--> Proceedings --> GISDECO

GIS and natural resource management : prospect and problems Kanyati Communal Lands, Zimbabwe

H. Huizing1 , A.G. Toxopeus 1 , E. Dopheide 1 , and B.M. Kariaga 2
1) International Institute for Geo-Information Science and Earth Observation (ITC), PO Box 6, 7500

Enschede, The Netherlands
2) Monash University, Private Bag X60, Ruimsig, Roodepoort, South Africa

Information on the variability and distribution of natural resources and natural resource problems is needed to support decisions in natural resource management (NRM). However, despite availability of this information, its actual use in decision-making processes at the local level is often disappointing. The use of GIS might reduce this problem. A GIS makes it easier to visualize the spatial diversity of resources; to analyze and integrate data sets; and to assess impacts of interventions, thereby enhancing the transparency of decisions regarding natural resource use.

In this paper, potentials and limitations of introducing GIS are explored for the Kanyati Communal Lands, Nyaminyami Rural District, Zimbabwe. Rural District Councils in Zimbabwe have the legal power to manage natural resources of communal lands for the benefit of the people. Detailed natural resource maps and reports covering soils, vegetation, land use, land capability and population were available, but not utilized to support decisions on NRM. It was expected that GIS could contribute to a more effective use of the available information.

The exploration started with a problem identification workshop with local stakeholders. They identified as main problems: lack of water for livestock and domestic use; inadequate land and facilities for crop and livestock production; and damage of crops by wildlife. Subsequently, utilizing available natural resource and socio-economic data, staff of ITC and the University of Zimbabwe prepared a GIS-based information system and developed the following GIS applications based on the identified problems:
  • Comparison of villages using “indicators” related to natural resource use
  • Spatial variation of overgrazing among villages
  • Spatial variation of access to water for domestic use and livestock (current and proposed)
  • Spatial distribution of crop damage by wildlife
  • Comparison of land use plans with the actual land use
During the development of the information system, ITC and UZ staff regularly consulted staff of the Rural District Council. During the consultations, it became clear that leading District Council staff lacked insight of how an information system could support decisions with regard to the use of natural resources in situations where priorities had to be set because of scarce resources. Neither the availability of the minimum hardware and software, nor a short training course and a demonstration of applications of the information system did substantially increase the interests. Middle-level technical staff of the Council and its supporting technical agencies, however, clearly became aware of the benefits of the system. It allowed them to visualize and localize problems and the impacts of possible interventions.
  • The lessons learnt are clear. A GIS-based information system has the potential to make decisions with respect to natural resources more efficient and transparent. Lack of training in GIS use and limited software and hardware resources are certainly hampering the effective use of GIS. However, lack of interest and appreciation of the role of existing information among key stakeholders can be the most limiting factor in establishing an operational GIS-based information system as an integral part of an organization’s planning activities. Although this finding is in line with other studies on GIS implementation in developing countries (e.g. Ramasubramanian, 1999), it is still often overlooked in the practice of GIS development.

    1. Introduction
    Natural resources play a critical role in the welfare of developing countries. With growing populations and increasing pressures on natural resources, efforts of people and institutions to improve the efficiency and sustainability of resource use are increasing. For such efforts to be effective, knowledge and information on the resources and their distribution in space and time are essential (Young, 1998; FAO and UNEP, 1999). However, experiences in many developing countries indicate that, even when information is available, this information is often not used to support decisions on their use and management. Reasons for the under-utilization of resource information are various (Dalal-Clayton and Dent, 1993):
    • Information is sectoral, gathered by different institutions without coordination, not integrated, and sometimes incomplete or out of date
    • Information is in scattered reports and maps in different offices and not easily accessible, therefore
    • Information is presented in terms that are not easily comprehensible by resource managers
    • Information collection is not geared towards solving urgent resource problems or conflicts
    • Information often does not reflect the knowledge and information needs of the actual resource users
    • Local institutions do not have sufficient capabilities to understand and use the available information for decision making
    The use of a geographic information system (GIS) can improve the efficient use of available information on natural resources and their use. Possible contributions of GIS to a more effective use of natural resource information include:
    • A better visualization of the spatial diversity of resources, thereby improving the transparency of decisions regarding alternative uses of natural resource and the allocation of scarce funds to areas most in need
    • Improved possibilities to analyze and integrate data from different sources
    • The possibility to assess impacts of alternative interventions, e.g. the spatial distribution of benefits of an intervention in relation to the needs of local communities
    Minimum requirements for a GIS-based information system to become effective as a tool that supports natural resource management in existing institutions are: (i) a personal computer, colour printer and digitizer; (ii) user-friendly GIS software, a word processor and spreadsheet programme; (iii) trained staff, knowledgeable with the use of GIS software; and (iv) awareness that a GIS-based information system will improve the institution’s natural resource management capabilities

    2. The Kanyati Communal Lands
    All communal land in Zimbabwe is state land. Rural District Councils have the legal power to manage communal resources for the benefits of the people. The Kanyati Communal Lands belong to the Nyaminyami Rural District and are located in North Zimbabwe. Resources like grazing land and woodland in Kanyati are held under communal tenure. Arable land and residential land in the communal areas, on the other hand, are held under a traditional freehold tenure that gives exclusive rights of use to families. These rights are transferable within a family (Sibanda, 2001). Kanyati consists of two wards and ten villages. Each village has a Village Development Committee (VIDCO) headed by a chairman. Ward councillors represent the VIDCO chairmen in the Nyaminyami Rural District Council. Several ministries and government departments have offices or representatives at the headquarters of the Council.

    The study area lies 800 m to 1,000 m above sea level and is mainly underlain by metamorphic and igneous rocks. Soils vary in depth, but are mostly shallow and sandy. The climate is semi-arid with an annual rainfall of between 600 and 900 mm and a rainy season of 3 to 5 months. In the early 1980’s, few people lived in the area. Eradication of tsetse in the mid-1980’s caused a great influx of people from other parts of Zimbabwe. In 1992, about 5,600 persons lived in the area. Their main occupation is crop growing, mainly maize and cotton, and animal husbandry (mainly cattle). The main source of fuel is firewood obtained from the remaining woodlands. In villages close to wildlife areas, damage of crops by wildlife is common.

    3. Objectives and approach
    In this paper, potentials and limitations of introducing GIS are explored for the Kanyati Communal Lands. The paper is based on joint work of staff of the Department of Geography and Environmental Science, University of Zimbabwe, and ITC in The Netherlands. The objectives of the study are:
    1. To develop a GIS-based information system for the study area and demonstrate applications of its use for the analysis of various natural resource management problems
    2. To assess potentials and limitations to operationalize the information system in the local planning procedures of the Nyaminyami Rural District Council (NRDC), the institution responsible for the management of natural resources in the Kanyati area
    The study approach included the following steps:
    1. Identification and structuring of problems by representatives of local communities.
    2. Development of a digital spatial database for the area.
    3. Organization of a one-week GIS training course for staff of the NRDC at the University of Zimbabwe to familiarize them with GIS and its applications.
    4. Development of sample applications of the information system to demonstrate its potentials to assess and find solutions for natural resource related problems identified by the local communities.
    5. Further training of staff of the NRDC and affiliated local agencies after installation of hardware, software and the database at the District’s headquarters
    Local problems were identified during a one-day “problem analysis and structuring” workshop in Kanyati in which one ward councillor, all VIDCO chairmen and some resident staff of specialized agencies participated. Table 1 shows the main output of the workshop. The table indicates that natural resource related problems (water, crops/livestock/food and wildlife) accounted for about 60% of the perceived problems.

    Table 1. Ranking of problems identified during workshop in Kanyati.

    Water (lack of wells, water for dip tanks and irrigation, dams, boreholes)
    Roads/transport/communication problems
    Crops/livestock/food related problems (e.g. lack of cropland,
    Food security, draught animals, soil erosion)
    Wildlife incidents
    3 rd
    % of total problems

    4. The GIS database
    For the ten villages of the Kanyati Communal Lands, natural resource maps and reports (ARDA, 1992) covering themes like soils, vegetation, land use, land capability, proposed land use, location of boreholes and settlements, population characteristics, village and ward boundaries and village-based land use plans are available at a scale of 1:12,500. Until now, little use was made of these maps and reports. In addition, information is available on contours, drainage and infrastructure from 1:50,000 topographic maps. Data on wildlife and crop damage exist in different government departments and WWF. Rainfall data is available for several rainfall stations in the study area and its neighbourhood.

    The main data source for the GIS database is ARDA data and maps of 1992. ARDA maps are hand-drawn, not very accurate and include geo-referencing mistakes that had to be corrected during the digitizing of the maps. Satellite data (Landsat TM images covering the period 1973 to 1998) and aerial photos of the 1970’s, 1980’s and late 1990’s and the results of several MSc studies by students of ITC and the University of Zimbabwe contributed to the completion of the database.

    5. GIS applications for natural resource management
    Lack of water, transport and communication, and problems related to crops, livestock and wildlife are the main natural resource problems perceived in the study area (Table 1). A GIS database should be able to help in the analysis of these problems and to contribute to the setting of priorities for spending scarce financial resources to reduce the problems (Ceccarelli, 1997). The identified problems were the basis for the development of sample applications that were used to demonstrate the potentials of a GIS-based information system. They include:
    1. Comparison of villages on the basis of “indicators” related to natural resource management
    2. Spatial variability of access to water for domestic use (current and proposed)
    3. Spatial variability of access to water for livestock (current and proposed)
    4. Spatial variation of overgrazing among villages
    5. Spatial distribution of crop damage by wildlife
    6. Comparison of proposed land use plans of 1992 with the actual land use
    Application A. Comparison of villages using ‘indicators’ related to natural resource management
    The study area contains ten villages (Figure 1). Table 2 shows data related to natural resources in the villages and their use (ARDA, 1992). The table shows that there are considerable differences between villages with respect to cropped area, number of livestock per household and grazing area per livestock unit.

    Figure 1. The ten villages of the Kanyati Communal Lands

    Table 2. Village data and natural resource use, Kanyati Communal Lands

    Nr. of hh
    Nr. of settlements
    % of hhctivatingLand
    Total cropped areaha)
    Arable land (ha)
    % cropland/village
    Cropped area (ha/hh)
    Maize (ha/hh)
    Cotton (ha/hh)
    LUs (nr./hh
    Grazing land (ha/LU)

    Notes: hh = household; LU = livestock unit; “arable land” (ha) is based on land capability assessment and mainly includes land marginal for crop growing (ARDA, 1992)

    Application B. Access to water for domestic use; Water sources for human consumption are mainly boreholes and wells, as most rivers and reservoirs are without water during the dry season. Locally, permanent pools with standing water occur during the dry season, but water from these pools is often too dirty for drinking purposes. The preferred maximum distance from settlements to water collecting places is 500 meter with a maximum of 2,000 m. Important information for the planning of new wells or boreholes is how many households have access to drinking water at distances within 500 m, 500-2,000 m and more than 2,000 m. This information can be used to plan new boreholes or to evaluate to which extent plans for new boreholes will be effective in solving the problem of access to water for domestic use. For this case, maps showing (i) settlements; (ii) boreholes/wells; and (iii) proposed boreholes/wells are used as input maps. From these maps, a new map is created that shows settlements and their distance from existing and proposed water collecting points (Figure 2).

    Availability of dinking water is a major problem in the study area (Table 1). Table 3 confirms this. Only 19 percent of all settlements in the area lie within 500 m of a well/borehole. In Chitete village, 39 percent of the settlements are located 2 to 5 km away from a well/borehole. The location of the proposed new boreholes contributes little to improvement of this situation in the most problematic villages (Chitete, Hurenje and Nyajena). Figure 2 and Table 3 show the variation of the problem of access to domestic water and the limited effect of the proposed new boreholes and wells in solving the problem.

    Figure 2. Distance between existing and proposed drinking water (collecting) places (boreholes and wells) and settlements.

    Table 3. Settlements within villages and distance from well/boreholes.

    Village Number of  settlements   < 500 m    500 m – 1 km   1 – 2 km    2 – 5 km
    Chebere   77   30%   43%    23%    4%
    Chitete    142    2%    15%    44%    39%
    Hurenje    64   30%   13%    55%    3%
    Hwamira    42    7%    45%    36%    12%
    Kadziru    39    10%    31%    46%    13%
    Kanyati    74    19%    42%    35%    4%
    Kemengwandi   50    34%    32%    22%    12%
    Makande   71    3%    44%    54%    0%
    Nyadara    60    52%    33%   15%   0%
    Nyajena    62    16%   18% 65% 2%

    Application C. Access to water for livestock
    Livestock in the study area grazes on communal land, usually up-hill in summer and in lower areas and arable fields in winter (Agritex, 1996; Nguru, 1998). Livestock is herded by households and sometimes by groups of households. During nights, livestock is mainly kept in fenced areas (“kraals”) near residential sites. In daytime, livestock has to walk to watering places. The water sources in the dry season are mainly boreholes, pools and wells, as most rivers and small reservoirs dry up. Preferred distances to boreholes, wells and pools are less than 2 km; the maximum distance is 5 km. ARDA (1992) developed a plan with locations of new boreholes and wells. The aim of this case is to find out whether the proposed new boreholes/wells are useful and have a correct location. Another aim is to determine alternative locations for boreholes/wells in each village.

    Figure 3 shows distances of grazing areas from existing watering points for livestock. The location of the proposed new boreholes indicates whether or not these will be effective in reducing access to water for livestock.

    Figure 3. Distance grazing areas and drinking places for livestock in dry season.

    Availability of water is a problem for local communities. It is clear from Figure 3 that the proposed boreholes in are roughly in proper locations for livestock watering, although they could be spread a bit more. In some villages (central Makande and east Nyadara), the proposed boreholes are not useful at all for livestock watering because watering points already exist at proper locations. Several potential grazing areas will remain too far away from water after construction of the proposed boreholes.

    Application D. Livestock carrying capacity and current stocking densities
    This application shows the variation among the villages regarding amount of forage available for livestock and actual stocking densities. The numbers of livestock are expressed in Tropical Livestock Unit Equivalents (LUs1) for this purpose. By comparing the potential LU numbers (carrying capacity) of each village with the actual LU numbers, the LU balance per village was estimated. When the potential LUs are lower than the actual LUs, the village is expected to be overgrazed (Table 4).

    The total number of livestock (expressed in LUs) was calculated for each village based on (i) actual livestock numbers per species per village in 1992 and (ii) the conversion values shown in footnote 1. Next, the carrying capacity (potential livestock numbers LUs per village) was calculated based on field staff estimates of the carrying capacity of mapping units shown on the available vegetation maps produced by ARDA (Figure 4).
    Conversion values of livestock species kept in the study area:
    Cattle: bulls (LU = 0.75); oxen (LU = 0.94); cows (LU = 0.64); steers (LU = 0.61); heifers (LU = 0.55); calves (LU = 0.28)
    Other: goats (LU = 0.07); sheep (LU = 0.07); donkey (LU = 0.40).

    Figure 4. Carrying capacity for livestock (ha/LU) of different vegetation units

    Table 4. Comparison of livestock carrying capacity, number of livestock in 1992 and LU-balance which shows whether village land is likely to be overstocked (negative values) or understocked.

     Village  Carrying capacity (LUs) based on vegetation cover  Actual LUs in 1992  LU balance
    Chebere   257  146   111
    Chitete  350  284   66
    Hurenje  158  179   -21
    Hwamira  330  97   233
    Kadziru  386   273  113
    Kanyati  256   331  -75
    Kemangwandi  423  273   150
    Makande  138  333   -195
    Nyadara  200  314   -114
    Nyajena  362  192   170

    Table 4 indicates that Makande, Nyadara and Kanyati villages are overstocked. Most of the other villages still have surplus of forage, when considering the number of livestock in relation to potentially available forage.

    Case E. Crop damage by wildlife
    Wildlife incidents in the ten villages have been recorded by the Campfire Association from 1993 up till 1998 (Mwiya, 1998) and by Agritex/WWF. Data from the two sources (Table 5) are not the same but of the same orders of magnitude. Table 5 shows that the most northern villages that border conservation areas suffer most.

    Table 5. Number of incidents with elephants and other wildlife species (column “other”) and total number of wildlife incidents from 1993 to 1998 based on Agritex/WWF data. The column Campfire shows total wildlife incidents over the same period, recorded by Campfire.

    Village Crop damage by wildlife (1993-98)
    Agritex/WWF data Campfire Data
    Elephants Other tota
    Chebere  59  3  62   46
    Chitete  0  10  10   6
    Hurenje  279  14  293   293
    Hwamira  19  14  33   24
    Kadziru  0  0   0
    Kanyati  5  0  5   0
    Kemangwandi  0  0  0   0
    Makande  0   0  0
    Nyadara  1  0  1   0
    Nyajena  61  2  63   64

    Application F. Land use plan of 1992 and actual land use in 1992 and 1999
    A final application was developed to compare (i) the land use of 1992, (ii) the land use as proposed in a plan of 1992 and (iii) the actual situation seven years later. This was done for Makande village for which information on the 1999 land use situation was available (Sadiki, 2000). In the land use plan, land allocated to grazing is divided in paddocks. In 1999, four large paddocks had been fenced. These paddocks are used for grazing in the wet season. Figure 5 shows that cropland expanded at the expense of grazing land from 1992 to 1999, but not exactly according to the plan of 1992. It is likely that the fencing of grazing land reduced further expansion of cropland into grazing land. Residential areas were scattered in 1999, probably similar to the situation in 1992, while the land use plan proposed consolidation of residential areas.

    6. Discussion
    The applications shown in the previous sections show that a GIS-based information system makes it possible to visualize and make transparent the variation of the magnitude of natural resource management problems between spatial units, in this case villages. Ranking the villages on the basis of some of the identified problems shows, for example, that the village of Nyajena is relatively worse of than other villages in terms of the problems of access to water for domestic use; overgrazing; and crop damage by wildlife. Although the selected indicators do not give a comprehensive overview of the problems, a more complete set of indicators has the potential to support decisions on the allocation of resources between the villages. No information was obtained on the actual spending of public resources in the various villages.

    In addition to support decisions on spatial resource allocation, the developed GIS applications have the potential to monitor and evaluate the effectiveness and the implementation of various natural resource management plans, e.g. planned locations of boreholes for domestic use as well a for livestock; land use plans. Finally, the applications show that information from different sources, e.g. location of residential areas and locations of water points, can be related and integrated.

    Figure 5. Land cover/use in 1992, proposed land cover/use (land use plan of 1992) and actual land use in 1999, Makande village, Kanyati Communal Lands

    A well-structured database makes it possible to provide answers to many questions with regard to land resources and their use and, thereby, is useful tool for decision making with regard to the use of natural resources. Such a tool can increase the efficiency of institutions responsible for land use planning and natural resource management.

    However, reviewing the development of the GIS applications in this real-world situation and evaluating the response of the local institutions, particularly the Rural District Council, the reported case suggests that the actual use and success of GIS for natural resource management in developing countries is certainly not self-evident. There may be limitations and/or conditions that are not optimal for the implementation and use of a GIS-based information system. As far as implementation issues are concerned a number of obstacles will have to be overcome. In our case, we experienced that some essential data were of low quality, out of date or lacking, funds for new data acquisition was not be available and trained staff for implementing and updating the GIS database were not present. Also the initial awareness of the potential of a GIS information system was virtually absent. But even when these barriers to GIS implementation were overcome through the acceptance of lower data quality, the provision of the minimum hardware and software, a short training course and demonstration of the developed applications, the response among leading local staff remained low. Middle-level technical local staff clearly became aware of the benefits of the system, but interest and willingness of higher staff to actual use and develop the GIS applications further for their own use was lacking.

    This lack of interest among higher staff can be partially explained by a still limited awareness of the potential of a GIS-information system for local level decision making and a lack of insight of how an information system could support decisions with regard to the use of natural resources. But the reluctance to actually start using the information system might also be inherent in the prevailing governance settings. Information on the magnitude of the problems among the various villages, for example, is possibly not the most required for actual public decision making in the area. The potential of an information system to make public decisions more efficient and transparent may in some governmental settings be less desired than is often assumed from an outsiders' point of view.

    7. Conclusions
    Applications of a GIS-based information system shown in this paper have the potential to make decisions with respect to natural resources more efficient and transparent. Although the development of the GIS information-system for the Nyaminyami Rural District Council was, on purpose, low profile, the lessons learnt are clear. A number of barriers that hamper GIS implementation, like lack of training in GIS and limited software and hardware resources in local institutions need to be overcome. However, lack of interest among key decision-makers to understand and use available information appeared to be the most limiting factor in establishing an operational GIS-based information system as an integral part of the natural resource management planning activities at the local level. The low response of the local institutions reveals and confirms the need to pay sufficient attention to issues of utilisation of GIS in actual government in addition to issues of implementation.

    • AGRITEX, undated. Lecture Notes for Land Use Planning Course I. AGRITEX Planning Branch, Harare. 63 pp.
    • AGRITEX, 1996. Grazing schemes by-laws. AGRITEX Planning Branch, Harare.
    • ARDA, 1991. Towards a sustainable farming system for the Kanyati and Gatshe Gatshe Communal Lands. In: The International
    • Evaluation of the Kanyati and Gatshe Gatsche Land Use Project, 22-25 April, 1991. The Project Management Unit of ADA,
      Kariba. 29 pp.
    • ARDA, 1992. Kanyati Land Use Project. Plan of operations for the 10 villages composing the Kanyati Communal Lands.
    • CAMPFIRE, 1998. Kanyati and Gatshe Gatshe CAMPFIRE Problem Animal Control Records (1993-1998).
    • Cecarrelli, T., 1997. Towards a Planning Support System for Communal Areas in the Zambesi Valley, Zimbabwe. A multi criteria evaluation linking farm household analysis, land evaluation and GIS. ITC publication No. 51, ITC, Enschede. 282 pp.
    • Child, G., 1995. Wildlife and People: the Zimbabwean Success. Published by: Wisdom Foundation, Harare. Pp 147-173.
    • Dalal-Clayton, B. and D. Dent, 1993. Surveys, plans and people. A review of Land Resource Information and its use in
    • developing countries. IIED, Environmental Planning Issues No. 2. FAO and UNEP, 1999. The future of our land: facing the challenge. Guidelines for integrated planning for sustainable management of land resources.
    • Murwira, A. and D. Mutendera, 1997. Operationalising GIS: The development of the Integrated Resource Information System (IRIS) Graphical User Interface (GUI) in the Department of Natural Resources (DNR), Zimbabwe. Proc. AfricaGIS97 Conference, 23-28 June 1997.
    • Mwiya, S., 1998. Analysis of crop damage by wild animals in Zimbabwe communal lands: A case study of Kanyati Communal Land. Unpublished report, ITC, Enschede, The Netherlands.
    • Nguru, P.M., 1998. Stocking densities and the status of grazing lands in Hurungwe Communal Lands, Zimbabwe. Unpublished report, ITC, Enschede, The Netherlands.
    • Ramasubramanian, L. (1999). GIS Implementation in Developing Countries: Learning from Organisational Theory and Reflective Practice in Transaction in GIS, 3(4): 359-380
    • Sadiki, P. J., 2000. Driving forces underlying land use conversions in land use planning. A case study of resources utilization in communal lands of Zimbabwe. MSc thesis. ITC, Enschede, The Netherlands.
    • Sibanda, B., 2001. Wildlife and communities at the crossroads: Is Zimbabwe’s Campfire the way forward? SAPES Books, Harare, Zimbabwe.
    • Stour Consultants (Pvt) Ltd, 2000. District hydrogeological report, Kariba (Nyaminyami) District. Report prepared for Department of Water Development, Ministry of Rural Resources, Zimbabwe.
    • Young A., 1998. Land resources, now and for the future. Cambridge University Press.
    • Vincent, V. (Ed.), 1961. An agricultural survey of Southern Rhodesia. Part I: Agro-ecological survey. Part II: Agro-economic survey. Federal Ministry of Agriculture, Salisbury.
  • © All rights reserved.