3. Cost-effectiveness

3.1 Review of cost-effectivenes of habitat rehabilitation

Estimates of cost-effectiveness (i.e. increase in habitat area or fish numbers per unit cost) are even less frequently conducted than physical or biological evaluations of rehabilitation projects (Roni, 2005). This is in part because many projects are designed as experiments and accurate costs are not often recorded. Furthermore, many projects include in kind contributions, volunteers and other intangibles such as public education, which are difficult to quantify economically. Few studies we examined reported project costs and only a handful did any cost-benefit or economic analysis (e.g. Cederholm et al.,1997; House et al.,1989). However, the examination of the cost-effectiveness of fisheries habitat rehabilitation is a critical tool for assisting in the prioritization of restoration activities and allocation of resources for habitat rehabilitation and important part of the FAO Code of Conduct for Responsible Fisheries.

The available cost benefit or economic assessments of rehabilitation projects fall into three categories: 1) estimates of the benefit to the economy through estimates of expenditures by recreational fishers, 2) cost-benefit based on increased number of fish per dollar, and 3) a simple reporting of project costs. Below we review published cost-benefit analyses of freshwater rehabilitation projects in these three categories as well as provide estimates of average costs for different types of rehabilitation and discuss information needed for a proper cost-benefit or economic analysis.

Studies on the total benefit to the economy of aquatic habitat rehabilitation are limited. Of all the studies we examined only two examined total benefit to the economy and both of these were on large North American lakes. Hushak et al. (1999) examined two artificial reefs in a large lake and estimated that the total value of the reefs to the economy was US$276 000 per year versus a construction cost of $100 000 (1991 US$). Stockner and MacIsaac (1996) estimate that nutrient enrichment of British Columbia lakes had increased sockeye salmon returns to coastal British Columbia lakes worth approximately $16 million (1996 Canadian dollars) annually to the Canadian economy on an investment of about $17 million between 1976 and 1996. In a modeling effort on British Columbia sockeye salmon lakes, Guthrie and Peterman (1988) examined pulsed versus annual nutrient enrichment and found that the net economic benefit for annual nutrient enrichment was greater than for pulsed. These evaluations and modelling efforts demonstrate that lake enhancement can have large economic benefits.

The value to the economy of fish harvest has also been used but there is considerable debate about the economic value of various species and most work has focused on recreationally important species, most notably Atlantic salmon. The value of a single sport caught Atlantic salmon in 1991 ranged from 250 to 1 000 or more British pounds (Kennedy and Crozier, 1991). Thus if a project estimates the benefit in number of adult Atlantic salmon, a range of economic values can be applied to estimate benefits to the project. Atlantic salmon are unique because their harvest is highly regulated and access to a given section of stream is by fee, making calculations of their economic value more straightforward compared to other species.

Other studies have examined the cost versus the benefit in increased fish numbers and the value of those fish to the commercial fishery or the cost of producing fish through other means. House et al. (1989) examined the economics of 15 stream habitat enhancement projects along the Oregon coast (USA), and estimated that the economic benefits based on increased salmon and trout returns and harvest were more than three times the cost. Estimates such as House et al. (1989) do not contain the true economic value but the value of a landed fish (i.e. the price per kilogram paid to the fisher). O’Grady et al. (1991) estimated that the placement of rubble mats (boulders and cobble to create riffle or habitat diversity) in an Irish stream cost about 14 571 (IR) pounds, but provided an annual increase in juvenile and adult Atlantic salmon and brown trout worth 4 160 (IR) pounds based on the cost to rear and stock hatchery salmon and trout. In a more detailed evaluation of placement of instream structures and brush removal, O’Grady (1995) estimated that costs per returning adult salmon ranged from 0.4 to 8.3 (IR) pounds per additional returning adult Atlantic salmon. Based on their monitoring of nutrient additions in the Keogh River, British Columbia, Canada, Ashley and Slaney (1997) estimated the cost of producing an addition adult steelhead was between $11 and $20 Canadian depending upon the size of the stream.

Comparisons of the cost-effectiveness of different rehabilitation techniques are also rare and typically limited to structural manipulations. Examinations of Johnson and Lynch (1992) found that stake-bed structures used in lake enhancement in the midwestern US were not cost-effective in terms of fish response and angler use when compared with brush bundles or evergreen trees. O’Grady (1995) compared cost-effectiveness (cost per additional returning adult Atlantic salmon) of placement of instream structures, shrub-pruning (debrushing) and stocking of hatchery smolts and found shrub pruning and instream structures more cost-effective than stocking (1.38, 2.14, and 33.34 (IR) pounds, respectively). Cederholm et al. (1997) compared the cost per coho salmon for placing artificial structures versus simply falling trees in the stream and reported costs of US$14.82 versus $13.00 per smolt, respectively. However, he did not incorporate the time value of money over the 25-year period he examined, which would have greatly increased the cost of the artificial structure. This highlights a common problem in simple cost-benefit analysis conducted by fisheries scientists: ignoring basic financial and economic principles when conducting cost-benefit analysis (Plummer, 2005). Habitat protection is believed to be more cost-effective than structural manipulations or other habitat rehabilitation techniques, but similar to other techniques, little information exists on effectiveness of habitat protection measures (Lucchetti et al., 2005).

Cost-benefit analyses can also be done using various "ecosystem services" such as water clarity, fish and wildlife habitat, allowable water uses, and naturalness (Holmes et al., 2004). Using these values Holmes et al. (2004) estimated that riparian rehabilitation projects in the Little Tennessee River basin in North Carolina (USA) had a benefit to cost ratio ranging from approximately 4 to 16, suggesting that these efforts were economically feasible and that this approach might also be used to prioritize projects.

Given that existing economic evaluations of rehabilitation activities make a cost-benefit analysis or comparison of various project types virtually impossible, the next logical step would be to provide cost estimates for various types of rehabilitation. Estimating project cost would seem to be relatively straightforward; however, it is also difficult for most rehabilitation projects, as many have multiple government agencies and other entities that may not track their "in kind" contributions (staff time). Moreover, costs can vary considerably by site or region because of differences in site size, geology, biology, ownership, politics, permitting, and other factors. Thom et al. (1995) examined cost information on more than 90 wetland, instream, and estuarine rehabilitation or projects in the United States and found highly variable rates for basic construction activities (e.g. hauling logs, gravel). For example, gravel removal activity ranged in cost from US$327 to $3 239 dollars per ton and many other rehabilitation activities varied by two or three fold. Even within one type or area of stream rehabilitation costs can be highly variable.

The costs of similar projects per unit area will often vary by stream size. Ashley and Slaney (1997) reported costs of nutrient enrichment of British Columbia streams ranging from $180/km, $346/km, $571/km (Canadian dollars). In a review of Wyoming (USA) instream rehabilitation projects, Binns (1999) reported costs per kilometre in 1995 US dollars of $13 175, $16 780, $ 21 500, $22 000, and $43 000 for first, second, third, fourth, and fifth order streams respectively. Because thousands of dams built in the last hundred years are obsolete and in need of repair or removal in many developed countries and these projects are particularly costly, dam removal has recently received detailed economic analysis. Chisholm (1999) in a review of dam removal projects to date in the United States reported costs of removal ranging from $1500 to $3.2 million. Born et al. (1998) examined small dam removal projects that had occurred in 30 Wisconsin rivers and found that the estimated costs of repairing these dams were on average three times higher than the cost to remove them.

In an effort to provide a broader estimate of what various types of habitat rehabilitation might cost, we summarized the costs of a handful of studies reviewed in other sections of this document that reported costs (Table 15). These averaged from a US$3 500 per kilometre for decommissioning of a forest road to more than US$500 000 per kilometre for remeandering streams. The large variation in costs among and within project types emphasizes how site conditions and other factors can affect project costs. These numbers are probably much higher as many studies did not indicate specifically what year costs were incurred and we did not convert these to 2005 dollars. We also examined cost estimates for different categories of aquatic habitat rehabilitation projects implemented between 2000 and 2003 under the Pacific Coastal Salmon Recovery Fund, a $400 million plus initiative for stream rehabilitation in the northwestern United States and the National River Restoration Synthesis (NRRS) project (www.nrrss.umd.edu); Bernhardt et al., 2005). Mean project costs within a category such as fish passage, ranged from a few thousand to several million dollars and averaged $1.2 million per project between 2001 and 2004 (Table 16). In contrast, costs of rehabilitation measures reported throughout the United States under the NRRS project were much higher. For example, average cost of instream habitat rehabilitation projects US$138 474 versus 365 545 for the Pacific Salmon Coastal Recovery Fund, and the NRRS project respectively. The differences in these two databases (Table 16) and from the literature (Table 15) emphasize differences in costs among and within regions and project types and the difficulty in obtaining accurate costs for a number of projects.

The time value of money (discount rate) and the annual benefits in dollars were examined in only a few studies - emphasizing the need for more detailed economic analysis. Plummer (2005) and Scarfe (1997) provide detailed information on how to conduct cost-benefit analysis. It is important to realize that when conducting cost-benefit analysis the metric chosen will have an effect on the cost-effectiveness of the project. For example, Plummer (2005) found differences in cost-effectiveness of several Pacific salmon habitat enhancement projects if increase in pool habitat or salmon production per unit cost were used.

3.2 Conclusions - cost-effectiveness

In summary, we were not able to compare the cost and benefits of various habitat rehabilitation actions because actual costs are rarely reported for rehabilitation projects. Key items that need to be reported and considered for future projects so that cost benefit analysis will be possible for a single or multiple projects include:

• Full cost of project and year(s) incurred (including value of "in kind" contributions such as unpaid salaries etc.)

• Length or area of project

• Benefits in terms of physical habitat or fish numbers

• Discount rate when estimating multiyear benefits of projects

• Information on the economic value of fish, education, aesthetics, and other values.

TABLE 15
Average and range of costs of various rehabilitation activities reported in studies reviewed in previous sections of this document. We were unable to locate any grazing or riparian studies that provided a cost-benefit analysis. Holmes (1998), Ashley and Slaney (1997), Cederholm et al. (1997), and Lister and Finnegan (1997) are in 1995 US$ all others are as reported in reference. Unlike other techniques, nutrient enrichment needs to be repeated annually.

NA: not available

Some of this information can and should be collected or reported while others will require detailed scientific as well as socioeconomic studies to determine the value of fisheries resources. This information can be used to assist in planning and prioritization of projects. The value of rehabilitated ecosystems to objectives other than fisheries (e.g. erosion control, climate amelioration, carbon sequestering) is extensive (Costanza et al., 1997), but not considered here.