Seeing the water for the trees

Biodiversity Climate Forests Investments Water Jan 1, 2001
Mark Nicholls

Everyone knows that planting trees is good for water management, right? Wrong. A new report has challenged the assumption that "more trees are always better"–and has raised questions about the science underpinning water markets. Ecosystem Marketplace talks to the report's authors. For advocates of ecosystem service markets, water represents–after carbon–the most promising candidate for payment schemes to reward environmental stewards (see also, Watershed Services: The New Carbon? and Watershed markets get a dose of myth-busting science). Like carbon, measurement–in this case, for water quantity and quality–is relatively straightforward. Like carbon, there is 'demand'– from users who are prepared to pay for reliable and clean water supplies. And, like carbon, it is thought, there is a clear relationship between the behaviour of stewards–such as farmers and other landowners–and the quality and quantity of water flows. But the results of a four-year study by the UK's Forestry Research Programme (FRP)–funded by the Department for International Development (DFID) (see FRP News)–have raised questions over the basis for existing and potential payment schemes for water management. Writing in From the Mountain to the Tap, the report setting out the study findings, the manager of the DFID FRP, John Palmer, says "the research findings will be a disappointment to some enthusiastic promoters of schemes for payments for environmental services, which are one of the most promising avenues for putting reliable amounts of cash into the hands of upstream managers." The study also urges a rethink in the assumptions underpinning "major investments" in forestry-based water resource projects, stating that "trees [are] overplayed as solutions to the world's water problems". The report summarises the findings of a series of research projects carried out in Costa Rica, India, South Africa and Tanzania by water experts based at the University of Newcastle Upon Tyne in the UK, and at the Free University of Amsterdam. The main finding of the report is that "massive amounts of money" are being wasted by government agencies and development banks pursuing tree planting initiatives under the assumption that "forested land always conserves and supplies more water than grasslands or other treeless areas". Using more advanced instrumentation, better mathematical modelling and more powerful geographical information systems, scientists have been able to plug gaps in their understanding of what happens to watersheds between 'water in'–via precipitation–and 'water out'–that is, as streamflow. For example, one study in India found that tree-planting was reducing downstream water flows, suggesting that converting agricultural lands to forests in Himachal Pradesh and Madyha Pradesh caused a 16-26% reduction in water yields. This is partly explained by the higher rates of evaporation–transpiration–of water from forests compared to grassland. "We're not saying that forests never produce water benefits or that they don't have an important role in the ecosystem," says Ian Calder at the University of Newcastle upon Tyne. "But if we are trying to manage our water resources effectively, the overenthusiastic adoption of the simple view that 'more trees are always better' is a prime example of how a failure to root a decision in scientific evidence leads to bad water policy." "The key message," says Palmer, "is there are no blanket recommendations." The study's authors have particular warnings about market-based water payment schemes, based particularly on a study in Costa Rica, which has operated an extensive environmental services programme since 1996. In that year, the government established a national forestry finance fund (FONAFIFO) (see Overview: Costa Rica Water-Based Ecosystem Services Markets) and a payment scheme to pay upland farmers to sustainably manage their land, and to fund reforestation. The programme recognises four environmental services: carbon sequestration; hydrological services (stablising streamflow and reducing sedimentation downstream); biodiversity protection; and the provision of scenic beauty. FONAFIFO acts as an intermediary between buyers of carbon credits and participants in the scheme, and it disburses the proceeds from a domestic fuel tax to landowners. It also distributes funds to farmers from private hydroelectric companies–who are particularly concerned about problems caused by sedimentation. In Costa Rica, the report notes, part of the rationale for providing payments for water services has been that 'cloud forests' capture water from the atmosphere, stabilising–or even increasing–water flows downstream. However, "only scanty scientific data or analysis existed [in 1996] to prove that these beliefs were true." So, the DFID FRP funded a team of hydrologists, lead by Sampurno Bruijnzeel of the Free University of Amsterdam, to carry out a study of the relationship between cloud forests and water flow. Improved data collection–such as the use of horizontal rain gauges–helped to plug gaps in the scientific knowledge, such as why stream flows from cloud forest was higher than might be expected (it turns out that traditional rain gauges failed to capture horizontal wind-driven rain). The data collected was fed into a model–called 'FIESTA'–developed by Mark Mulligan of Kings College London. This national-scale model, which focused on water production rather than stream flow, predicts the amount of water produced by forests and other types of land. However, the report notes that water can either leave upland areas through run-off, be stored in the soil, or percolate downwards–and, "in highly fractured bedrocks…deeply penetrating water, or 'leakage', can cross watersheds underground. This multitude of possible routes presents great challenges in tracing water flow back to a specific location downstream." The fundamental problem is that it is often not possible to directly tie improved water flows to the particular landowner who is being paid to manage his land, says Palmer: "Unfortunately, it seems just now that models for reliable prediction of dry season streamflows in relation to the management of catchment vegetation upstream will not routinely support payments for water services." This conclusion was supported by another piece of research carried out by the FRP. Ina Porras, of the International Institute for Environment and Development (IIED), looked at the Arenal reservoir, which feeds hydroelectric generation, in Costa Rica's Rio Chiquito catchment. Porras found that "regardless of the conversion of cloud forests, the Arenal reservoir tended to maintain a stable flow throughout the year, and from year to year." "What we're really concerned about," says Palmer, "is, for example, a brewery putting a pipeline into a forest, and paying the farmer to keep the water clean. The leakages are a concern–they're very difficult to measure." Indeed, the report notes that water payment schemes could be open to "complex litigation" if, for example, payments were made and water quality failed to improve. "Local biophysical relationships are too complex to be translated into direct economic trading relationships and, because of the difficulty in providing absolute proof, could be challenged legally," the report says. Palmer does not, however, write off the concept altogether–or rubbish the efforts made thus far in Costa Rica. The FRP team has been in discussions with officials in Costa Rica, "who are very keen to get it right", he says. The answer to the leakage question, Palmer says, may be that schemes should be increased in scope, to a regional or even national level. This, he concedes, could lead to a prisoner's dilemma: "the critical thing would be to persuade enough land-owners and managers to farm in a way that no-one ruins it." "Regional schemes may be workable if reliable minimum estimates, rather than statistical mean values, of downstream water flow are taken as the basis for payments, and long-term trading arrangements are guaranteed," the report says. Also, water payment schemes would require more careful monitoring of water flows than previously anticipated, Palmer says, which in turn would make them more expensive to operate. "Such schemes are practical," he says. But "this is pointing in the direction of 'clean water is a good thing, we'll pay for it out of a general tax'. Payments can't be as precisely targeted as we thought." "A nationally collected payment, such as a central government tax on domestic water supplies, which recognises the national benefit of a positive hydrological outcome in a specific region, could be a more workable, secure and equitable financing mechanism for payments to watershed service providers," the report argues. Indeed, it also raises concerns that any market mechanisms for addressing water are adequately "pro-poor". While most of these concerns are directed at South Africa's Working for Water programme, the reports section on the Costa Rica environmental payment programme notes that many smaller landowners are unable to gain access to the scheme. While the scheme is open to parcels of land between one and 300 hectares, FONAFIFO adopts a qualification threshold of 10 hectares, to be consistent with the minimum area of "forest" defined by the Costa Rican government. Furthermore, farmers must be able to demonstrate government-authorised land-title–which also tends to exclude small, traditional farmers. "Another problem is that small farmers often do not have the education or resources to deal with an application to the scheme," the report adds. However, while the research carried out by the FRP programme has raised a number of questions about forests, water flows, and environmental markets, Palmer stresses that there is much more still to learn. "We definitely need more research to understand the mosaic of environmental services that forests provide," he concludes. Mark Nicholls, a regular contributor to The Ecosystem Marketplace, is the London-based editor of Environmental Finance magazine, and consulting editor to its sister publication, Carbon Finance. First published: August 10, 2005