GISdevelopment.net ---> AARS ---> ACRS 1989 ---> Land use

PC based GIS land use capability evaluation

Toshiaki Hashimoto, Shunji Murai
Institute of Industrial Science, Univ. of Tokyo
7-22, Roppoongi, Minato-ku, Tokyo 106

Norihisa Ohno
N.C.M Co. Ltd. 4-33-22, Hakusan,
Bunkyo-Ku, Tokyo 112 Japan


Abstract
Land use capability can be evaluated by the superposition of some kinds of geographic informations such as land uses. Soil conditions, slope gradients, etc. This procedure can be automatically carried out by a computer when the geographic informations are digitized. This paper deals with the system developed by the authors for land use capability classification which runs on a personal computer. The contents processed by this system are the land use capabilities of flood damage risk, slope failure risk and the selection of suitable land for agriculture or forestry.

Introduction
Land use capability evaluations had been carried out by manual superposition of some kinds of existing maps. This procedure was very time consuming and had the probability that the results depend on the skillfulness of interpreters. But the problems are resolved by the existence of a computer which has the capability of fast data handling as well as powerful information system. Computer processing has another advantage of the ability to generate new data from existing data such as slope gradient data derived from elevation data.

On the other hand, the processing ability of a personal computer has been so improved that a large volume of data like remote sensing data can be managed on it. The usage of a personal computer leads to the achievement of a simple and cheap system for GIS.

Evaluation Technique
  1. General
    The evaluation is based on the superposition of some kinds of different factors which are independently categorized into same ranks. The factors which would affect the objects to be evaluated should be selected. Fig. 1 shows the objects and their factors in this system. It also shows the data sources for factors. This paper introduces the details of the flood damage risk and the case of developing paddy fields in the selection of suitable land for agriculture.

    The factors selected are related with natural land condition and not so related with agro-economical condition, so the results means only the evaluations from the view point of natural aspects.


    Fig. 1. Objects, factors and data sources


  2. Categorization of factors (score giving)
    The factors are categorized into some ranks according to the degrees of effect on the object. The scores for the selection of suitable land for agriculture are different from each other because of the difference f effects, although the same factors are adopted. The scores may be changes in consideration of the conditions of the areas to be evaluated.

    1. Land use for flood damage
      This factor indicates the damage potential at a flood.

      3 : cities, villages, densely inhabited areas

      2 : others (covered area like factory sites, etc.)
      1 : rice fields, farms, others(uncovered area like golf courses)
      0 : grass fields, forests, lakes, rivers

    2. Land form for flood damage
      This factor indicates the flood risk which is based on the results of correlation analyses between land form and the records of former floods.

      3 : plain in valleys, deltas, flood plains, marsh, peat areas, artificiality modified areas 1 (flat, surf zones, beaches)
      2 : alluvial fans, small hills (or small hilly areas), dunes, artificiality modified areas 11 (exception of 1), depressions, gravel sedimentations
      1 : table land
      0 : mountains, hills

    3. Land use for the selection of suitable land for agriculture
      This information is used for determining the area to be evaluated

      • paddy fields
      • crop lands
      • orchards
      • grass fields
      • forests
      • lakes, ponds, rivers
      • cities, villages, artificial structure ( airports, roads, etc.)

    4. Slope gradient and soil depth for the selection of suitable land for developing paddy fields
      Slope gradient indicates work efficiency and the probability of erosion and soil depth is related with work efficiency and productivity.

      Slope gradient soil depth
      3 : < 8 degree 100cm <
      2 : 8-15 100-70
      1 : 15 -20 70-40
      0 : 20 < < 40

    5. Soil character for the selection of suitable land for developing paddy fields
      This factor indicates work efficiency, productivity and the probability of erosion. If the characteristic of soil is evaluated as the lack of phosphorous and/or potassium or the great acidity, the score should be lowered by one rank. Soils are expressed by the following acronyms.

      S : sand or sandy L : loam or loamy
      C : clay SI : silt
      Li : Light H : heavy
      V : volcanic G : gravel
      P : PEAR (ex. SL : sandy loam)

      3 : L, SiL, SCL, CL, SiCL, SC, LiC, Sic
      2 : SL, VL, VCL
      1 : S, LS, VSL, HC, P(lower)
      0 : G, VS, VSG, P(upper)

    6. Saline distribution for the selection of suitable land for developing paddy fields
      This factor may be ignored if there is no probability for salinity damages.

      3 : not probable for salinity damage
      2 : probable for salinity damage
      1 : covered with salinity of less than 10%
      0 : covered with salinity of more than 10%

  3. Evaluation

    1. Evaluation of flood damage risks
      The evaluation is carried out by the superposition of two factors categorized into four ranks. Both ranks are multiplied as shown in Fig. 2. The ranking matrix can be modified according to the conditions of the areas to be evaluated. Land use and form represent the damage potential and the flood risk, respectfully. So the multiplied scores can be interpreted as Table 1.

    2. Evaluation of suitability for developing paddy fields
      In land use planning, each land use has the different importance according to the reasons of economy, politics, etc. So it is supposed


      Fig. 2. Ranking matrix and evaluation flow


      Table 1 Evaluation for flood damage risk
      Score remarks
      0 No Probability (not in undated)
      No damage even when in undated
      1 Smaller damage with low probability
      2 Small or medium damage with low or medium probability
      3 Serious damage with low probability
      Small damage with high probability
      4 Medium damage with medium probability
      6 Medium damage with high probability
      Serious damage with medium probability
      9 Serious damage with high probability

      that it is probable to develop paddy fields into urban area while the reverse is improbable. In this system, the present land uses to be evaluated can be selected. For example, when the present land use of pasture is selected for the evaluation for developing paddy fields crop lands, the evaluation is carried out only for developing the specified land uses.

      From the view point that every factor used for the evaluation of agricultural suitability limits the productivity independently. The total score of any point is expressed by the minimum of every score of three (ore four in case of the usage of saline distribution) factors. The score expressed by minimum is interpreted as Table 2.


Fig. 3. Evaluation flow

Table 2. Evaluation for the selection of suitable land
Score remarks
3 well suited for developing specified land use having no or slight significant li limitations
2 moderately suited for developing specified land use having some limitations limitations
1 poorly suited for developing specified land use . having severe limitations limitations.
0 not suited for developing specified land use.

System
  1. Hardware
    Fig. 4 shows the hardware configuration. The personal computer of "NEC PC-9801 RA5" performs as a CPU. The hard disk has the capacity 40 Mbytes. The CRT can display an image by 640 x400 pixels in 125 colors by dither technique.

    The scanner is used to acquire digitized data from existing maps and photographs. It has the functions as follows: a scan size is up to A3 (297 x420 mm) by one scan : a color resolution is 1 byte in R.G.B. respectively : a spatial resolution is up to 600 dots per inch.


    Fig. 4. Hardware configuration

  2. Data handling
    As mentioned above, seven kinds of data are used in this system. Among these data, land use data are derived from DTM which is generated from topo maps (1), other data are directly derived from existing maps.

    All image data are in raster form and every pixel is coded by a polygon ID number. There is prepared the look up table where coded data can be compared with the corresponding legends. Other tables for color informations and the scores for each object are also prepared.

  3. Software
    This system can be operated interactively by using menu styles. The functions of the system are shown in Table 3.

    Table. 3. Functions of the system
    1 To display a geographic Information
    2 To display the results of evaluation
    3 To set up scores
    4 To set up color codes

    The image data like geographic information are displayed in colors selected among 125 colors. The colors can be changed by changing color code in the color table. The scale and the location to be displayed are also changeable. The image of results of evaluation can be stored.

    The number of image to be displayed is only one, but geographic informations are overlaid. That is to say, the legend of a certain pixel can be seen on the display when the type of informations to know and the location are specified. For example, the legend of land use on a certain pixels appears on the display while the image of land from is displayed. The specification of legend and location is operated by a mouse.
Applications
  1. Flood damage potential at Nakhon SI Thammarat, Thailand
    The area is located on the north-west of Nakhon Si Tammarat, the southern part of Thailand where a serious flood damage occurred in November, 1988.

    Land use data are prepared by a conventional supervised classification method from a TM image acquired on March 30, 1988. The image consist of 480z480 pixels and 1 pixel correspondents to 50m on the ground. The land form data are derived from DTM generated from topo maps, of 1/50,000. The data are adjusted o the land use data. The result of evaluation is shown in Fig. 5.


    Fig. 5. Flood damage risk


  2. Suitability for developing paddy fields at Khon Kaen, Thailand.
    The area is located around Khon Kaen, the north-eastern part of Thailand where the level of agro-economics is very low.

    TM image acquired on December 18, 1987 Are classified to generate, the land use data of the pixel size of 66.7m on the ground. The image consists of 334p x 433 lines. The slope gradient data are derived from DTM generated from topo maps, of 1/50,000. The soil character and soil depth data are derived directly from soil maps, of 1/250,000. The saline distribution data are digitized from saline distribution maps, of 250,000. All data are adjusted to the land use data. Fig . 6 shows the result of evaluation.

Fig. 6. Suitable land for developing paddy field


Conclusion
A new system for land use capability classifications has been developed. The system runs on a personal computer and is easy to operate. GIS is useful for suitability studies in land use planning and this system will contribute to this field.

Reference
  • S. Viseshin, T. Hashimoto, S. Murai :Interactive system for Automated Generation of DTM from Existing Topo-graphical MAP of Thailand", The proceeding of the ninth Asian Conference on Remote Sensing, Nov. 23-29, 1988, Bangkok, Thailand, pp s-5-1-s-5-6.
---------------------------------------------------
Presented at the tenth Asian Conference on Remote Sensing held in Kuala Lumpur, Malaysia. November 23-29, 1989