ABSTRACT: The sixth mass extinction in the history of the planet is underway. To curb this nonreversible destruction, the wholesale killing of animals and plants needs to stop, and habitat destruction needs to be curtailed. Achieving these two goals will require ecosystem management policies that are effective at stopping the killing and destruction. These policies need to be derived from credible models of those political-ecological systems that host endangered species. Such models need to be fitted to dynamic data streams consisting of both online news articles, and ecological datasets. Because the resources of private enterprise are unmatched, private firms could stem this biodiversity crisis by marketing profitable offerings that are tied to biodiversity projects. Advances in Computer Science would be used to ensure the profitability of these offering and the ecological effectiveness of their associated projects. Advances would be needed in:

1. Computational linguistics methods to detect ecosystem-affecting actions from online news outlets and social media sites;

2. IoT-based persistent remote sensing and reporting of animal and plant populations;

3. Efficient optimization methods for high-dimensional black-box objective functions to support the statistical fitting of political-ecological models;

4. Interpretable graphical displays of nominally-valued political-ecological time series.

5. Non-blockchain protocols for the sharing of criminal intelligence among members of a nonhierarchical, mutually-distrustful group of wildlife crime investigators.

6. Social network analysis of wildlife trafficking networks.

Examples will be given of this Computer Science-enabled business approach to conserving biodiversity.

BIO: Timothy C. Haas earned a Ph.D. in Statistics from Colorado State University in 1989, served as an acting assistant professor in the Statistics department at the University of Washington during 1989-1990, and, apart from sabbaticals at the National Center for Atmospheric Research during 1999 and Stanford’s department of statistics during 2006- 2007, has been at the Lubar School of Business, University of Wisconsin-Milwaukee since 1990. Professor Haas has developed semi-parametric methods for prediction of nonstationary spatio-temporal processes, algorithms for the redesign of monitoring networks, Bayesian network models of aspen stand survival, forestry ranger decision making, and integrated, agent-based models of human-wildlife conflict. Support for these endeavors has come from grants awarded by the United States Department of Agriculture, the United States Environmental Protection Agency, and the World Wildlife Fund. This work has been published in the Journal of the American Statistical Association, Forest Science, Atmospheric Environment, Environmetrics, AI Applications, Stochastic Environmental Research and Risk Assessment, Security Informatics, IEEE Transactions on Cybernetics, Ecological Applications, PLoS One, Frontiers in Conservation Science, the Journal of Cybersecurity, and Cogent Social Sciences. In addition, Professor Haas has published two books with Wiley on ecosystem management. Currently, Professor Haas is developing a consultancy to help firms embark on profitable offerings that conserve biodiversity.

See www.profitablebiodiversity.com, https://sites.uwm.edu/haas/

Hosted by Professor Paul Eggert