CyberTracker and MIKE

by Rene Beyers, April 2002

Mike is an acronym for Monitoring Illegal Killing of Elephants. The program was put in place by CITES at the 1997 Conference of the Parties to CITES in Harare.

At this conference it was decided to transfer the African elephant populations of Botswana, Namibia, and Zimbabwe from CITES Appendix I to Appendix II, and to allow limited commercial trade in raw ivory from these countries. The objective of the monitoring programme was to better inform decision makers responsible for the listing and protected status of the elephant by monitoring illegal killing of elephants, trends in elephant populations and factors affecting these trends.

The Wildlife Conservation Society in Central Africa undertook a pilot project and conducted surveys of forest elephant populations at three pilot sites.

CyberTracker was used in one of the sites (Odzala) in collaboration with ECOFAC to assess its potential for collecting data during the surveys.

How can CyberTracker help MIKE?

MIKE collects a variety of information on elephants and factors that relate to elephants. There are three main aspects of the programme in which CyberTracker could be used to facilitate data collection and data transfer.

Elephant population surveys

Elephant population surveys will be conducted regularly in about 54 sites across Africa and several more in Asia. The amount of data that will be collected is potentially enormous. In savannah, surveys will primarily rely on aerial counts of live elephants and carcasses.

In forest, surveys will be done using line transects and reconnaissance walks counting elephant dung and recording information that relates to elephants, including habitat, signs of poaching etc. It is here that CyberTracker can be play a useful role in collecting the data and transmitting it to higher levels.

Sampling units (straight line transects or a combination of recce transects) are laid out systematically across the entire survey zone (figure 1). Longer reconnaissance walks on which some data is recorded continuously or at regular intervals connect these sampling units.


Figure 1: systematic layout of transects (or recce-transects) connected by travel recces across the survey zone.

Line transects are a well-established method for wildlife censuses. A straight transect is cut through the vegetation and all animal sign that is observed from walking along the transect line, in this case elephant dung, is recorded. Perpendicular distance from the transect line to the centre of the dung pile is accurately measured. A modelling technique (distance) is used to estimate dung densities based on the probability of detecting dung piles from the transect line. The line transect is marked by a thin thread (topofil) and distance along this thread is also measured.

Recces follow a path of least resistance in a general direction and all animal sign from that path is recorded without measuring perpendicular distance. Recces are easier and faster to perform than transects but produce biased data. On these recces the number of dung piles are often over-estimated (e.g. when more dung piles are found than the average for the area by following elephant trails) or under-estimated (when less dung piles are found than the average for the area by going though dense vegetation) without really knowing the magnitude of the bias.

Cybertracker can be used for data collection on both recces and transects. On transects it is not necessary to record a GPS position because the length along the transect is measured using the topofil. Since a transect often cuts through very dense vegetation (one has to literally hack his way through lianas and shrubs) it is often difficult to receive GPS satellite signals and therefore the GPS function on cybertracker is best switched off at the beginning of the transect. Only the positions of the beginning and the end of the transect are logged.

For recces the GPS is turned on again. In order to compare recces and transects we take data from a section of the recce and we need to know the distance for each observation along the recce. Cybertracker can produce that data and do away with the need of a topofil along which distance is measured.

For longer recces (e.g. travel recces) it is also interesting to have positions for recces which can then be spatially analysed and modelled.

In any case it is critical that all observations of dung are recorded and overriding the GPS should be possible when it takes too long to get a signal or no signals can be obtained at all.

Law Enforcement Monitoring

To evaluate the effectiveness of management and protection of a park, reserve or wildlife resource Law Enforcement Monitoring is a valuable if not essential tool.

Law enforcement monitoring, LEM in brief, is a systematic approach to the monitoring of illegal use of wildlife or other protected resources, measures of protection and the results of protection.

LEM looks at the amount of protection that goes into a protected or managed site, signs of illegal activities within and around that site and results of law enforcement (e.g. ivory seizures, arrests of poachers and seizures of guns etc.).

Protection effort can be measured in terms of number of staff protecting the area, available budget, frequency and number of patrols, area covered by patrols etc.

LEM plays an important role in MIKE as illegal killing of elephants is closely linked to protection and one cannot evaluate real changes in elephant populations over time without taking into account changes in levels of protections.

For MIKE several data collection protocols and field forms (table 1) were developed representing several steps in the LEM process. A training session was organized in Central Africa to guide the use of these forms and train park guards in developing a LEM system in their area.

PROTOCOL Description
Field patrol authorization Authorization of a field patrol by the wildlife warden or his equal
Daily patrol report Filled in by field patrols

Measures patrol effort and spatial coverage

Records indicators of illegal activities and results of patrol actions

Records observations of elephant or other key wildlife species and features related to these species

Elephant carcass and ivory report Records information on elephant carcass and ivory finds

Filled in by patrols, monitoring teams, researchers and other field staff

Debriefing and field patrol summary Process whereby information gathered by field patrols is verified, amplified and communicated to the wildlife warden and park management.
Daily checkpoint Filled in by field patrols at a stationary base or checkpoint (e.g. roadblocks, guard posts etc.)

Records information on illegal transport of wildlife products, arms and other items.

Intelligence monitoring Measures effort and resources expended on intelligence gathering by protected area staff, as this relates to illegal activity.

Table 1: Protocols and forms used in LEM

The use of cybertracker in this data collection scheme should be seen as part of broader program and its main objectives are to speed up data collection in the field and transmission at the base and to assist the evaluation of the collected information during debriefing. Specific sequences for daily patrols were developed by ECOFAC (Odzala National Park) and a pilot project was recently started in the Samburu – Laikipia area in Northern Kenya.

Cybertracker is especially useful for daily patrols and daily checkpoint monitoring especially when large amounts of data are being collected (e.g. in Odzala).

Table 2 gives an overview of the kind of information being collected on daily patrols.

Type of information Key information collected
Daily patrol activity
  • Patrol route
  • Patrolling time
  • Patrol transport
  • Number of persons patrolling
Signs of illegal activity
  • Primary indicators of elephant poaching
    (elephant carcasses, elephant poaching camps, armed poachers, etc.)
  • Secondary indicators of poaching and other illegal activities
    (snares, poaching camps, human footprints etc.)
Results of antipoaching
  • Arrest of poachers, interrogation..
  • Seizure of ivory, guns
Animal and animal – related observations
  • Observations of key species
  • Observations of geographic features relating to elephants (e.g. forest clearings with fresh elephant sign…)
Elephant carcass
  • Information on carcasses and ivory found by patrols or other field staff

Table 2: key information collected on daily patrols with cybertracker

One of the major advantages of cybertracker above paper forms and GPS is that patrol route and patrolling time are automatically recorded. This avoids downloading and converting GPS trackpoints to GIS compatible formats, the rather awkward entering of named waypoints on a GPS, transcription errors etc. Cybertracker can keep an accurate track of the whereabouts of a patrol. Patrol coverage and patrolling time are important indicators of patrol effort.

Cybertracker records a GPS position automatically for each observation (provided satellite reception is possible), which facilitates spatial treatment and analysis of the data.

Some situations that occur on patrols however will make collecting data with cybertracker difficult (nobody’s first thought during shootout with armed poachers is to switch on cybertracker) and some data will have to be added later either in the field or during debriefing. We therefore recommend taking a notebook alongside cybertracker for more complex or descriptive data entry. In case of technical failure patrol team members should have backup forms or form templates (and be motivated!) to continue recording data using pen and paper.

Upon return to the base, patrols have to be debriefed. Debriefing takes place between the patrol secretary assisted by other patrol team members and the wildlife or park warden. During debriefing, information captured on the daily reports is verified, corrected and other information can be added. The PC Cybertracker software on the PC can be used in this process. Data can be checked by going through the table and GPS positions can be checked on the map. Changes to the data can be made directly in the cybertracker table or in an exported spreadsheet. Other information can be added and paper field notes should be integrated in the database.

Editing spatial data is better done in a GIS package such as Mapinfo or Arcview. If no GIS is available on site, maps can be printed from within PC cybertracker and editing can be done by hand. Corrections and additions should be made in a different colour than the one that was used in the field to distinguish original field data from edited and added data during debriefing. Maps either on screen or on paper are an efficient way of communicating patrol information to other park and wildlife management staff.

Graphs on frequency of observations, effort and the relationship between effort and "catch" (e.g. seized weapons, arrested poachers…) can be produced either using cybertracker software or a spreadsheet programme.

Ancillary data

Many factors that have an impact on elephants are spatial and information on these is collected from satellite images and field data or a combination of both. Elephant densities can then be related to these factors. Because of the spatial nature of much of this information the cybertracker is a very useful tool in collecting this data.

Table 3 shows information that is being collected for MIKE. Cybertracker can be used to collect field data on human access, presence and activity and habitat. This information is combined with satellite image data to produce a spatial database and basemaps of a site.

Some of the variables thus measured can be used in a spatial modelling framework to explain variation in elephant densities and is also informative on threats to wildlife and protection.

A "socio-economic" sequence was developed with ECOFAC for collecting this information (except habitat). A separate geographic and habitat sequence may be developed in the future to collect specific information on habitat, ground control points and selected geographic features for ground verification of satellite images.



Satellite data Cybertracker
Human access Roads Road access and condition, road point features (bridges etc.)
Rivers River access and river point features (waterfalls, etc.)
Human Presence Major towns and human occupation

Socio-economic indicators

Human settlements (villages, camps) and socio-economic data (demography, number of houses, ethnic groups, schools, clinics, main economic activities…).

Presence of administrative units (local and county administration, police, military,…)

Human Activity Forest exploitation, mining Human activity sign

Forestry concession boundary

Exploitation settlements

Conservation status Verification of protected area boundaries Demarcation of protected area.
Protected area infrastructure (e.g. guard posts), tourist and other infrastructure
Patrol routes, law enforcement results
Hydrology Main Rivers and lakes River junctions, navigable rivers
Habitat Vegetation Vegetation ground verification

Table 3: Geographic features of basemaps collected with cybertracker in combination with satellite imagery.

Technical issues

locate satellites in thick forest cover, batteries may run out during the patrol, mistakes may be made, the GPS may get dropped in a river or otherwise damaged. In addition, in times of stress, such as encouters with poachers, taking a GPS reading may not be first priority!

Cybertracker units proved to be quite reliable in the field although they remain fragile and should be handled with great care in a tropical forest environment. Teams are often out in the field for several days to weeks and if a unit breaks down one has to make sure that data doesn’t get lost. While making backups of the data was not possible with the Palm Pilot III, with the current handspring visor it is possible to insert memory cards in the expansion slot on which backups can be easily made. It is recommended to take at least one backup of the whole dataset daily and preferably on at least two different memory cards that are stored in lockable plastic bags with different people.

GPS signal reception is not always sufficient in the forest to get a position fix within a reasonable time. GPS should be turned off on transects if it slows down the team. Positions of observations on a transect can be calculated afterwards if necessary.

Sometimes it will also be necessary to override the GPS on recces when it takes too much time to get a reading. In any case observations should not be discarded when no GPS position can be taken.