Applications of the small-scale and large-scale SOTER databases in Hainan, China

Size: px

Start display at page:

Download "Applications of the small-scale and large-scale SOTER databases in Hainan, China"

Lynne Lewis
5 years ago
Views:

1 Symposium no. 44 Paper no. 655 Presentation: oral Applications of the small-scale and large-scale SOTER databases in Hainan, China ZHANG Gan-Lin (1), GONG Zi-Tong (1), ZHANG Xue-lei (1), ZHAO Yu-guo (1), QI Zhiping (2), VAN ENGELEN Vincent (3) and MANTEL Stephan (3) (1) Institute of Soil Science, Chinese Academy of Sciences, Nanjing , China (2) Chinese Academy of Tropical Agricultural Science, Hainan , China (3) International Soil Reference and Information Centre, Wageningen, The Netherlands Abstract SOTER methodology was set up upon the demand of updating global and national soil and soil environmental information by combining spatial information of soil-- terrain units and their attribute data. The system was tested in the South America and has been used extensively in Kenya, Central Europe and other regions and has been proved as a useful tool in regional land management. When its traditional industry of rubber production meets a falling profit, and meanwhile more competent products-tropical fruits are finding their increasing share in domestic and international markets, the local government of Hainan, China, is looking for an effective tool in order to plan the provincial land according to physical land quality. Medium(1/250,000) to large scale(1/50,000) SOTER was adopted for the sustainable land management plan in Hainan, China, oriented for the agricultural production. Two spatial scales were adopted in which medium scale SOTER was designed for the whole Hainan province (HaiSOTER), an area of 34,000 square kilometers, while large-scale SOTER was for four typical "windows" of the island (WinSOTER). Both SOTER databases contained standard attribute data and georeferenced ploygon data as well as annexed climate and land use data. Locally-tailored application models, including land quality evaluation models based on ALES(Automated Land Evaluation System) and the model for soil water erosion assessment based on USLE principles were established. Physical suitability for main tropical crops, such as rubber tree, banana, mango, coffee, in different input and technology levels were assessed. Soil water erosion in current land use and different scenarios were estimated. HaiSOTER provided overall guidance for the provincial plan of main kinds of tropical cash crops and the county SOTER helped to set regional demonstrations of crop production. Comprehensive use of the two SOTER databases should be very helpful for the sustainable land management of the tropical island. Keywords: SOTER, land quality, land management, Hainan, China Introduction Policy-makers, resource managers and the scientific community at large have repeatedly expressed the need for ready access to soil and terrain resources through georeferenced databases, in order to make assessments of the productive capacity of soils, 655-1

2 to have a better understanding about the risks and rates of soil degradation, and to better quantify processes of global change (Oldeman, 1994). SOTER(Soil and Terrain Digital Database) is a relational database handling system which can store detailed information on natural resources in such a way that these data can be readily accessed, combined and analyzed from the point of view of potential use, in relation to food requirements, environmental impact and conservation (Van Engelen and Wen, 1995). SOTER methodology was set up upon the demand of updating global and national soil and soil environmental information by combining spatial information of soil--terrain units and their attribute data. The system was tested in South America and has been used extensively in Kenya, Central Europe and other regions and has been proved as a useful tool in regional land management (Oldeman and van Engelen, 1993; Mantel and van Engelen, 1999). Hainan Province covers million hectares and has an estimated population of 7.24 million. It is located in the north fringe of the tropical zone and most of China's tropical resources are concentrated in this region. Hainan has a favorable thermic climate throughout the year allowing potentially triple cropping. The climate is subhumid, it includes a rainy season with abundant reliable rainfall for rain-fed double cropping. In the dry season, the deficit of moisture for crop growth is large and a third crop requires irrigation. Extreme climate conditions hamper often efficient crop production. For examples, typhoons, which generally occur in the summer and autumn seasons, form a serious climate risk. The frequency of occurrence is 5.8 times y -1 and the wind speed reach maximum levels of 12 m s -1 with devastating consequences for agriculture. Therefore sites for rubber trees need to be situated in areas with adequate geographical shelter (Zeng and Zeng, 1989). Hainan is a major rubber producing region, contributing for more than 60%of the national rubber production. Although the rubber plantation industry is a pillar industry, rubber yields are about 25% lower compared to those in advanced countries, partly because of climate stress, partly the low soil fertility and an inefficient application of fertilizers, which do not take into account the site-specific soil conditions and the crops. In addition to rubber, Hainan produces sugar cane and various tropical fruits. An acceleration of the production of the tropical crops, fruits, off-season vegetables and quality seed for export to the mainland and abroad is being urged. The shift of current land use to specific tropical crops requires an economic and ecological sound assessment and planning. It is very much needed to up-date soil and terrain information of Hainan to locate the investments to suitable agricultural land without endangering the soils and other natural resources. As supported by UNDP, we have established SOTER for the whole Hainan Province (HaiSOTER) and for four typical ecology pilot areas (windows) with different natural settings (WinSOTER). The former is in scale of 1:250,000 and the later 1:50,000. The databases containing spatial and attribute data are operated with ARC/INFO and ARC/VIEW GIS systems. The SOTER databases were used as basis for further applications, in combination with other data sets such as land use map, climate record and transformed climate parameters

3 Materials and Methods Establishment of HaiSOTER and WinSOTER HaiSOTER and WinSOTER were compiled according to the standard methodology developed by International Soil Reference and Information Centre (Van Engelen and Wen, 1995). Because SOTER was developed for national or international resource inventory, medium to large map scale databases need a revision for some of the criteria determining terrain or SOTER units. Data source for the HaiSOTER construction and further interpretation included: - 1:100,000 Hainan Topography map - 1:200,000 Hainan Geological map - 1:200,000 Hainan Soil map - 1:200,000 Land use map, as interpreted from TM images of year Climate record of 18 stations all over the island Rich in attribute data is a major feature of SOTER but this also means a strict requirement of soil and environmental data. Topography map and geological map together meet the need for Terrain Unit grouping, but soil information was a bottleneck, as usually seen in other regions. In order to fill in the gap, extra soil investigation was made and supplementary representative soils were sampled and analysed, allowing every SOTER unit have at least one representative soil data set, which is essential for further model running and complete interpretation. For Win SOTER construction, detailed source maps of topography, soil were used. Typical landscape segments were identified in order to determine the proportions of different terrain components. In a large map scale (1:500,000), soil component proportions usually typify that of terrain components. The compilation of provincial SOTER database applied the procedures as given by Figure 1 WinSOTER followed the same methodology but used detailed source information from different pilot areas. Figure 1 Procedure of HaiSOTER compilation

4 Interpretation and application SOTER database provides a full spectrum of data from topography, lithology to soil physical and chemical tests, which are very useful for physical interpretations on various purposes. At this stage several kinds of interpretation and applications were made, which were basic thematic attribute assessment, soil water erosion risk assessment and land quality for main tropical crops of the island. In order to facilitate the application of SOTER data, an AEZ (Agro-ecological Zones) concept was adopted in which the whole region was divided into different agroecological zones according to mainly climate data. Climate stations were considered representative for the AEZ in which they are located. Within each AEZ, boundaries were drawn using iso-lines of rainfall and typhoon rainfall which are the main controllers of crop production, in addition to Thiessen polygons generation if each isoline zone can be further divided. Calculated LGP (Length of Growing Period) according to Driessen and Konjin (1992) was used as the main index to indicate water availability. The modelling process of both water erosion risk assessment and tropical crop suitability were illustrated as in Figure 2 Figure 2 Flow chart of modelling process. Water erosion risk assessment model An erosion risk assessment model that calculates the hazard for erosion for the land use under consideration was developed. SWEAP, the SOTER Erosion Assessment Program (Van den Berg and Tempel, 1995), was used to facilitate mapping of water erosion risk using SOTER data. The model was based on the Universal Soil Loss Equation (USLE, Wischmeier and Smith, 1978). For model running, all parameters were extracted from SOTER database (K, LS) and the annexed climate data (R) as well as land use data (C, P). The scale of the study (1:200,000) does not permit the quantification of soil loss in t ha -1. Instead, a qualitative unit (erosion hazard unit) was used and final soil erosion risk classes were made

5 Land quality analysis for tropical crop suitability Land evaluation is the assessment of the suitability of a tract of land for a specified kind of use, it provides objective sets of data on potentials and constraints, which can contribute to decisions on a sustainable land use. In practice this involves the comparison ( matching ) of the requirements of a specified land use with the properties of the land. A SOTER-based, automatic procedure for qualitative land evaluation is developed. This procedure was created in the Automated Land Evaluation System (ALES) (Rossiter, 1990). The ALES is a computer program that allows land evaluators to build their own knowledge-based system with which they can compute the physical and economical suitability of map units in accordance with FAO Framework for Land Evaluation (FAO, 1976). The ALES program works with so called decision trees being hierarchical multi-way keys in which the leaves are results e.g. severity levels of land qualities and the interior nodes of the tree are decision criteria. e.g. 1and characteristic values. These trees are traversed by the program to compute an evaluation using actual land data for each map unit SOTAL is a SOTER-based qualitative model developed in ALES for physical land evaluation in which presently three land utilization types LUTs are distinguished cultivated banana, coffee and rubber under different input and technology conditions. These LUTs are characterized by 11 land use requirements and evaluated by matching the land use requirements with the corresponding land qualities (Zhang et al, 2001). The objective was to design a procedure that allows for a quick separation of potentially suitable from non-suitable SOTER units for the intended land use indicating constraints to different kinds of land use. Results and Discussion Water erosion risk assessment Water erosion risk assessment was conducted using SOTER data, climate data and land use information. After overlaying AEZ map, SOTER unit map and land use map, each generated polygon was identified by PAT (polygon attribute table). In which parameters for SWEAP running are available. The calculated results were again linked to polygon and were then classified according to given criteria. A map showing the predicted result of one pilot area-baoting window, based on WinSOTER is given by Figure 3. Erosion sediment yield was estimated for current land use and a scenario that there wre no vegetation covers for the whole island. Table 1 gives the calculated results of Hainan island under the two land use conditions. It was obviously seen that under current land use condition soil erosion was weak and more than 90% of the land had no or very slight erosion. Because Hainan is a tropical island with not too intensive population, vegetation coverage is high and cultivated land on slope is normally terraced, so at the moment visible erosion takes place only in medium slope cultivated land, especially under summer storms. However, in a scenario that no vegetation covers, the calculated erosion would be high and more than 70% of the land would undertake strong to extreme strong water erosion. A regional distribution analysis of erosion hazard showed that central mountains would be vulnerable to erosion due to sharp slope

6 Figure 3 Soil erosion hazard of Baoting window. Table 1 Modelled water erosion risk in current land use and in scenario of bare land. Class Current Bare land Sediment yield (t/km 2 Area Proportion Area Proportion of the.a) (km 2 ) of the island (km 2 ) island No erosion Invisible Very Slight Slight Medium Strong Very strong Extreme > However at current land use condition the actual calculated erosion risk was relatively lower, because dense forest decreased C value and soil erodibility was not high due to adequate organic matter accumulation and good soil structure. The USLE has been designed for the calculation of soil losses in t ha -1. Considering the scale of study and the fact that the parametric model is not calibrated for the range of conditions as represented by the climatic and SOTER databases of the studied province, it is not considered justifiable to present the results on an absolute scale. Therefore, a presentation of the results in qualitative terms should be chosen. Validation of model results with actual erosion monitoring data is essential and this work is being conducted in some typical areas of the island. The current results allow for a comparison between the various areas of the island but does not give an absolute soil loss yield. Physical land suitability of tropical crops Banana, Coffee and Rubber are the dominant tropical crops in Hainan island. The LUT s (Land Use Types) as broadly defined for this study were: Banana/low input and low technology, Banana/medium input and low technology, Banana/medium input and medium technology, Banana/high input/irrigated and technology, Coffee/ low input and low technology, Coffee/medium input and low technology, Coffee/medium input and 655-6

7 medium technology, Rubber/mediu m input and low technology, Rubber/medium input and medium technology. The definition of these LUT's is listed in Table 2. Table 2 General definition of Land Use Types (LUT's) evaluated. Land Use Types (LUT's) Banana/low input and low technology Banana/medium input and low technology Banana/medium input and medium technology Banana/high input/irrigated and technology Coffee/ low input and low technology Coffee/medium input and low technology Coffee/medium input and medium technology Rubber/medium input and low technology Rubber/medium input and medium technology General management description low organic fertilizer, simple weeding and soil tillage, no terracing modest fertilizer with no mechanised weeding and soil tillage, no terracing modest fertilizer and mechanised weeding and soil tillage, artificial drainage high fertilizer, mechanised weeding and soil tillage and also good irrigation low lime and fertilizer and no mechanised weeding and soil tillage, no terracing modest lime and fertilizer with simple weeding and soil tillage, no terracing modest lime and fertilizer and mechanised weeding and soil tillage, no terracing modest inputs such as P fertilizer, availability for weeding, no terracing modest inputs such as P fertilizer, cover crop, weeding and soil tillage, no terracing Suitability assessment for Banana The interpreted results were given in Table 3, in which suitability of banana under different management levels were evaluated. Respectively, 61.45%, 38.45%, 83.47%, 61.46% of Hainan island were considered suitable for banana cultivation under different management of A, B, C, D. The extent of suitability of banana cultivation increased with input and technology levels. Severe restrictions of the land qualities availability of moisture, availability of nutrients and available foothold for roots were the main causes for constraining land suitability. Table 3 Extent (km 2, numbers) of Agroecological Units grouped for Banana suitability class. Suitability class Number of AEU s Extent AEU(km 2 ) A B C D A B C D Highly suitable (class1, 2) Moderately suitable (class3) Marginally suitable (class4) Not suitable (class5) Urban area/water/other Total A-high input,irrigated and mechanized; B-low input & low technology; C-medium input & low technology; D- medium input & medium technology

8 Maps (Figure 4) showing the spatial distribution of the physical suitability for banana under various input and technology levels are given. The main suitable areas for banana were at north and centre-north of the island which is flat or low slope land with deep soil and, with abundant rainfall. Soils are normally poor in chemical fertility (low in nutrients, esp. P), but given adequate fertilizer input the area can be good for most of tropical fruits. But mechanization level did not improve essentially the land quality concerned, esp. The Southern part of the island, because of its hilly condition. Banana/low input & low technol Banana/medium input & low technology Legen Legend S1 - highly suitable S1 - highly suitable S2 - moderately suitable S3 - marginally suitable N-notsuitable N W E Kilometers S Banana/medium input & medium technology Legend S1 - highly suitable S1 - highly suitable S2 - moderately suitable S3 - marginally suitable N - not suitable N W E Kilometers Figure 4 Banana suitability analysis under different management levels. S As for banana, the modeled results of the spatial extent and tabular data of the suitable areas for coffee and rubber under different management levels were also generated, but not presented in this paper. The general occurrence of suitable sites for these tropical crops was at north and centre-north region. For rubber tree, typhoon rains can be very damageable so the facing side of the east part of the island had a lower suitability compared to the central low hill regions. Conclusion SOTER databases for the whole Hainan island and for four pilot areas were built, according to the standard methodology, after some minor revision for the requirement of large and medium map scales. Interpretation and applications were made, for single attribute analysis, soil water erosion risk prediction and for exploring the suitability of crops under specified management. Methodologies are presented for above applications

9 This study aimed to highlight the status of current soil erosion of the island and, trends in productive capacity under different conditions, whereby the emphasis should be on the relative difference between scenarios and between land units rather than on absolute values presented. The approach may be used to support strategic decisions seeking to optimize land use, prioritize research, and guide conservation planning. It is shown that assuming appropriate conservation measures, the more management to be taken the more extent suitable for all the three tropical crops in Hainan island. The methodology can be repeated for relevant LUT s and can provide, especially when economic parameters are included, information on options for land use, resulting in a sustainable land use planning. Acknowledgement This project was supported by UNDP (CPR/96/105) and National Natural Science Foundation of China ( ). References Driessen, P.M. and N.T. Konjin Land Use System Analysis. Wageningen Agricultural University, Wageningen. FAO A framework for land evaluation. Soil Bulletin 32, FAO, Rome. Mantel, S. and V. Van Engelen Assessment of the impact of water erosion on productivity of maize in Kenya, an integrated modelling approach. Land Degradation and Development 10: Oldeman, L.R The global extent of soil degradation, pp In D.J. Greenland and I. Szabolcs (eds.). Soil Resilience and Sustainable Land se CAB International, Wallingford. Oldeman, L.R. and V. van Engelen A world soil and terrain digital database (SOTER)-an improved assessment of land resources. Geoderma 60: Rossiter, D.G ALES: a framework for land evaluation using a micro-computer. Soil Use and Management 6:7-20. Van den Berg, M. and P. Tempel SWEAP, A Computer Program for Water Erosion Assessment applied to SOTER. Documentation version 1.5 SOTER Report 7. ISSS. Wageningen. Van Engelen, V. and T.T. Wen Global and National Soils and Terrain Digital Databases(SOTER). Procedures Manual (revised edition). UNEP, FAO, ISSS, ISRIC. Wageningen. Wischmeir, W.H. and D.H. Smith Predicting rainfall erosion losse-a guide to conservation planning. Agriculture Handbook No US Dept. of Agriculture, Washington, DC. Zeng, Z.X. and X.Z. Zeng Physical Geography of Hainan Island. Science Press. Zhang, X.L., G.L. Zhang and Z.T. Gong Evaluation of some tropical crops in Hainan Province by using ALES based on HaiSOTER. Scientia Geographic Sinica 21(4):

Land Resources Information Systems A Source of Information and Basic Tool for Modelling and Monitoring Land-use and Land-cover Changes

1 Land Resources Information Systems A Source of Information and Basic Tool for Modelling and Monitoring Land-use and Land-cover Changes Otto SPAARGAREN and Vincent VAN ENGELEN ISRIC International Soil