Wildlife criminal activity exhibits distinct spatial patterns influenced by environmental and anthropogenic drivers. This study applies Maximum Entropy (MaxEnt) modelling, a presence-only and robust machine learning technique originally designed for species distribution, to estimate the potential distribution of suitable sites based on recorded wildlife crime incidents, in Gonarezhou National Park (GNP), Zimbabwe. A total of 1,305 georeferenced crime incident locations were analysed alongside a suite of environmental predictors including elevation, slope, Normalised Difference Vegetation Index (NDVI), distance to roads, distance to settlements, distance to water bodies, and distance to park boundary. The model achieved an Area Under Curve (AUC) of 0.9, indicating excellent predictive performance. Among the predictors, elevation, distance to settlements and roads emerged as the most influential variables. The spatial distribution of crime suite ability revealed heightened crime risk near park boundaries and adjacent communities, reflecting the interplay between terrain, accessibility, and land-use gradients as key determinants for wildlife crime. These findings highlight the value of integrating ecological modelling techniques into conservation criminology and support the implementation of spatially targeted law enforcement strategies within protected areas. The resulting surface reflects spatial patterns of recorded incidents influenced by patrol detection effort rather than unbiased estimates of actual crime occurrence.
Wildlife criminal activity exhibits distinct spatial patterns influenced by environmental and anthropogenic drivers. This study applies Maximum Entropy (MaxEnt) modelling, a presence-only and robust machine learning technique originally designed for species distribution, to estimate the potential distribution of suitable sites based on recorded wildlife crime incidents, in Gonarezhou National Park (GNP), Zimbabwe. A total of 1,305 georeferenced crime incident locations were analysed alongside a suite of environmental predictors including elevation, slope, Normalised Difference Vegetation Index (NDVI), distance to roads, distance to settlements, distance to water bodies, and distance to park boundary. The model achieved an Area Under Curve (AUC) of 0.9, indicating excellent predictive performance. Among the predictors, elevation, distance to settlements and roads emerged as the most influential variables. The spatial distribution of crime suite ability revealed heightened crime risk near park boundaries and adjacent communities, reflecting the interplay between terrain, accessibility, and land-use gradients as key determinants for wildlife crime. These findings highlight the value of integrating ecological modelling techniques into conservation criminology and support the implementation of spatially targeted law enforcement strategies within protected areas. The resulting surface reflects spatial patterns of recorded incidents influenced by patrol detection effort rather than unbiased estimates of actual crime occurrence.
Price includes VAT (Russian Federation)
Instant access to the full article PDF.
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request .
Anagnostou M, Moreto WD, Gardner CJ, Doberstein B (2021) Poverty, pandemics, and wildlife crime. Conserv Soc 19(4):294–306
Anderson DR, Burnham KP, White GC (1998) Comparison of Akaike information criterion and consistent Akaike information criterion for model selection and statistical inference from capture-recapture studies. J Applied Statistics 25(2):283–282
Brantingham PJ, Brantingham PL (1981) Environmental criminology. Sage, London
Brantingham PL, Brantingham PJ (1993) Nodes, paths, and edges: Considerations on the complexity of crime and the physical environment. Classics Enviro Criminol, ental (8) pp. 289–326
Brantingham PL, Brantingham PJ (2017) Environment, routine, and situation: Toward a pattern theory of crime. In Ronald V. Clarke & Marcus Felson (Eds) Routine activity and rational choice. Routledge, pp 259–294
Chadwick FJ et al (2024) LIES of omission: complex observation processes in ecology. Trends Ecol Evol 39(4):368–380
Cohen LE, Felson M (2010) Social change and crime rate trends: A routine activity approach (1979). In M. A. Anderson, P. J. Brantingham & B. J. Kinney (Eds). Classics in environmental criminology. Routledge, Milton Park, pp 203–232
Cohen LE, Marcus F (1979) Social change and crime rate trends: In M.A. Anderson, P. J. Brantingham, &J.B. Kinney (Eds). A routine activity approach. Am Sociol Rev, pp. 588–608
Cowan D et al (2020) Applying crime pattern theory and risk terrain modeling to examine environmental crime in Cambodia. J Contemp Crim Justice 36(3):327–350
Critchlow R et al (2015) Spatiotemporal trends of illegal activities from ranger-collected data in a Ugandan national park. Conserv Biol 29(5):1458–1470
Cunliffe R, Tom M, Mapaura A (2012) Vegetation survey of Gonarezhou National Park, Zimbabwe. Zimbabwe Parks and Wildlife Management Authority, Harare
Da Re D et al (2023) USE it: Uniformly sampling pseudo-absences within the environmental space for applications in habitat suiteability models. Methods Ecol Evolutions 14(11):2873–2887
Dudley N, Stolton S, Wendy E (2013) Wildlife crime poses unique challenges to protected areas. Parks 19(1):7–12
Duffy R (2022) Crime, security, and illegal wildlife trade: Political ecologies of international conservation. Glob Environ Politics 22(2):23–44
Elith J et al (2011) A statistical explanation of MaxEnt for ecologists. Divers Distrib 17(1):43–57
Esmail N et al (2019) Emerging illegal wildlife trade issues in 2018: a global horizon scan. Conserv Lett 13(4):e12715
Gavin M, Solomon J, Blank S (2010) Measuring and monitoring illegal use of natural resources. Conserv Biol 24(1):89–100
Goga T et al (2025) Detection of potential illegal environmental activities in Slovakia based on earth observation data. J Maps 21(1):2464054
Gore ML, Escouflaire L, Wieland M (2021) Sanction avoidance and the illegal wildlife trade: A case study of an urban wild meat supply chain. J Illicit Economies Dev 3(1):43–67
Hansen AJ, DeFries R (2007) Ecological mechanisms linking protected areas to surrounding lands. Ecol Appl 17(4):974–988
Huberman DB (2020) Advances in Spatial Statistics for Ecological and Environmental Data. North Carolina State University, North Carolina
Kahler JS, Gore ML (2015) Local perceptions of risk associated with poaching of wildlife implicated in human-wildlife conflicts in Namibia. Biol Conserv 189:49–58
Kaky E, Nolan V, Alatawi A, Gilbert F (2020) A Comparison Between Ensemble and MaxEnt Species Distribution Modelling Approaches for Conservation: A Case Study With Egyptian Medicinal Plants. Ecol Inf 60:101150
Kurland J, Pires S, Moreto W (2014) Wildlife crime: a conceptual integration, literature review, and methodological critique. Crime Sci 3(1):1–15
Lemieux A, Clarke R (2009) The international ban on ivory sales and its effects on elephant poaching in Africa. Br J Criminol 49(4):451–471
Lindsey P et al (2015) Illegal hunting and the bush-meat trade in Savanna Africa. Zoological Society of London Wildlife Conservation Society, Panthera
Low BW, Yiwen Z, Heok HT, Yeo DCJ (2021) Predictor complexity and feature selection affect Maxent model transferability: Evidence from global freshwater invasive species. Divers Distrib 27(3):497–511
Manel S, Jean-Marie D, Ormerod SJ (1999) Comparing discriminant analysis, neural networks and logistic regression for predicting species distributions: a case study with a Himalayan river bird. Ecol Model 120(2–3):337–347
Moreto WD (2016) Occupational stress among law enforcement rangers: Insights from Uganda. Oryx 50(4):646–654
Moreto WD, Lemieux AM (2015) Poaching in Uganda: Perspectives of law enforcement rangers. Deviant Behav 36(11):853–873
Muboko N (2011) Conflict and sustainable development: the case of the Great Limpopo Transfrontier Park (GLTP); Southern Africa. Nelson Mandela Metropolitan University, Port Elisabeth
Murwendo T, Murwira A, Masocha M (2023) Vegetation phenology patterns in semi-arid savannah woodlands of Gonarezhou National Park, Southeastern Zimbabwe. Int J Geoheritage Parks, 11(2) pp. 298–309
Mutanga CN, Gandiwa E, Muboko N (2017) An analysis of tourist trends in northern Gonarezhou National Park, Zimbabwe. Cogent Social Sci 3(1):1392921
Muzhingi DT (2021) Environmental niche modelling of elephant poaching sites in the Matusadona National Park. University of Johannesburg, Zimbabwe. Johannesburg
Naimi B et al (2014) Where is positional uncertainty a problem for species distribution modelling? Ecography 37(4):191–203
Pearce J, Simon F (2000) Evaluating the predictive performance of habitat models developed using logistic regression. Ecol Model 133(3):225–245
Phillips SJ, Miroslav D (2008) Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation. Ecography 31(2):161–175
Pires SF, Moreto WD (2017) Preventing wildlife crimes: Solutions that can overcome the ‘Tragedy of the Commons’. Transnational environmental crime, pp. 419–442
R Studio Team (2023) RStudio: Integrated Development Environment for R. RStudio, PBC. Available at https://posit.co/products/open-source/rstudio/. Accessed 20 July 2025
Raschka S (2018) Model evaluation, model selection, and algorithm selection in machine learning. arXiv preprint arXiv Volume 1811:12808
Reineking B (2006) Constrain to perform: regularization of habitat models. Ecol Model 3(4):675–690
Schmitt S et al (2017) ssdm: An r package to predict distribution of species richness and composition based on stacked species distribution models. Methods Ecol Evol 8:1795–1803
Shaminja TS, Igyo JA, Terhile AJ (2025) Crimes Against Biodiversity in Nigeria: A Search for Sustainable Development. Torkwase J Agricultural Res 2(1):1–15
Skidmore A (2023) Exploring the motivations associated with the poaching and trafficking of Amur tigers in the Russian Far East. Deviant Behav 44(3):331–358
Solomon JN, Gavin MC, Gore ML (2015) Detecting and understanding non-compliance with conservation rules. Biol Conserv 189:1–4
Swets JA (1988) Measuring the accuracy of diagnostic systems. Science 240(4857):1285–1293
Tavuyanago B (2016) Living on the fringes of a protected area: Gonarezhou National Park (GNP) and the indigenous communities of South East Zimbabwe 1934–2008. University of Pretoria, Pretoria
Titurus B, Friswell MI (2008) Regularization in model updating. Int J Numer Methods Eng 75(4):440–478
Valavi R, Guillera-Arroita G, Lahoz‐Monfort JJ, Elith J (2022) Predictive Performance of Presence‐Only Species Distribution Models: A Benchmark Study With Reproducible Code. Ecol Monogr 92:e01486
Walker JT (2017) Advancing science and research in criminal justice/criminology: Complex systems theory and non-linear analyses. In J. T. Walker (Ed.), Social, Ecological and Environmental Theories of Crime. Routledge, Oxfordshire, pp 499–525
Wyatt T, van Uhm D, Nurse A (2020) Differentiating criminal networks in the illegal wildlife trade: organised, corporate and disorganised crime. Trends Organised Crime 23(4):350–366
The authors have no relevant financial or non-financial interests to disclose. The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.
Department of Geography, Geospatial Sciences and Earth Observation, University of Zimbabwe, Harare, Zimbabwe
Cliff Jawe, Mark Zvidzai, Honour Chinoitezvi, Paul Dzikiti & Ratidzo Blessing Mapfumo
International Conservation Affairs Department, Zimbabwe Parks and Wildlife Management Authority, Harare, Zimbabwe
Patience Gandiwa
Scientific Services, Parks and Wildlife Management Authority, Causeway, Zimbabwe
Honour Chinoitezvi
Authors
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Cliff Jawe, Mark Zvidzai, Paul Dzikiti, Ratidzo Blessing Mapfumo and Fadzai Michelle Zengeya. The first draft of the manuscript was written by [Cliff Jawe] and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Correspondence to Mark Zvidzai.
The authors declare no competing interests.
Communicated by Vinicius R. Tonetti
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
Jawe, C., Zvidzai, M., Chinoitezvi, H. et al. Integrating environmental and anthropogenic drivers in MaxEnt models to understand the spatial patterns of wildlife crime. Biodivers Conserv 35, 207 (2026). https://doi.org/10.1007/s10531-026-03414-9
Received: 14 October 2025
Revised: 31 May 2026
Accepted: 27 June 2026
Published: 06 July 2026
Version of record: 06 July 2026
DOI: https://doi.org/10.1007/s10531-026-03414-9