Elsevier

Biological Conservation

Volume 189, September 2015, Pages 33-38
Biological Conservation

Special Issue Article: Conservation Crime
Mobile decision-tree tool technology as a means to detect wildlife crimes and build enforcement networks

https://doi.org/10.1016/j.biocon.2014.08.018Get rights and content

Highlights

  • Enforcement of illegally traded wildlife requires ability of customs officers and police to detect crime.

  • Present case studies of mobile tools to identify wildlife crime in China, Vietnam, and the U.S.

  • Law enforcement agents in China find tool “easy to use”; and >80% expect to use tool in future.

  • After one year of trial, users in Vietnam are increasing; and buy-in from government agencies is secure.

  • Limited trial of decision-tree tools for species ID is promising for enforcement of wildlife crimes.

Abstract

Accurate field identification of illegally traded wildlife and wildlife products is critically important in the detection and suppression of wildlife crimes. Yet many law enforcement officers and concerned citizens lack access to resources for identifying species and products; this is particularly true for those with no formal expertise in biology, zoology or wildlife training. Emerging digital technologies such as mobile applications may provide important easy-to-use decision-tree style tools for in situ identification. With local government and civil society partners, we are piloting such tools in China and Vietnam to identify whole animals and ivory products; and in the United States developing tools that will be used at U.S. military bases in Afghanistan to identify species from wildlife products. We are coordinating these efforts to minimize redundancy and overhead; we benefit from shared backend support for a photo database and species ID keys that can be translated easily to ensure enough flexibility for targeting needs of the specific country and audience. Planned inclusion of ‘ask the expert’ and geolocation functions will increase accuracy in identification and aid monitoring and research of supply chains. For these emerging technologies to be successful, deployment must be accompanied with on-the-ground trainings to recruit and retain enforcement personnel. The establishment of a supporting network of experts and a user community will be critical for long-term implementation and evaluation of success. Preliminary response from users of a pilot app in China demonstrates high potential for employing these technologies as routine tools to help fight wildlife crimes.

Introduction

The illegal trade of animals and products made from parts of animals is a major global threat to wildlife. In the last 25 years, populations of many iconic species such as elephant, rhinoceros, tiger, and saiga antelope have decreased dramatically or become locally extinct, and much of this decline is due to illegal hunting for coveted products such as ivory, medicines, and clothing (Emslie et al., 2013, Li et al., 2007, Maisels et al., 2013, Milner-Gulland et al., 2001, Walston et al., 2010). Weak enforcement systems have enabled wildlife to be poached and traded to markets where sales of wildlife and products made from wildlife parts fuel illegal business (Bennett, 2011). Responsibility for enforcement is spread across individuals along supply chains, from rangers to transportation and border guards. Lack of awareness and interest in wildlife conservation are one of the major obstacles, along with lack of training and technical resources, for combating these crimes (Bennett, 2011, Kaaria and Muchiri, 2011). Law enforcement authorities stationed at borders, including customs, immigration, police, military, agricultural and transportation personnel (The World Bank, 2013) are at the front line of the wildlife crime trade-chain battlefield in ensuring illicitly traded products are not crossing borders or being offered for sale in local markets. The rapidly increasing international trade and flow of travelers has put considerable pressure on border authorities, who must screen a broad range of goods and decide whether or not to release a shipment in a matter of minutes or even seconds (Gerson et al., 2008). Building capacity of enforcement personnel can increase the quality of market surveillance for illegally traded wildlife. However, the ability for these law enforcement entities to impact illegal wildlife trade requires having access to easy-to-use tools and resources to aid in the identification of traded wildlife and products.

Knowledge, motivation, and technical resources specific to wildlife are three main factors that can improve the impact of law enforcement agencies on illegal wildlife trade (Bennett, 2011, Gerson et al., 2008, Kaaria and Muchiri, 2011). Ideally, customs agents would have extensive training as well as easy access to the latest forensic techniques and laboratories to facilitate identification of wildlife to the species level (Bell, 2011, Rosen and Smith, 2010). In reality, training is limited; and, while promising, forensic science must occur in properly equipped laboratories (Anderson, 2013), far away from the front lines of trade activity. In the past, organizations have experimented with wild species identification prototypes mimicking field guides (Tallant et al., 2010). This approach revealed significant practical limitations as it required some previous knowledge of wildlife to ultimately be successful in identifying to a species level. Capitalizing on this experience, the “decision-tree style” may be a more viable approach to allow successful operation by officers with relatively little training to make quick and reliable identification of potential crimes in the field (Rosen and Smith, 2010).

Decision-trees are a system to provide classification based on inputs and organization of information. They have wide-spread application, from determining water treatments, to diagnosing medical conditions and classifying astronomical images (Murphy, 2001, Murphy and Olson, 1996, Salzberg et al., 1995). Often, decision-trees facilitate analysis by assigning probabilities to supplied data selections, with additional weights to account for potential errors, to calculate the outcome of system manipulations (Murphy, 2001). Decision-trees can also be simple dichotomous or multichotomous keys employed to identify items of the natural world such as flowering plants, minerals, or insects (Carle, 2010, Alan Plante et al., 2003, Watts, 1998). These keys use a process of elimination to narrow the possibilities at each selection and can be effective as a tool to classify or identify based on different factors. This approach could be equally useful for identification of traded wildlife and wildlife products but until now has yet to be tested.

Emerging technologies such as interactive web pages and mobile applications provide an opportunity to make multichotomous keys available to a variety of audiences at a relatively low cost, as use of mobile technologies is booming around the globe (Asia Digital Marketing Association, 2012, Smith, 2013). For these tools to curb wildlife trade at the scale necessary to prevent decline of wildlife populations, organizations must deploy them in strategic locations and to a network of individuals engaged in using them for specific purposes. The Wildlife Conservation Society (WCS) has been involved in the enforcement of wildlife trade for over a decade supporting networks of trained park rangers and training customs, police and forestry officers in places such as Congo, Thailand, China, Indonesia, Malaysia, and Vietnam (Bennett, 2012, Lee et al., 2005, Roberton, 2013, Walston et al., 2010, Wildlife Conservation Society, 2013a) and training military police to secure on-base markets (Kretser, 2012, Kretser et al., 2012). These networks represent opportunities to test whether emerging technologies using decision-tree approaches will be a viable method for enforcement. We present three cases from WCS field programs in China, Vietnam, and the United States illustrating how to build a network of users employing decision-tree style tools to improve detection of wildlife crimes.

Section snippets

Methods

WCS is developing and pilot-testing decision-tree style technological resources to identify wildlife and products made from wildlife parts to a species level across three diverse contexts and audiences in China, Vietnam, and the United States. All authors on this paper are WCS staff and are intimately involved and leading this development process. To streamline mobile technology production efforts, we are creating a back-end database structure that can be adapted to multiple tools and includes

China

With a population of over 1.3 billion, China is potentially the largest demand market for several species of wildlife globally due to the traditional use of wildlife for medicine, food, and trophies and the increase of that use due to rapid economic development. In recent years, Chinese authorities have progressively acknowledged the impact illegal wildlife trade has on global biodiversity as well as the country’s international reputation (Hu Jintao, 2012). The destruction of 6.15 tons of ivory

Conclusions

Decision-tree style resources for field identification of species in potential wildlife crimes are promising as an emerging enforcement tool. The tools in either app or web-based format can be adapted to particular contexts and are expandable, easy to use, and helpful for individuals with little to no training in wildlife as is the case with many enforcement individuals (Bennett, 2011, Rosen and Smith, 2010). Mobile app technology also has an additional benefit in that a user does not require

References (47)

  • Carle, D.M. 2010. Identification keys to the larvae and pupae of New Jersey black flies (Diptera: Simuliidae. Entomol...
  • R. Chandan et al.

    Wildlife enforcement monitoring system (WEMS): a solution to support compliance to multilateral environmental agreements

    Gov. Inform. Q.

    (2011)
  • Emslie, R.H., Milliken T., Talukdar B., 2013. African and Asian rhinoceroses – status, conservation and trade report to...
  • H. Gerson et al.

    Monitoring international wildlife trade with coded species data

    Conserv. Biol.

    (2008)
  • Hu Jintao, C.H., Making great efforts to promote ecological progress. Report to 18th Party Congress, November, 2012,...
  • B.I. Kaaria et al.

    Enforcement challenges across borders: detecting and prosecuting illegal wildlife trafficking

    Ninth Int. Conf. Environ. Compliance Enforcement

    (2011)
  • Kretser, H.E. An unconventional battle on U.S. bases: training the military to curb wildlife trafficking. Wildlife...
  • H.E. Kretser et al.

    Demand for wildlife trade products available to U.S. military personnel serving abroad

    Biodivers. Conserv.

    (2012)
  • Kulish, N. Ivory culprit in kenya: Ex-official from U.S. New York Times, July 24, 2013. Accessed on January 27, 2014....
  • E. Kuntsche et al.

    Using personal cell phones for ecological momentary assessments: an overview of current development

    Eur. Psychologist

    (2013)
  • Li, L., Zhou, Y., Bennett, E.L., 2007, Report of a survey on saiga horn in markets in China. Report to the 14th meeting...
  • L. Liebenberg et al.

    Rhino tracking with the CyberTracker field computer

    Pachyderm

    (1999)
  • F. Maisels et al.

    Devastating decline of forest elephants in central Africa

    PLoS ONE

    (2013)
  • Cited by (16)

    • Challenges and perspectives on tackling illegal or unsustainable wildlife trade

      2021, Biological Conservation
      Citation Excerpt :

      The potential value of captive breeding for conservation has been undermined by fraudulent practices (Lyons and Natusch, 2011), and harvest from the wild may continue despite captive-bred alternatives (Macdonald et al., 2021). Some tools that could be more widely developed and adopted include user-friendly aids for border officers such as online ID guides (CEC, 2021) and apps such as Wildlife Alert (for helping identification of illegal wildlife products) or iNaturalist (for automated or community-driven species identification via photography of lifeforms or their traces) (Kretser et al., 2015; iNaturalist, 2021; Wildlife Conservation Society, 2021). Forensic DNA databases optimised towards endangered and traded species (Ahlers et al., 2017), such as the Barcode of Wildlife Project (http://www.barcodeofwildlife.org/) can identify species, trace the origin of traded products (Mwale et al., 2016), and detect counterfeits (Hellberg et al., 2019).

    • Systematic reassessment of Laos’ protected area network

      2019, Biological Conservation
      Citation Excerpt :

      Geospatial social media content is also providing interesting new opportunities for environmental monitoring (Di Minin et al., 2015), while counterfactual analyses are providing ways to avoid selection bias in the monitoring (Geldmann et al., 2013). These newly available datasets and ways to analyze data should be made more accessible for developing country governments through readymade platforms, generalized training packages and support to expert networks, while governments should be encouraged to include these new datasets and tools to their traditional monitoring efforts (Kretser et al., 2015). Even though datasets based on earth observation, species distribution models, social media content and crowdsourcing have improved systematic conservation planning and monitoring enormously in recent decades (Dickinson et al., 2012; Turner, 2014), these are not able to replace biodiversity data produced through field surveys and inventories.

    • Understanding non-compliance: Local people's perceptions of natural resource exploitation inside two national parks in northeast Brazil

      2017, Journal for Nature Conservation
      Citation Excerpt :

      Nevertheless, many conservationists remain skeptical, fearing that CBC approaches may be more effective at changing users’ attitudes rather than their behaviours (Infield and Namara, 2001; Lepp and Holland, 2006). New technologies (remote-sensing, drones, etc.) and global information infrastructure (i.e. the Internet) have also recently provided an innovative and low-cost way to support the enforcement of conservation and PAs regulations (Fuller, 2006; Kretser et al., 2015). Nevertheless, these new tools raise new concerns about data privacy, data security, and regulatory reach (di Vimercati, Genovese, Livraga, Piuri, & Scotti, 2013).

    View all citing articles on Scopus
    View full text