Home Articles Wildlife GIS: Spatial Analysis and Visualisation in Masai Mara

Wildlife GIS: Spatial Analysis and Visualisation in Masai Mara

Wildlife GIS: Spatial Analysis and Visualisation in Masai Mara

Joram Nduati Kinuthia
Recently graduated
University of Nairobi,
Email: [email protected]

Prof A. J. Rodrigues
University Of Nairobi

Richard Oluoch
Kenya Wildlife Services
Email: [email protected]

Through the years, organizations have continued to accumulate information. This information is mainly on the organizations’ business processes, transactions and the operating environment.
Due to the very nature of organizations being geographically distributed, it is increasingly becoming important to convey the information spatially.

As thus, Geographical Information Systems (GIS) are becoming useful tools in collection, storage, manipulation and representation of spatial data. GISs are able to present a large amount of data in a short period of time on a map, using a geographical coordinate system. It also becomes very easy to grasp spatial data as compared to data stored in relational database.

GISs have been used to support strategic decisions in a variety of government and business activities in areas such as housing, healthcare, land use, natural resources, environmental monitoring, public health, transportation, retail and routing.

This project aims to analyze and visualize different datasets (wildlife, vegetation, contour, rivers) on the Masai Mara Game Reserve. This will be through creating a geo-database, suitable user interface, and availing tools that will aid in geo-processing. The results from the system will aid various stakeholders in decision making.

Information and telecommunications technologies (ICTs) have seen explosive growth and application in many areas, especially commerce. However, not much has been done in the management of natural resources.
Kenya is naturally endowed with beautiful scenes and wildlife. Such natural resources have brought the country monetary income. To sustain this, game reserves and national parks have been created to ensure continuity through conservation of the fauna and flora.

Through the years, Kenya Wildlife Services (KWS) has collected data on wildlife distribution, with an aim of making meaningful analysis that would help in policy making and management. KWS spends over ksh 40 million annually to collect information on the resource base and changes that occur in national parks, game reserves, and biodiversity areas.

Thus, this paper aims to look at how ICT can be applied, through the use of GIS to structure selected data of Masai Mara game reserve, analyze, and present the results in an easy to interpret and understand way, thus aiding decision making.

A model system was developed, on which this paper investigates and reports.

This spectacular expanse of open grassland covers 320 sq km in the south-west corner of Kenya. It is an area of gently rolling hills, woodland and acacia trees which is watered by the Mara and Talek rivers and opens onto the Serengeti plains of Tanzania.

The Masai Mara Game Reserve is often called simply “The Mara” which is the Maa word meaning “Mottled” – a reference to the patchy landscape. The Masai Mara is a Game Reserve (sometimes called a National Reserve) although an inner area is treated as a National Park.

The highlight of the Mara is undoubtedly the great migration of wildebeest which move north from the Serengeti in July and August in search of lush grass. They return south in October before the rainy season.

The Masai Mara region is the traditional land of the Masai people who often find themselves in competition with the wildlife and tourists for scarce resources.
Three year wildlife census data was available, generously through Mr. Odongo (co-author).


GIS have been used to effectively store, retrieve, manipulate, analyze and display geographic information.
Masai Mara Wildlife GIS enabled manipulation of the data, extraction of relevant information and representing the same in a meaningful way to the users.

The end product of the development process saw the delivery of the following;

  • A geo-database. Provided the framework and structures to store the spatial and attribute data in an organized way.
  • A custom VB program embedded in ArcGIS. This provided an easy-to-use user specified functionalities, without necessarily interacting directly with ArcGIS.
  • A web-enabled GIS client system developed using Apache, Mapserver and Chameleon. This was accessed via web browsers within a networked environment.

The system was designed run on a networked environment. The core program (custom VB embedded in ArcGIS) runs on a server. The geo-database resides in the same machine.
The GIS clients can be accessed from any machine within the network. The user is required to type the uniform resource locator (url) of the client and supply his username and password on his favourite web browser

Creation of a geo-database
Currently, there is no database structure in place that can efficiently store and allow manipulation of spatial data.

The datasets provided were animal wet counts for years 2000, 2001 and 2002. They fall under the feature dataset count.

Administrative shape file for Masai Mara was also provided. Other datasets provided include vegetation (Africover), roads, and contours.

Summary of datasets
Three year population datasets for Masai Mara were provided. These were wet counts for 2000, 2001, and 2002. The species in the data include buffalo, cattle, sheep, rhino, lion, giraffe, leopard, elands, etc.
Vegetation data from Africover was also used. Vegetation in Mara mainly includes scattered trees, shrubs, thickets, grasslands and savannah.

For purposes of simplicity, a few species were chosen for analysis.

Figure 1 The Mara geo-database structure

Custom VB program

Figure 2 Custom VB program: User interface and dialogue design

GIS client

Figure 3 GIS client: User interface and dialogue design

System architecture

Figure 4 System architecture

ArcGIS 9.0 provides the platform on which the custom VB application runs. It is located on a central computer (server) where the main user will be a GIS developer or advanced user.
A communication network (local area network) provides the link between the central computer and the GIS clients running from different locations.

Users at the client side have to login in order to access information from the geo-database.

Several outputs were expected from the system. Of notable interest was the ease of use of the application, supported by the easily accessible menu and toolbar.

System performance was highly dependent on which tool was executed on what dataset. For example, creating a new layer from a population count took less than 30 seconds where as running the Nearest Neighbour tool took 2-3 minutes on a single species of animals.

The qualities of outputs (reports, maps) were very high and aesthetically appealing.


This provided for generation and output of analysis results. ‘Data Charts’ menu allowed for graphical output of summary tables. This required the user to first generate a summary table before executing this tool.

‘Thematic Reports’ menu provided a mechanism for loading already processed shape files and layers.
Based on the outputs of the system, decision makers can easily be able to extract and quantify the needed information, and present the same in a usable form.

Figure 5 3 year buffalo trends 2000-2002

Figure 6 GIS Client showing elephant distribution and open-low shrubs

Figure 7 All blocks 2000 population summary report


This work has gone further by;

  • Using more recent tools in implementing the GIS application (ArcGIS 9.0).
  • Providing a single, easy to launch and use interface for custom ArcGIS application.
  • Interfacing ArcGIS with Visual Basic, which provides more flexibility in terms of user interface design, and functionality.
  • Highly customizing the ArcGIS 9.0 interface through automation of routine tasks, hiding unused tools and features, providing mechanism for minimal system configuration during migration.
  • Integrating the system with a web-based client, that allows as many clients as possible to access reports (maps, shape files).

Further Work

A substantial amount of work was put into the completion the project. It can however still be augmented in the following ways:

  1. Communication module
    Allow GIS client users to electronically place data/report requests from the web site.
  2. User groups
    GIS clients could be configured to allow different classes of users access only the relevant data. This means different users will have different settings and views of the GIS client (web site).
  3. Data upload module
    Currently, data that exists in the database was manually appended. A module can be created that automatically receives batch data, and logically places it the geo-database.
  4. User training
    Basic training in GIS is needed to ensure full productivity of users. Users will also be able to easily figure out the kind of functions needed in future systems.


Though it looked like an insurmountable task initially, the set objectives of the system were successfully met. The accuracy of the reports, however, depends on the accuracy of the input spatial data. The system can be very useful if a large centralized database can be set up. Historical information can be used to make future predictions, and further aid in decision making and management of the Masai Mara.

This system can be further extended to cover other natural resources areas with some modifications.


  1. Fischer, Manfred M., Nijkamp, Peter, Geographic Information System, Spatial Modelling, and Policy Evaluation.
  2. A. B. Cox, Fred Gifford, An overview to GIS, 1999.
  3. K. Chang, Introduction to GIS, 2004
  4. Lyon, McCarthy, Wetland and Environmental Applications of GIS, 1995.
  5. University of Arkansas, “Getting started in GIS”, (2006), Available at: www.libinfo.uark.edu/gis/tutorial.asp
  6. J. Youngman, “Kenyan Wildlife”, (2006), Available at: www.magicalkenya.com
  7. ESRI, “What is GIS”, (2005), Available at: www.gis.com
  8. Kenya Wildlife Services, “Parks and Reserves”, (2005), Available at:www..kws.org
  9. O. R. Odongo, “Application of GIS on Natural Resource Management, A case study of information system for coastal wetland ecosystem”, School of Computing and Informatics-University of Nairobi Masters Project Report, 2004.
  10. O. O. Hesbon, “Crime mapping analysis using a GIS- based system”, School of Computing and Informatics-University of Nairobi Undergraduate Project Report, 2005.
  11. E. G. Anduvate, “Disease Monitoring System”, School of Computing and Informatics-University of Nairobi Undergraduate Project Report, 2005.
  12. M. Melissa, “A GIS for analyzing arid and semi-arid lands”, School of Computing and Informatics-University of Nairobi Undergraduate Project Report, 2005.