Fuzzy logic for fine-scale soil mapping: A case study in Thailand

R. Moonjun*, Dhruba Pikha Shrestha, V.G. Jetten

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

8 Citations (Scopus)
52 Downloads (Pure)


Conventional soil survey methods are labor intensive and prohibitively expensive considering the area to be covered. Unfortunately, the current soil survey products are not adequate, either categorically or cartographically, and cannot be easily downscaled for its application at farm-level. On the other hand soil is a continuous variable and does not have abrupt boundaries in nature. One soil type can change gradually to become another class. This creates problem in delineating soil boundaries due to overlapping of classes. In this situation fuzzy logic can be useful. In conventional soil survey, this is solved by creating mapping units such as consociation, association or complex. In consociation the delineated areas are dominated by one soil type (at least 75%) with some inclusion of other soils. When dissimilar soils occur in a consistent repeating pattern, it is mapped as an association. In classification using fuzzy logic a pixel may have multiple class membership and the one with the highest membership or similarity value gets the class label. The main objective of the study is to assess the usefulness of fuzzy logic in increasing efficiency in soil mapping. The study was conducted in Lomsak, Phetchabun province in Thailand. An expert system was used whereby rule-based reasoning was applied for mapping soil series and topsoil texture in which the soil-landscape relationship was taken into account. Lithology and terrain parameters were used as predictor variables. This resulted in mapping 17 soil series and 10 topsoil texture classes in a complex landscape. In the conventional soil survey technique, it is possible to map only 8 soil series at 1:50,000 scale, indicating that detail soil mapping is possible by using fuzzy logic. The accuracy of the fuzzy logic derived soil series map was tested using a set of evaluation data. The result showed an average accuracy of 70%. Fuzzy logic has the potential for reducing inconsistency and costs associated with the traditional soil mapping processes as mapping can be carried out with a relatively low density of soil samples. The research results can be used to support soil survey works in complex landscapes at sub-watershed scale.
Original languageEnglish
Article number104456
Pages (from-to)1-19
Number of pages19
Early online date3 Mar 2020
Publication statusPublished - 1 Jul 2020


  • 22/2 OA procedure


Dive into the research topics of 'Fuzzy logic for fine-scale soil mapping: A case study in Thailand'. Together they form a unique fingerprint.

Cite this