17.2.2 Examples of adaptation practices

There is a long record of practices to adapt to the impacts of weather as well as natural climate variability on seasonal to interannual time-scales – particularly to the El Niño-Southern Oscillation (ENSO). These include proactive measures such as crop and livelihood diversification, seasonal climate forecasting, community-based disaster risk reduction, famine early warning systems, insurance, water storage, supplementary irrigation and so on. They also include reactive or ex-poste adaptations, for example, emergency response, disaster recovery, and migration (Sperling and Szekely, 2005). Recent reviews indicate that a ‘wait and see’ or reactive approach is often inefficient and could be particularly unsuccessful in addressing irreversible damages, such as species extinction or unrecoverable ecosystem damages, that may result from climate change (Smith, 1997; Easterling et al., 2004).

Proactive practices to adapt to climate variability have advanced significantly in recent decades with the development of operational capability to forecast several months in advance the onset of El Niño and La Niña events related to ENSO (Cane et al., 1986), as well as improvements in climate monitoring and remote sensing to provide better early warnings on complex climate-related hazards (Dilley, 2000). Since the mid 1990s, a number of mechanisms have also been established to facilitate proactive adaptation to seasonal to interannual climate variability. These include institutions that generate and disseminate regular seasonal climate forecasts (NOAA, 1999), and the regular regional and national forums and implementation projects worldwide to engage with local and national decision makers to design and implement anticipatory adaptation measures in agriculture, water resource management, food security, and a number of other sectors (Basher et al., 2000; Broad and Agrawala, 2000; Meinke et al., 2001; Patt and Gwata, 2002; De Mello Lemos, 2003; O’Brien and Vogel, 2003; Ziervogel, 2004). An evaluation of the responses to the 1997-98 El Niño across 16 developing countries in Asia, Asia-Pacific, Africa, and Latin America highlighted a number of barriers to effective adaptation, including: spatial and temporal uncertainties associated with forecasts of regional climate, low level of awareness among decision makers of the local and regional impacts of El Niño, limited national capacities in climate monitoring and forecasting, and lack of co-ordination in the formulation of responses (Glantz, 2001). Recent research also highlights that technological solutions such as seasonal forecasting are not sufficient to address the underlying social drivers of vulnerabilities to climate (Agrawala and Broad, 2002). Furthermore, social inequities in access to climate information and the lack of resources to respond can severely constrain anticipatory adaptation (Pfaff et al., 1999).

Table 17.1 provides an illustrative list of various types of adaptations that have been implemented by a range of actors including individuals, communities, governments and the private sector. Such measures involve a mix of institutional and behavioural responses, the use of technologies, and the design of climate resilient infrastructure. They are typically undertaken in response to multiple risks, and often as part of existing processes or programmes, such as livelihood enhancement, water resource management, and drought relief.

A growing number of measures are now also being put in place to adapt to the impacts of observed medium- to long-term trends in climate, as well as to scenarios of climate change. In particular, numerous measures have been put in place in the winter tourism sector in Alpine regions of many Organisation for Economic Co-operation and Development (OECD) countries to respond to observed impacts such as reduced snow cover and glacier retreat. These measures include technologies such as artificial snow-making and associated structures such as high altitude water reservoirs, economic and regional diversification, and the use of market-based instruments such as weather derivatives and insurance (e.g., Konig, 1999, for Australia; Burki et al., 2005, for Switzerland; Harrison et al., 2005, for Scotland; Scott et al., 2005, for North America). Adaptation measures are also being put in place in developing country contexts to respond to glacier retreat and associated risks, such as the expansion of glacial lakes, which pose serious risks to livelihoods and infrastructure. The Tsho Rolpa risk-reduction project in Nepal is an example of adaptation measures being implemented to address the creeping threat of glacial lake outburst flooding as a result of rising temperatures (see Box 17.1).

Box 17.1. Tsho Rolpa Risk Reduction Project in Nepal as observed anticipatory adaptation

The Tsho Rolpa is a glacial lake located at an altitude of about 4,580 m in Nepal. Glacier retreat and ice melt as a result of warmer temperature increased the size of the Tsho Rolpa from 0.23 km2 in 1957/58 to 1.65 km2 in 1997 (Figure 17.1). The 90-100 million m3 of water, which the lake contained by this time, were only held by a moraine dam – a hazard that called for urgent action to reduce the risk of a catastrophic glacial lake outburst flood (GLOF).

Figure 17.1. Tsho Rolpa Risk Reduction Project in Nepal as observed anticipatory adaptation.

If the dam were breached, one third or more of the water could flood downstream. Among other considerations, this posed a major risk to the Khimti hydropower plant, which was under construction downstream. These concerns spurred the Government of Nepal, with the support of international donors, to initiate a project in 1998 to lower the level of the lake through drainage. An expert group recommended that, to reduce the risk of a GLOF, the lake should be lowered three metres by cutting a channel in the moraine. A gate was constructed to allow for controlled release of water. Meanwhile, an early warning system was established in 19 villages downstream in case a Tsho Rolpa GLOF should occur despite these efforts. Local villagers were actively involved in the design of the system, and drills are carried out periodically. In 2002, the four-year construction project was completed at a cost of US$3.2 million. Clearly, reducing GLOF risks involves substantial costs and is time-consuming as complete prevention of a GLOF would require further drainage to lower the lake level.

Sources: Mool et al. (2001); OECD (2003b); Shrestha and Shrestha (2004).

Recent observed weather extremes, particularly heatwaves (e.g., 1995 heatwave in Chicago; the 1998 heatwave in Toronto; and the 2003 heatwave in Europe), have also provided the trigger for the design of hot-weather alert plans. While such measures have been initiated primarily in response to current weather extremes, at times there is implicit or explicit recognition that hot weather events might become more frequent or worsen under climate change and that present adaptations have often been inadequate and created new vulnerabilities (Poumadère et al., 2005). Public health adaptation measures have now been put in place that combine weather monitoring, early warning, and response measures in a number of places including metropolitan Toronto (Smoyer-Tomic and Rainham, 2001; Ligeti, 2004; Mehdi, 2006), Shanghai (Sheridan and Kalkstein, 2004) and several cities in Italy and France (ONERC, 2005). Weather and climate extremes have also led to a number of adaptation responses in the financial sector (see Box 17.2).

There are now also examples of adaptation measures being put in place that take into account scenarios of future climate change and associated impacts. This is particularly the case for long-lived infrastructure which may be exposed to climate change impacts over its lifespan or, in cases, where business-as-usual activities would irreversibly constrain future adaptation to the impacts of climate change. Early examples where climate change scenarios have already been incorporated in infrastructure design include the Confederation Bridge in Canada and the Deer Island sewage treatment plant in Boston harbour in the United States. The Confederation Bridge is a 13 km bridge between Prince Edward Island and the mainland. The bridge provides a navigation channel for ocean-going vessels with vertical clearance of about 50 m (McKenzie and Parlee, 2003). Sea-level rise was recognised as a principal concern during the design process and the bridge was built one metre higher than currently required to accommodate sea-level rise over its hundred-year lifespan (Lee, 2000). In the case of the Deer Island sewage facility, the design called for raw sewage collected from communities onshore to be pumped under Boston harbour and then up to the treatment plant on Deer Island. After waste treatment, the effluent would be discharged into the harbour through a downhill pipe. Design engineers were concerned that sea-level rise would necessitate the construction of a protective wall around the plant, which would then require installation of expensive pumping equipment to transport the effluent over the wall (Easterling et al., 2004). To avoid such a future cost the designers decided to keep the treatment plant at a higher elevation, and the facility was completed in 1998. Other examples where ongoing planning is considering scenarios of climate change in project design are the Konkan Railway in western India (Shukla et al., 2004); a coastal highway in Micronesia (ADB, 2005); the Copenhagen Metro in Denmark (Fenger, 2000); and the Thames Barrier in the United Kingdom (Dawson et al., 2005; Hall et al., 2006).

A majority of examples of infrastructure-related adaptation measures relate primarily to the implications of sea-level rise. In this context, the Qinghai-Tibet Railway is an exception. The railway crosses the Tibetan Plateau with about a thousand kilometres of the railway at least 13,000 feet (4,000 m) above sea level. Five hundred kilometres of the railway rests on permafrost, with roughly half of it ‘high temperature permafrost’ which is only 1 to 2°C below freezing. The railway line would affect the permafrost layer, which will also be impacted by thawing as a result of rising temperatures, thus in turn affecting the stability of the railway line. To reduce these risks, design engineers have put in place a combination of insulation and cooling systems to minimise the amount of heat absorbed by the permafrost (Brown, 2005).

In addition to specific infrastructure projects, there are now also examples where climate change scenarios are being considered in more comprehensive risk management policies and plans. Efforts are underway to integrate adaptation to current and future climate within the Environmental Impact Assessment (EIA) procedures of several countries in the Caribbean (Vergara, 2006), as well as Canada (Lee, 2000). A number of other policy initiatives have also been put in place within OECD countries that take future climate change (particularly sea-level rise) into account (Moser, 2005; Gagnon-Lebrun and Agrawala, 2006). In the Netherlands, for example, the Technical Advisory Committee on Water Defence recommended the design of new engineering works with a long lifetime, such as storm surge barriers and dams, to take a 50 cm sea-level rise into account (Government of the Netherlands, 1997). Climate change is explicitly taken into consideration in the National Water Management Plan (NWMP) of Bangladesh, which was set up to guide the implementation of the National Water Policy. It recognises climate change as a determining factor for future water supply and demand, as well as coastal erosion due to sea-level rise and increased tidal range (OECD, 2003a).

There are now also examples of consideration of climate change as part of comprehensive risk management strategies at the city, regional and national level. France, Finland and the United Kingdom have developed national strategies and frameworks to adapt to climate change (MMM, 2005; ONERC, 2005; DEFRA, 2006). At the city level, meanwhile, climate change scenarios are being considered by New York City as part of the review of its water supply system. Changes in temperature and precipitation, sea-level rise, and extreme events have been identified as important parameters for water supply impacts and adaptation in the New York region (Rosenzweig and Solecki, 2001). A nine-step adaptation assessment procedure has now been developed (Rosenzweig et al., 2007). A key feature of these procedures is explicit consideration of several climate variables, uncertainties associated with climate change projections, and time horizons for different adaptation responses. Adaptations can be divided into managerial, infrastructure, and policy categories and assessed in terms of time frame (immediate, interim, long-term) and in terms of the capital cycle for different types of infrastructure. As an example of adaptation measures that have been examined, a managerial adaptation that can be implemented quickly is a tightening of water regulations in the event of more frequent droughts. Also under examination are longer-term infrastructure adaptations such as the construction of flood-walls around low-lying wastewater treatment plants to protect against sea-level rise and higher storm surges.