Home Articles A study on nature reserve Geographic Information System in Xishuangbanna

A study on nature reserve Geographic Information System in Xishuangbanna

ACRS 1997

Poster Session 3

A Study on Nature Reserve Geographic Information System in Xishuangbanna

Li Hongga, Cui Weihong
(Institution of Remote Sensing Application, CAS)
P. O. Box 9718, Beiging, 100101, P. R. C.
Li Zhixi
(Institute of Remote Sensing & GIS, Southwest Forestry College)
Kunming, 650224, P. R.C.


The protection of forest, especially tropical rain forest which has great economic and ecological effects, is a major task in the world. In this paper, we offer an approach to protect and manage tropical rain forest in Xishuangbanna nature reserves by using Geographic Information System. The study not only provided the basic spatial database, but also evaluated the qualities of nature reserves and environment e.g. soil erosion and elephant habitat and developed regional plans for the forest fire risk and agroforest. Through the practical application, the GIS priority has been established in the work of the management of nature reserves and villagers’ sustainable development. The results demonstrated the potential for applying GIS for nature reserves and sustainable development as well as the effect of the integration of Remote Sensing, GIS and Sampling.

Keywords :

Geographic Information System, Sustainable Development, Remote Sensing

1. Introduction

The United Nations Conference on Environment and Development (UNCED) which took place in Rio de Janeiro, Brazil, 3-14 June 1992, represented the threshold of a fundamental change in the attitudes towards environment preservation and sustainable development, signed the 21st Century Agenda and Biodiversity Convention, which put forward hat the protection of natural environment and resources is imperative and important to safeguard the basic heritage of mankind and the quality of the earth environment, not only for the present but also for the future. The function of forest, especially the tropical primitive forest, which is regarded as the main component of the earth environment and the valuable natural resource, has been more and more paid attention by the world. It has been learned from painful experience throughout the world’s humid tropics that it is generally disastrous to cut tropical forest for wood or to replace them with monocultures either as crops or as plantations. The result is not only the loss of the precious tropical rainforest and biodiversity, but also the challenge to cultivate and develop in tropics.

This problem occurs in Xishuangbanna, the southern prefecture of the Chinese province Yunnan, N 21.08-22.35, E 99 53-101.80. Five nature reserves have been set up in this mountainous area to preserve the tropical, subtropical forest ecosystems and biodiversity. But increasing population with increasing demands for foods, fiber and others is now threatening these five nature reserves. As a result, the villagers go deeper and deeper into the primitive tropical forest. From 1965 to 1988, forest coverage (including primitive forest, rubber forest and so on) had decreased from 46.46% to 33.72%, with an average of 0.55% annually. In the 1995, the primitive forest coverage lowered the rate of 27%.

Although a lot of urgent and important protecting work needs to be done in Xishuangbanna Nature Reserve, the establishment of the spatial management information system is put to the priority. This priority become to base for developing short and long term financial and personnel staffing plans for the protection of tropical forest. Remote, as tools of the information management and analysis technique, have been applied for evaluating and protecting natural sources and environment ; see for instance Y. Jim, Lee (1987) using Landsat TM to analyze Cygnets habitat along the coast of British Columbia, and R. Aspinall (1992) using GIS and Digital Elevation Model to map the Chequered Skipper and Curlew distribution and migration in Scotland. Furthermore, Youngynt Trisurat (1992) applied GIS to manage nature reserves in Thailand.

In This paper, for the sake of needs of the preservation and management of the tropical forest in the natural resources, at first, we complete the spatial database through the integrated way of Remote Sensing, Geographic Information System and Sampling. Secondly, based on the spatial database, some spatial models have been systematically built in special on synthetic studies. Finally, due to this project’s running long in Xishuangbanna, we not only focus on the protection of the tropical ecosystem and biota, but also study a way to harmonize the relationship between the local villager’s socio-economic development and the protection of the tropical forest in nature reserves.

2. Study Area

In this study, the study area mainly lies in Mengyang Nature Reserve, the largest one of the five reserves in Xishuangbanna, between the Xiaohei stream and the Mekong River. Its total area is 118,410 ha and the elevation varies from 550 metes to 2250 meters. The forest vegetation types mainly are comprised of seasonal rain forest, mountain rain forest, monsoon forest and evergreen broad-leaf forest, etc. At the same time, some applications have also been implemented in the other four nature reserves such as Menglun, Mengla, Mankao, Shangyong.

ACRS 1997

Poster Session 3

A Study on Nature Reserve Geographic Information System in Xishuangbanna

3. Method

The framework includes the establishment of the geographic information database, the evaluation of the characteristics of natural resources and environment, and the regional plan and design or the sustainable development for the natural resources management and local village’s socio-economy.

3.1 The establishment of geographic information database.

Before the geographic information system is set up, a realizable system design and analysis has been completed with the consultation of the World Wide Found for Nature’s experts, the ESRI’s experts, the Xishuangbanna Nature Reserve Management Bureau’s officer and the some local villagers. The special database dictionary was compiled in 1994 at Redland, California, USA. Due to the natural resources surrounded by the complicate natural and social environment, the database includes twelve layers topography, hydrography, forest vegetation, traffic net, administrate boundary, soil, humid, precipitation, temperature, slop, aspect, landuse, village situation including population, nationality, income, education health, cultivated model and so on. Among them, the topography is transferred to Digital Elevation Model and the aspect and the slope can be derived from DEM. The scale of main layers is 1:50,000. The entire database has been completed in the three years by using ARC/INFO version 7.02

Because of the scare of the special forest data, at the same time, considering the significance of vegetation, obtaining vegetation and landuse data including the dynamic changes is the core object. The vegetation and landuse data mainly come from the interpretation on the panchromatic aerophotos with the scale of 1:20,000 taken in March 1989, and 28-May 1992 respectively, Spot XS data recorded on 16-Feb. 1988, etc. After image processing such as contract enhancement and synthesis, considering the TM234 susceptibility to vegetation type and partly eliminating hill shadows, we classify it by means of Maximum Likelihood. Meanwhile, the rules of the ecological law and the vertical distribution are also applied for the classification. Finally, we get 11 types such as forest, bamboo, tea, rubber woods, shrub, meadow, farmland and so on with 79% precision at the reliability 95%.

3.2. The evaluation of the characteristics of natural resources and environment.

Based on spatial database, we stared to quantitatively assess the characteristics of environment and sources in the nature reserves step by step. The content of assessment is very wide. In this paper, two typical application are reviewed.

Soil erosion is the crucial factor to influence the reforestory and the form of agriculture in tropics. Red soil, the main soil type distributed in natural resources, is not suited for intensive agriculture, even though it is supported under natural conditions of a rich and diverse plant cover. The steep and mountainous lands even accentuate these problems, as clearcutting on hillsides lead to erosion. Soil is too precious to let it be eroded away without trying to hold it in place. GIS can play a major role in the curtailment of erosion.

Two series of soil erosion have been developed. One is used for villages’ erosion hazard; the other is for entire nature reserve’s risk of soil loss. Both are being derived from the Universal Soil Loss Equation, in order to keep consistency.

Soil loss by water erosion can be expressed by the empirical equation (Wischmeier and Smith, 1978):

A= R*K*L*S*C*P

where A= computed soil loss per unit area, expressed in the units selected for K and for the period selected for R;

R = rainfall and runoff factor (N/hr);
K = soil erodibility factor (kg hr/N m2)
L = slope-length factor;
S = slope-steepness factor;
C = cover and management facor;
and P = support practice factor.

Because of the limited data that were available, R and K have been estimated by the approximate method proposed by Arnoldus (1980) and K. Romkens et. al (1988), respectively. Topographic factor L and S can be derived from the DEM. C is the ratio of soil loss from the cropped land under specified conditions to the corresponding loss from clean-tilled, bare land. Cropland including paddy, dry land, fallow, tea, rubber tree, brushes, grass, bamboo, forest including rain forest, mountain rain forest, monsoon forest, evergreen broadleaf forest etc. have been estimated by field sampling and approximate method proposed by Wischemeier & Smiths. In view of the only support practice in Xishuangbanna planting rubber and tea on the contour, P varies between 0.6 and 0.9 according to the gradient of the slope. The beginning of rainy reason i.e. May and the mean level of year have been computed, respectively. The maximum and the mean of the soil erosion of May are 329.295 t/ha, 1.595 t/ha with the Standard Deviation of 9.093 respectively. The maximum and the mean of every year are 3140.85t/ha, 15.210t/ha a with the SD of 86.712. The whole reserve has been divided into 4 erosion classes according to erosion hazard for year.

The Asian Elephant habitat is obtained by means of field sampling and Multivariate Regression. In field survey, elephant trails and relevant ecological factor, biological factor and human activity have been marked. The final results can be gotten by the following formula:

Y= b0+b1X1+b2X2+b3X3

Where, Y = the frequency of elephants activity (unit is trails per ha);

b0,b1,b2,b3,b4,b5,b6,b7,b8 are regression coefficients;

and X1,X2,X3,X4,X5,X6,X7,X8,X9
represent correlative factors, i.e. vegetation, crown density, distance from water, slope, aspect, location, elevation and human activity. These can be obtained from spatial database. And the precision is obtained at 82% under the reliability of 95%.

3.3 The regional plan for natural resources management and local village socio-economy.

Based on the database and environment assessment, according to the practical needs, the detail plan and design have been applied. Two priorities have been considered, i.e. wildfire hazard and agroforestry.

According to the historic record in nature reserves such as to location, area, intensity and so on, we find that the main forest fire risk comes from the slash and burn by local villagers, and the intensity results from the inflammability of woods an dryness. So the distance form road and villages, the aspect and the inflammability of vegetation and so on which are deduced from database are clustered after they have been quantitatively normalized. A reliable classification with five wildfire hazards has been plotted. It is easy to plot the patrol lines and the fire break for the Nature Reserve Management Bureau.

There is an urgent need to apply improved agroforestry systems and ecological villages fro China’s deteriorating farmlands where fertility loss exceeds crop yields. It is particularly urgent for the watersheds of the Lancang Jiang(Mekong River) in Xishuangbanna where the Farming of slash and burn is practiced on a diminishing suitable landbase.

There is an urgent need to apply improved agroforestry systems and ecological villages for China’s deteriorating farmlands where fertility loss exceeds crop yields. It is particularly urgent for the watersheds of the Lancang jiang(Mekong River) in Xishuangbanna where the farming of slash and burn is practiced on a diminishing suitable landbase.

A great quantity of villages are situated around or in the nature reserves. In order to prevent people from cutting woods, hunting wildlife, destroying the nature reserves, the aims of agroforestry and ecological villages include improving the forest coverage around ecological villages, increasing the villager’s incomes only in limited lands, and finally raising the fertility of soils. At the same time, some sketches have been finished to move villages out of he core zone of nature reserves. The villager’s GIS for agroforestry has been implemented in Zhongtianba and Bajiazhai with the scale of 5,000. Zhongtianba ecological village has been built under the support of WWF. By means of planting nitrogen fixed woods, herb and comprehensive management, the villagers improve the incomes. Meanwhile, environment is protected.

ACRS 1997

Poster Session 3

A Study on Nature Reserve Geographic Information System in Xishuangbanna

4. Results and Discussion

4.1 Results

Firstly, we have complete the whole database of Mengyang, the part database of the other fours. The database has been provided to the Management Bureau of Xishungbanna Nature Reserve and the other relevant organizations. These become the base for developing short and long term studies.

Secondly, based on the database, the evaluations of resource and environment in nature reserves have been developed step by step e.g. the soil erosion and the endangered Asian Elephant habitat. These models with the practical parameters for assessing and protecting the Xishuangbanna’s biodiversity are essential. Some areas will be marked to special use patterns which are used for soil erosion and wildlife use patterns, that can be plotted and then coordinated with the management of nature reserves to help reduce the risk.

Thirdly, based on the database, the regional plans for the management of nature resources and local villages socio-economy have been implemented. Cooperated with experts from Nature Reserve Management Bureau, by the education and promulgation of the agroforest, plans and designs have been widespread accepted by the officer and the local villagers. It is a efficient and effective way to adjust the relationship between the villagers and the Nature Reserve Management Bureau. Furthermore, it will pay a important role in sustainable development for nature reserves and villages.

4.2 Discussion

In this study, Forest Sampling, Remote Sensing and Geographic Information System are effectually and efficiently integrated. In the scare data area, it is a reliable way to obtain data sources, to identify the practical parameters of the assessment models and to adjust the regional plans and designs. According to Remote Sensing and Permanent Plot Techniques for World Forest Monitoring Conference and International Guidelines for Forest Monitoring, the integration of FS, RS and GIS is the developing tendency for natural resources monitoring. The paper may be regarded as a test of the integration of FS, RS and GIS. Meanwhile, the activities in Xishuangbanna provide and excellent opportunity to expand the working knowledge of GIS in the protection of nature reserves and the sustainable development.


  • R. G. Best. 1983, Remote Sensing Approaches for Wildlife management. Proceeding of the RNRF Symposium on the Application of Remote Sensing to Resource Management, Seattle, Washington, p55-96.

  • Richard Aspinall, 1992. Spatial Pattern Analysis tools in GIS for Environment Planning. Proceeding of the IUFRO Centennial meeting, Berlin, p.223-233.
  • Youngyut Trisurat, 1992. Application of GIS in Protected Area Management Remote Sensing and Permanent Plot Technique for World Forest Monitoring. Proceeding of the IUFRO S4.02.05 Pattaya, Thailand. P215-225.
  • R. H. Cydelund Songkram. Risto Paivinen Thammincha, 1992. Remote Sensing and Permenemt Plot Techniques for World Forest Monitoring Proceeding in the IUFRO s4.02.05, Pattaya, Thailand.