Prioritizing Invasive Species Threats Under Uncertainty

Agricultural and Resource Economics Review, Apr 2006 by Moffitt, L Joe, Osteen, Craig D

In effect, simple decision rules would prioritize pests for assessment of the likelihood and magnitude of threats, development of pest detection and management options, research on the costs and consequences of those options, or analyses to select such options, all of which are costly and subject to limited budgets. For example, the minimax criterion indicates that, if rapid response panels were organized to assess the subjective probabilities of the consequences of pest events and responses, pests with potentially large adverse consequences would have a higher priority for inclusion in the set of pests for assessment, even if the unknown probability of events might be small, than would pests with smaller but potentially more likely consequences. More sophisticated analyses using models that account for bioeconomic systems, time, space, or risk, and that have large data requirements, can be focused on high-priority threats in order to highlight decision trade-offs and help select actions.

Research on high-priority threats would also provide information to (i) re-estimate pest damage and reprioritize pest threats, which could identify other needs for research, (ii) examine the costs, effectiveness, and potential adverse effects of responses, and (iii) conduct more sophisticated analyses to select options. In the third case, when probabilities are appropriate for representing uncertainty of events and good estimates become available through the learning process, the traditional risk management approach, as expressed by equations (1) or (2), could be used to evaluate decisions. If probabilities are not available or appropriate for representing uncertainty but other relevant information is available, then the minimax criterion in expression (3) or other uncertainty criteria could be used to select actions.

Crop Protection Priorities

We developed a ranking of potential agricultural crop pests not currently or previously in the United States to demonstrate use of the combined minimax and relative cost criteria, as expressed in (4). A pessimistic presumption consistent with this approach is that significant pest problems in foreign locales will eventually lead to a similar significant domestic pest experience. While this presumption might not always be true, some empirical evidence supports using it. Reichard and Hamilton (1997) found that the most reliable characteristic for predicting invasiveness of woody plants was whether or not the species was known to invade elsewhere in the world. We assumed that no cost-effective options are available, so pests were ranked by worst-case damage estimates. A list developed with these procedures could be regarded as a first step in the decision making process, because it prioritizes pests for actions that obtain information about damage severity or the availability, cost, and effectiveness of exclusion or management actions in order to inform decisions that select such actions.

The procedure identified arthropod, weed, and disease pests that have never been in the United States but that could affect any of the 25 highestvalue U.S. crops. Focusing the analysis on highvalue crops assumes that prevention or management costs will be low relative to damage costs, which provides a rationale for preparedness actions. The procedure derived the 25 highest-value U.S. crops for a representative year, identified foreign countries where the crops are also produced, identified pests affecting the crops in the foreign countries, and recorded reported estimates of pest damage to the crops.