![Sample our Geography journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5937/TOC-Subject-Sample-2021-Geography.jpg"})
Abstract
This paper presents a cost-benefit analysis (CBA) of a trade-off between salmon and hydropower production in the Ume/Vindel River, northern Sweden. A distinctive element of this analysis is that estimated changes in resource conditions are based on detailed river-specific data. A salmon population model was used to develop the scenario and a novel willingness to pay (WTP) question, which caters for uncertainty in a different manner, provided an interval estimate. Non-use values are the major contributors to the benefit (96–517 millions of Swedish kronor (MSEK)) of increasing the stock of wild salmon. Sensitivity analysis suggests that the opportunity costs in terms of lost electricity are typically higher than the estimated benefits.
Acknowledgements
The author wishes to thank Yuri Belyaev, Bengt Kriström and Per-Olov Johansson for their valuable input to this study. The study was financed by CMF (Centrum för miljövetenskaplig forskning) and the Swedish EPA.
Notes
1. During the period 1995–2005.
2. This is a translated version of Version 1 of the WTP question in the study. See for differences between versions of the WTP question.
3. Northern Sweden is defined here as the counties of Norrbotten, Västerbotten and Västernorrland. Hence, southern Sweden is defined as the other counties in the country. If the population had not been divided into these two stratum before individuals were sampled, it was likely that the number of Swedes from northern Sweden (where the population is much smaller) in the sample would be few in number. If so, it would be hard to investigate distributional effects between different areas. Looking at variables such as age and gender the respondent group for northern and southern Sweden, respectively, was shown to be representative.
4. The respondents were informed that the wild salmon are not considered to be under threat of extinction, but increasing the stock would contribute to their long-term survival.
5. By multiplying the mean WTP for the North and South samples by the numbers of adults in the respective stratum then summing the benefit for the two strata (see Appendix).
6. That is, the extent to which the estimated mean WTP corresponds to the true WTP.
7. This model uses data on turbine flow, flow in the bypass channel, water temperature, sex of salmon, length of salmon and the salmon's moving patterns in the river before reaching the fish ladder.
8. Håkansson et al. (Citation2005) analyse detailed daily data on water flow and daily data on the amount of salmon reaching the fish ladder at Stornorrfors power station. They find that more salmon made it through the fish ladder at a lower flow in the bypass channel than at a higher flow. That is, at status quo, i.e. at the current allocation of water to the salmon.
9. This can be found by combining information provided by the growth function for the salmon (Leonardsson et al. Citation2002b) and the model estimating the proportion of salmon that would succeed with an extra allocation of water (Leonardsson et al. Citation2005).
10. The average number of successful wild salmon migrations during the last 10 years has been approximately 3000 per year, however, the number of succeeding wild salmon during this period fluctuated between 1281 to 6065 per year due to factors such as variations in mortality (Fisktrappa Citation2007). Clearly, the numbers of salmon that reached their spawning grounds following an increase in water flow would not necessarily be approximately 4000.
11. For cost-efficient reasons, no more water than necessary to keep the number of salmon that reach the river at 4000 per year should be allocated to the salmon's passage ways.
12. During 1974–2005.
13. Due to the approximately linear relationship between E and ⋅w (t) Sw(t), it is not necessary to know the average flow during the spawning season. Given the information at Stornorrfors homepage (Vattenfall Citation2007b), it can be shown that it is only necessary to know the extra allocation of water to the salmon's passage ways in order to estimate the loss of electricity production.
14. Data on the flow through the turbines and the flow through the river basin show that this is not unusual in some years (see Leonardsson et al. Citation2005, Figure 4).
15. It could be argued that the market price of electricity for consumers is not necessarily the most appropriate price to use when estimating the value of reductions in electricity production, and that the marginal production cost, i.e. the spot market price for electricity, should be used instead. However, the model considers individuals' value losses, and thus it can be argued that the most appropriate price to use is the price consumers would have to pay.
16. This holds for both consumer prices and prices on the spot-market (see, for example, SCB Citation2007a, Citation2007b, Vattenfall Citation2007a).
17. It is not evident that this range of values is the most correct range to use. It covers future ‘unrealistic’ low electricity prices as well as prices that are considerable higher than prices not yet observed on the market.
18. To facilitate comparisons of different options is it advantageous to use the same discount rate in CBAs. The main roles of the European Commission's Environmental Directorate-General (DG) include the definition and introduction of new environmental legislation and ensuring that agreed measures are implemented in the EU member states. The DG recommends use of a 4% discount rate in CBAs, and the presentation of sensitivity analyses in which both 2% and 6% discount rates have been used. Other rates could be applied but, as shown in , the choice of discount rate does not have a major impact on the estimated total present cost for the salmon-hydropower case.
1. In this case the Swedish population (18 years old or older).
2. In this case to two strata; people in northern and southern Sweden respectively.
3. In this case an increase in the numbers of wild salmon in the Vindel River.
4. Given that Nj is rather large.
