The TAITA project: THEMES


The results of the land use change analysis based on aerial photo analysis, a Landsat TM image interpretation from Kilimanjaro, and SPOT XS from Taita Hills are the basis of the study. The comparison of the outputs of the Kilimanjaro and Taita projects aims at improving the understanding of various pathways of land use change and land use intensification, as well as the driving forces behind change in various agricultural systems in the East African highlands.

The results of the studies of Taita will contribute to the understanding of the driving forces and socio-ecologic impacts of land use change in the East African highlands.

To map and analyse changes of land use patterns, an aerial photo interpretation of land use change will be conducted, using aerial photographs from 1955, 1985 and 2003. Classification will be done visually by distinguishing patterns and tones of the surfaces.

Satellite remote sensing data will also be used, thus providing efficient use of pixel-based classifiers and also to study the land use change within the Tsavo plains, as well. Images from 1987, 1992 and 2002 will be used.

The land use change study also provides important information about the changes within the sacred forests. Applying the results of the land use changes in the past, models of future land use change will be made, answering questions like: what will the landscape look like if the current trend continues?


In this study the traditionally protected forests of Taita Hills will be identified and analysed. In Taita Hills, traditionally protected forests are practically the only remaining indigenous forests. They are small but ecologically significant because of their endemic species.

The emphasis is on the human context from the past to this day. According to recent studies there are only 12 indigenous forest patches remaining. Their sizes are similar to those in the neighbouring North Pare Mountains in Tanzania; however, here there can be more than 12 traditionally protected forests in one single village.

One of the specific questions to be studied within this research is the reason for these differences. Are the traditions so different, especially since according to the tradition, people in Pare stem from Taita? Could the reason for the small number of forests in Taita be the local culture, or the destruction or disappearance of the forests?

The size of the traditionally protected forests will be examined and the condition of traditionally protected forests will be studied. Aerial photographs from 1983 and 1993, a multispectral SPOT XS satellite image from 1987, and the digital camera mosaic acquired in 2003 will give us possibilities to analyse and explain the forest change, as well as represent it to the local and science communities.

The research aims at discussing new opportunities of generating an income and creating reasons for people to continue the protection of these forests. If the values and attitudes of local people and communities allow, these forests could e.g. be an interesting addition to the organised guided tours in the Tsavo National Parks, and they could be real income generators for the local communities around the forests.
One of the targets is to analyse the opinions of the local communities, on the forests and their future.

More information :
Jussi Ylhäisi (jussi.ylhaisi(at)
Department of Geography and Institute of Development Studies, University of Helsinki


Land degradation can be defined as a permanent or temporary lowering of the productive capacity of land. Since the productive capacity of land cannot be assessed by any single measure, several indicators of land degradation have been used on land degradation studies. A wide range of these indicators, such as changes in vegetation cover or distribution of gullies, can be assessed through remotely sensed data.

Soil erosion is recognised as one of the major aspects of land degradation. Gullies, as one of the most prominent erosion features, are almost always associated with accelerated erosion and landscape instability. Spatiotemporal distribution of gullies over land surface is controlled by several environmental factors including indirect human actions, such as land use and soil conservation practices.

The objectives of the land degradation studies in Taita are:
1) to develop gully erosion change detection methodologies from multitemporal sets of aerial photographs,
2) to study how gullies are distributed over the Taita Hills area,
3) to investigate the controlling factors affecting gully erosion with multivariate statistical techniques, and
4) to develop methodologies for gully erosion hazard assessments.

The database and the methods in the study consist of interpretation, photogrammetric and GIS techniques applied in multitemporal sets of aerial ortophotographs, field measurements and multivariate statistical analyses.

More information:
Tommi Sirviö (tommi.sirvio(at)
Dept. of Geography, University of Helsinki


It is generally accepted that one of the most important factors that hinders development in the third world is rapid population growth. This, together with continuing poverty and lack of basic needs of acceptable life (e.g. food, clean water, shelter, basic health care, security of tenure) imposes a great challenge for sustainable development. What is even more important, however, is the fact that most of the population growth in the world during the next 15 years will be urban growth, and the vast majority of it will take place in developing countries.

In Kenya, the level of urbanization in 2000 was 33.4 percent and it is estimated to go up to 50.8 percent by 2020, indicating an annual growth rate of 3.76 percent in the next 15 years. Of all urban population in Kenya, 70.7 percent lived in informal settlements, a total of 7.6 million people, which is alarmingly high when compared to the average of developing regions, 43.0 percent. Thus, a major focus in sustainable development and issues related to it in Kenya should be directed towards the urban environment and problems caused by rapid urban growth.

The study area is the township of Voi, located in South-Eastern Kenya, Coast Province, Taita Taveta District, 327 km South-East of Nairobi and 159 km North-West of Mombasa. It is situated at an altitude of approximately 580 m above sea level. In the north and east, it borders to the Tsavo East National Park, in the south to Sagala Hills, and in the west to Voi Sisal Estates. The newest population census from 1999 puts the population of Voi municipality at 33,077 and that of the Voi township at 24,404 residents. It is estimated that some 75% of the total population of Voi town live in informal settlements. Of the total housing units in Voi, 70% are constructed of temporary materials.

The principal aim of this study is to detect changes that have taken place in the built-up environment of the informal settlements in Voi during the time frame of investigation; 1985-2004. The main source of data are black and white and true-colour aerial photographs from 2004, 1993 and 1985.
From each temporal aerial photograph mosaic, a structure mask is extracted using two methods: (1) object-oriented image segmentation and (2) visual interpretation. The classification itself is very simple, consisting of built-up and non-built-up classes only. These masks are then compared using post-classification comparison change detection, in order to highlight the changes in built-up environment (either loss or gain of structures) in time.
This method was chosen because it is not dependent on pixel spectral values, which is essential when the data consists of both black-and-white and true-colour images. Results will be given for the growth of each informal settlement in Voi, from 1985 up to 2004.

The work is still in progress, but preliminary results will be given at the 5th AARSE (African Association of Remote Sensing of the Environment) Conference, held in Nairobi in October 2004. The Master’s thesis will be ready in 2005.

More information:
Pekka Hurskainen (pjhurska(at)
Department of Geography, University of Helsinki

Webmaster -- (C) 2003-2009 -- University of Helsinki