arguably, development plans (e.g. for cities), laws and regulations (e.g. for water allocation), and knowlegde bases (e.g. agricultural RD&D), although less tangible, are also long-lived and sensitive to changes in weather regimes.

The paper then goes on to discuss how investment decisions could account for a potential change in climate or weather parameters during the lifetime of a project. We argue that, given the prevailing uncertainty, the best way to account for potential climate change would be to increase the flexibility of systems to function under a wider range of climatic conditions, as well as their robustness to withstand more severe climatic shocks (Section 4). The same basic principles also hold for planning, as we briefly discuss in Section 5.

Another often heard assertion is that adaptation will be largely autonomous and will not require advance strategic policy intervention. We argue that, while individuals can certainly be expected to make adaptations to climate change, it is not certain that they will have the incentive, resources, knowledge and skills to adjust appropriately. Successful adaptation to a large extent depends on three elements: timely recognition of the need to adapt, an incentive to adapt, and ability to adapt. Timely recognition requires access to reliable and detailed information, and the ability to process such information. Section 6 deals with these issues and the role of research and monitoring. Proper incentives and an environment that allows economic agents to adjust are areas in which governments typically have a large stake. Section 7 discusses the role of government in climate change adaptation. It also discusses which tasks need to be carried out by the public sectors, and which can be safely left to individual agents.

The analysis makes clear that, while the impact of climate change on a regional scale remains uncertain, and the increase in mean temperature expected over the next few decades may be relatively modest, it would be shortsighted to postpone adaptive actions until impacts are better understood or more strongly felt. The paper therefore concludes with a series of anticipatory adaptation measures that could be considered now.

The paper is analytical rather than empirical. Worked examples of decision analysis about adaptation can be found in Smith et al. (1998). Historical perspectives on adaptation can be found in Lamb (1982), Langen and Tol (1996) and Wigley et al. (1981). Adaptation studies focusing on the recent past and current situation include Downing et al. (1997), Miller et al. (1997) and Smit et al. (1997). Adaptation to weather variability, particularly weather-related natural disasters, has a long tradition of study (see, for example, Alexander, 1993, Burton et al., 1993).