Taking It for a Test Drive: A Hybrid Spatio-Temporal Model for Wildlife Poaching Prediction Evaluated Through a Controlled Field Test
Worldwide, conservation agencies employ rangers to protect conservation areas from poachers. However, agencies lack the manpower to have rangers effectively patrol these vast areas frequently. While past work has modeled poachers’ behavior so as to aid rangers in planning future patrols, those models’ predictions were not validated by extensive field tests. In this paper, we present a hybrid spatio-temporal model that predicts poaching threat levels and results from a five-month field test of our model in Uganda’s Queen Elizabeth Protected Area (QEPA). To our knowledge, this is the first time that a predictive model has been evaluated through such an extensive field test in this domain. We present two major contributions. First, our hybrid model consists of two components: (i) an ensemble model which can work with the limited data common to this domain and (ii) a spatio-temporal model to boost the ensemble’s predictions when sufficient data are available. When evaluated on real-world historical data from QEPA, our hybrid model achieves significantly better performance than previous approaches with either temporally-aware dynamic Bayesian networks or an ensemble of spatially-aware models. Second, in collaboration with the Wildlife Conservation Society and Uganda Wildlife Authority, we present results from a five-month controlled experiment where rangers patrolled over 450 sq km across QEPA. We demonstrate that our model successfully predicted (1) where snaring activity would occur and (2) where it would not occur; in areas where we predicted a high rate of snaring activity, rangers found more snares and snared animals than in areas of lower predicted activity. These findings demonstrate that (1) our model’s predictions are selective, (2) our model’s superior laboratory performance extends to the real world, and (3) these predictive models can aid rangers in focusing their efforts to prevent wildlife poaching and save animals.
KeywordsPredictive models Ensemble techniques Graphical models Field test evaluation Wildlife protection Wildlife poaching
This research was supported by MURI grant W911NF-11-1-0332, NSF grant with Cornell University 72954-10598 and partially supported by Harvard Center for Research on Computation and Society fellowship. We are grateful to the Wildlife Conservation Society and the Uganda Wildlife Authority for supporting data collection in QEPA. We also thank Donnabell Dmello for her help in data processing.
- 1.Great Elephant Census: The great elephant census—a Paul G. Allen project. Press Release, August 2016Google Scholar
- 3.Critchlow, R., Plumptre, A.J., Alidria, B., Nsubuga, M., Driciru, M., Rwetsiba, A., Wanyama, F., Beale, C.M.: Improving law-enforcement effectiveness and efficiency in protected areas using ranger-collected monitoring data. Conserv. Lett. 10(5), 572–580 (2017). Wiley Online LibraryCrossRefGoogle Scholar
- 4.Kar, D., Ford, B., Gholami, S., Fang, F., Plumptre, A., Tambe, M., Driciru, M., Wanyama, F., Rwetsiba, A., Nsubaga, M., et al.: Cloudy with a chance of poaching: adversary behavior modeling and forecasting with real-world poaching data. In: Proceedings of the 16th Conference on Autonomous Agents and MultiAgent Systems, pp. 159–167 (2017)Google Scholar
- 5.Lee, W.S., Liu, B.: Learning with positive and unlabeled examples using weighted logistic regression. In: ICML, vol. 3 (2003)Google Scholar
- 6.Nguyen, T.H., Sinha, A., Gholami, S., Plumptre, A., Joppa, L., Tambe, M., Driciru, M., Wanyama, F., Rwetsiba, A., Critchlow, R., et al.: CAPTURE: a new predictive anti-poaching tool for wildlife protection. In: AAMAS, pp. 767–775 (2016)Google Scholar
- 7.O’Kelly, H.J.: Monitoring Conservation Threats, Interventions, and Impacts on Wildlife in a Cambodian Tropical Forest, p. 149. Imperial College, London (2013)Google Scholar
- 10.Seiffert, C., Khoshgoftaar, T.M., Van Hulse, J., Napolitano, A.: Rusboost: a hybrid approach to alleviating class imbalance. IEEE SMC-A Syst. Hum. 40(1), 185–197 (2010)Google Scholar
- 11.Solberg, A.H.S., Taxt, T., Jain, A.K.: A Markov random field model for classification of multisource satellite imagery. IEEE TGRS 34(1), 100–113 (1996)Google Scholar
- 12.Yin, Z., Collins, R.: Belief propagation in a 3d spatio-temporal MRF for moving object detection. In: IEEE CVPR, pp. 1–8. IEEE (2007)Google Scholar