Title
Analysing trends of illegal activities from Ranger-collected data in Queen Elizabeth National Park
Author(s)
R.Critchlow
Published
2014
Abstract
In this report we provide a draft manuscript of how our method can be applied to
analyse illegal activities using Management Information System (MIST) data from a
single national park: the Queen Elizabeth National Park (QENP), Uganda. In
addition, we detail: how we have gone about this process, any problems
encountered, the key results from the initial analysis, and finally the immediate
plans and future steps.
The aims of this primary analysis were to develop a method to use ranger-based
monitoring data (1) map the spatial distribution of illegal activities, (2) identify
the influential drivers of these activities, and (3) assess the spatial and temporal
trends of illegal activities. Our current approach can be applied across multiple
protected areas, and importantly accounts for observation effort. With accurate
knowledge of the locations and processes that drive different types of illegal
activities, rangers can more effectively target problems.
Existing methods to assess patterns of illegal activities from ranger based
monitoring include analysis of raw patterns or use of encounter rates. However,
these simple methods give highly biased results as the statistics used are
developed for situations where survey data is random or evenly spread across a
protected area, and ranger-based monitoring is focussed on areas where illegal
activities are expected to be highest. Encounter rates or catch per unit effort
(CPUE) are an improvement on analysis of raw or uncorrected data, but have their
own additional biases. For example, CPUE may not reflect the underlying trends of
illegal resource use if the efficiency of ranger patrols improves over time.
Additional pitfalls of the CPUE method are that it assumes reporting of illegal
activity is proportional to patrol effort and that observing illegal activities is
constant across space and time. This is unlikely because ranger patrols will rarely
perfectly cover a survey area, and proportionally more effort will be needed to
detect remaining illegal activity (Keane, Jones & Milner-Gulland 2011). Depending
on the particular assumptions made, the consequences of these biases may lead to
systematic over- or under-estimate of illegal activities with little information on
the scale of the bias, and always lead to uncertain trends.
Recognising this problem, we have taken an analysis approach that accounts for
surveillance effort by estimating the probability of reporting an illegal activity
independently from assessing the biotic and abiotic drivers of illegal activities.
This type of analysis is based on an approach used to analyse volunteer-based
records of bird distributions and change in regions with highly variable observer
effort in space and time (Beale et al. 2013) and is fully described in a recent paper
describing species distribution modelling (Beale, Brewer & Lennon 2014).
Keywords
MIST
Full Citation
R. Critchlow, C. Beale, M. Driciru, A. Rwetsiba, F.Wanyama, C. Tumwesigye, E. Stokes and A.J.Plumptre
Access Full Text
Back
DMX2721700000