Monitoring the Ilegal Killing of Elephants (MIKE) is one of two global monitoring programs for elephants that was initiated in 1998 by CITES (Convention for International Trade in Endangered Species). Specific objectives of MIKE are:
- to measure levels and trends in the illegal hunting of elephants;
- to determine changes in these trends over time; and
- to determine the factors causing or associated with such changes, and to try and assess in particular to what extent observed trends are a result of any decisions taken by the Conference of the Parties to CITES.
Data to meet MIKE objectives are obtained from a number of selected sites in Africa and Asia. At these sites surveys are carried out to estimate the size of elephant populations. In addition, anti-poaching patrols record the elephant carcasses that are found on patrol and whether or not the carcass was illegally killed. Making sense of the data from the anti-poaching patrols is difficult because they are not random surveys and because of the priority of an anti-poaching patrol is not to record data but to stop poaching.
MIKE must report to the CITES Conference of the Parties (CoP), the last one was is in March 2013.
I have been involved with MIKE since 2000, where I carried out an analysis of a pilot study in Central Africa, working with Steve Buckland at CREEM, St Andrews and the Wildlife Conservation Society.
In 2003 I worked with Bob Burn to produce a data analysis strategy for MIKE. This describes the flow of data from data collection on site through to sub-regional offices and to MIKE and then CITES. It also describes how information can also be used at each level in the hierarchy itself for management purposes and with added value can flow back from the top to those on the ground. For this to work capacity building is required at all stages in the hierarchy. The strategy also describes the range of analytical tools that may be required for: summarising and understanding data on the ground at one particular site; multi-site and cross-continent analyses for reporting to CoPs; integrated analyses with other elephant monitoring programs such as ETIS and the African Elephant Database.
Since 2004 I have been involved in the analysis of MIKE data for reporting to CITES. Initially, these reports were relatively simple summaries of the data with clear caveats included in the interpretation of the results. Since 2009, we have been working to provide more detailed analyses of the data to describe global trends in the illegal killing of elephants over time. These analyses must account for differences between sites in data collection. We have developed an indicator called PIKE (Proportion of illegally killed elephants) which should remove some of the differences between sites. Other differences between sites and countries ave been investigated by exploring the effects of different variables that describe site and regional characteristics. To be able to do this correctly and take account of the structure of the data (observations over time within a site which sits within a country) a Bayesian hiearchical model was implemented. The results were presented in a report to the CoP in 2010. The methodology,as well as short comings and difficulties of using these methods were publised in PLOS One in 2011.
More recently I worked with the Wildlife Conservation Society to look at the impact of varying natural mortality on PIKE and whether elicitation techniques can help improve estimates of poaching pressure.