| dc.description.abstracteng | Agriculture plays an essential role in Vietnam, especially central Vietnam, which is the least developed economically and socially compared to the other regions. Many administrative levels in Vietnam have implemented “top-down” agricultural land use planning. In this system, the government assigns a specific land use for each given area. However, in the past, they have not paid attention to land evaluation during this process. This shortcoming has often led to irrational agricultural land use. This study was conducted in A Luoi district, Thua Thien Hue Province, Vietnam with the overall objective of mapping soils properties and evaluating land suitability for potential agricultural land use types.
In the first part of this research, the differences among soil organic carbon (SOC), soil total nitrogen (STN), and pH under different land use types and topographic aspects were compared. Soil organic carbon contents in arable land and plantation forest are higher than in natural forest and grassland (p < 0.05). Conversely, the soil total nitrogen in natural forest was significantly lower compared to other land use types. The soil of grassland, natural forest, and plantation forest were more acidic than from the arable land use type. Soil organic carbon and soil total nitrogen decreased with increase of soil depth in all land use types. The soil pH in plantation forest and arable land use types showed no significant change in relation to soil depth. Significant differences were also not found between topographic aspect and soil organic carbon content.
The second part of this research consisted of an accuracy comparison between the ordinary kriging (OK) and regression kriging (RK) methods for soil organic carbon, soil total nitrogen, and soil pH (from 117 soil samples). The results show that land use type, transformed soil adjusted vegetation index, and topographic wetness index are not suitable variables in the regression kriging model for soil organic carbon and soil total nitrogen mapping, however land use type could improve the accuracy of soil pH mapping. In general, the OK method seemed more accurate than the RK method for SOC mapping (by 3.33%) and for STN mapping (by 10%), but the RK method was found to more precise for soil pH mapping (by 1.81%).
Finally, the result from the Participatory Rural Appraisal (PRA) method indicated that five crops have good development prospects at the research site, namely rice, cassava, acacia, banana, and rubber. The land suitability for each type of crop is different depending on the weighting of natural conditions, economic aspects, and social aspects. An agricultural land use plan for A Luoi district was proposed based on the land evaluation results. Overall, acacia and cassava are the most suitable land use type at the site. Rubber represents the preferential crop in only a very small patch (5 hectares) in A Luoi Town commune. The parts of the central valley region are the only places in which rubber cultivation would be feasible. Banana cropping is feasible only in small areas in A Luoi Town and the neighboring communes along the main road, totaling 437 ha. The areas in which rice represents the crop of choice is also limited with an extent of 1,388 ha, and are scattered across the northwestern communes of Hong Bac, Bac Son, Hong Trung, Hong Van, and Hong Thuy. With an area of 23,835 hectares, cassava has the highest suitability indices compared to all other land use options. The cassava regions are predominantly located in the remote communes along the western border of the district, as well as in Huong Nguyen and Hong Thuy. The largest contiguous area for cassava production can be found in southern Huong Phong commune. Acacia also has a high suitability with the largest contiguous area, amounting to 18,438 hectares. The preferential acacia region expands from the northern communes along the valley toward the southern and eastern parts of A Luoi district. The combination of scientific and local knowledge in land assessment based on Geographic Information Systems (GIS) technology, Analytical Hierarchy Process (AHP), and PRA method is highly feasible. Systematic integration of PRA and reviews of existing literature is an appropriate land evaluation method. | de |