An index of ‘deforestation pressure’ is suggested as useful for reserve planning alongside the currently used information on the species present at candidate sites. For any location, the index value is correlated with threats to habitat and thus also survival probabilities over time for members of species dependent on that habitat. Threats in the absence of reserves are key information for planning new reserves. The index is estimated using a regression approach derived from a dynamic, micro-economic model of land use, with data on observed clearing of forest over space and time as well as biophysical and socioeconomic factors in land returns. Applying an estimated threat (or probability of clearing) function for Costa Rica to locations of interest yields relevant estimates of sites’ deforestation pressure, which are used to evaluate proposed reserves and to suggest other candidate sites.
The development of seasonal-to-interannual climate predictions has spurred widespread claims that the dissemination of such forecasts will yield benefits for society. Based on the use as well as non-use of forecasts in the Peruvian fishery during the 1997–98 El Niño event, we identify: (1) potential constraints on the realization of benefits, such as limited access to and understanding of information, and unintended reactions; (2) the need for an appropriately detailed definition of societal benefit, considering whose welfare counts as a benefit among groups such as labor, industry, consumers, citizens of different regions, and future generations. We argue that consideration of who benefits, and an understanding of potential socioeconomic constraints and how they might be addressed, should be brought to bear on forecast dissemination choices. We conclude with examples of relevant dissemination choices made using this process.
Earth-science predictions of natural phenomena are increasingly seen as valuable aids to improved societal decision making. Pielke et al. recently (EOS 7/13/99) argued persuasively that good predictions alone won’t achieve better societal decisions. These authors’ call to change the decision environments in which scientific predictions are used, though, may be more relevant to the daily activities of policy makers than to those of scientists. We see a role also for changing the information that scientists feed into those decision environments. In particular, scientists could better serve societal needs by generating not only possible scenarios, but also improved probabilities that decision makers need, including for decisions to be taken in the near future.
Many firms conduct “environmental audits” to test compliance with a complex array of environmental regulations. Commentators suggest, however, that self-auditing is not as common as it should be, because firms fear that what they find will be used against them. This article analyzes self-auditing as a two-tiered incentive problem involving incentives both to test for and to effect compliance. After demonstrating the inadequacy of conventional remedies, we show that incentives can be properly aligned by conditioning fines on firms’ investigative effort. In practice, however, the regulator may not be able to observe such effort. Accordingly, we propose and evaluate the use of three observable proxies for self-investigation: the manner in which the regulator detected the violation; the firm’s own disclosure of violations; and the firm’s observed corrective actions. Each method has its own efficiency benefits and informational requirements, and each is distinct from EPA’s current audit policy.