Developing a User Friendly Decision Tool for Agricultural Land Use Allocation at a Regional Scale

Agricultural land use planning should always be guided by a reliable tool to ensure effective decision making in the allocation of land use and activities. The primary aim of this study is to develop a user friendly system on a spatial basis for agricultural land suitability evaluation of four groups of agriculture commodities, including food crops, horticultural crops, perennial (plantation) crops, grazing, and tambak (fish ponds) to guide land use planning. The procedure used is as follows: (i) conducting soil survey based on generated land mapping units; (ii) developing soil database in GIS; and (iii) designing a user friendly system. The data bases of the study were derived from satellite imagery, digital topographic map, soil characteristics at reconnaissance scale, as well as climate data. Land suitability evaluation in this study uses the FAO method. The study produces a spatial based decision support tool called SUFIG-Wilkom that can give decision makers sets of information interactively for land use allocation purposes.This user friendly system is also amenable to various operations in a vector GIS, so that the system may accommodate possible additional assessment of other land use types.


Introduction
Land use planning in agricultural development is necessary to guide decision makers in attempting to select appropriate types of land use, determining optimal spatial locations of the planned agricultural activity, identifying and formulating opportunities for land use change, and anticipating the consequences of changing land use policies. As a core component of land use planning, land suitability assessment is fundamental at all stages of planning and implementation, and at any stage of development of an area  to determine the suitability and availability of land for alternative uses.
In Indonesia, agricultural development has gradually shifted to using marginal land as agricultural land is undergoing a massive transformation. This is often accompanied by changing objectives of land use, owing to increasing demands for specific commodities in the rural environment. Such a practice threatens long-term ecologically sustainable production which is recognized as a key element of food security, particularly in resource-poor areas. Moreover, land use change from agriculture to non-agriculture as such becomes one of common phenomena in urban and rural areas that threaten the country's food security program. Therefore, a tool needs to be developed to assist land resource assessment in the decision making of land use choices for a dual and indivisible role: the first being that of maximizing benefits (productivity of the land), and the other being that of managing public goods: the environment (see Baja, 2009). To achieve this, it requires an effective management of geospatial information on the land on which such decisions should be based. Land suitability evaluation, as one of the branches of land resource evaluation (FAO, 1976) or land resource assessment, has been recognized among land use planners and policy makers as a primary tool for making appropriate land use choices (see Baja, 2009;Elsheikh et al., 2013).
From land resource assessment perspectives, collection of information over space and time has always outperformed our ability to interpret and apply the data. Consequently, the major goal of land suitability evaluation for site-specific management, which is to see a continuous improvement of management decisions,

Land Suitability Evaluation Method
Land evaluation procedures in this study involve eight main components ( Figure 4). The assessment of land quality for a specific type of land use should be based on land use requirements and constraints. Such requirements and constraints are then used as the basis for establishing what is termed evaluation criteria or decision criteria. This is then considered as a multicriteria decision making analysis (Sarkar et al., 2016). With reference to these decision criteria, a set of algorithms is then employed to match the existing quality of land and the requirements of that particular type of land use. The matching procedure then gives rise to a ranking of the potential of land for a given purpose , whether categorical or continuous grades. Regardless of the approaches employed, the final result of land evaluation is a map that portrays the divisions of the area of interest into suitability classes or indices of land units for a nominated land use.

Figure 4. Components of land suitability evaluation in this study
Database of the system contains various commands for depicting land suitability class in the study region. Land suitability classification in this study was undertaken based on the framework for land evaluation guidelines of FAO (1976). The classification is purely based on inherent land characteristics, which is popular among soil scientists and agronomists, where the main concern is classifying the land according to the goodness or appropriateness of soil characteristics for a specific purpose (see for example Baja et al., 2014;Davidson et al., 1994). As seen in CSR/ FAO (1983), the FAO's land suitability scheme is divided into Order, Class, Sub Class, and Unit. Order is the global land suitability group, and is divided into S (Suitable) and N (Not Suitable). Class is the land suitability group within the Order level. Land suitability classification is undertaken based on the level of detail of the data available. For example, at a semi detailed mapping activity the S order is divided into Highly Suitable (S1), Moderately Suitable (S2), and Marginally Suitable (S3). In the "Not Suitable" order no further division is made.
Furthermore, Sub-Class is indicated by the type and level or degree of limitations in each division. For example, land unit having a limiting factor of rooting condition at a marginal level is indicated by a Subclass S3rc. Further, detailed divisions of Sub-Classes into Units can be made according to differentiation in soil effective depths. The effective depths of 50 -70 cm and <50 cm, are respectively classified as S3rc1 and S3rc2 (Djaenuddin et al., 2003).

Land Units And Soil Characteristics
Soil characteristics surveyed and analysed for developing the SUFIG-Wilkom system include those forming the basic criteria for land suitability classification according to FAO (1976) ( Table 1). The data on climate such as average temperature, rainfall, number of dry months were obtained from local meteorological station.

GIS Database Processing
In this project, data processing and spatial analysis were undertaken in a vector GIS, on the basis of different themes developed earlier in system development. A GIS database table containing the names and codes of all unique attributes in the study area was constructed in a relational database system. As seen in a common GIS data base, unique soil identification numbers (IDs) were assigned to each record, and fields were added to indicate map unit types and a general classification of soil type at a sub-group level. Relationships between the various data sources as mentioned above were defined and well organized in a system (GIS and its derived subsystem SUFIG-Wilkom). At each stage of database processing, careful consideration was given when designing a spatially referenced soils database to ensure database table linkages, due to the nature of soil mapping units, taxonomic classifications (USDA, 2006), and the taxonomic level at which the soils were mapped, and the level at which soil attribute data was recorded. It also applies to other GIS data layers and attributes.
In terms of land suitability classification, SUFIG-Wilkom contains interactive commands that has a linkage to database of land and requirements of crops. In GIS systems, all these data sets provide a complex set of relationships between soil profile attribute data (i.e., characteristics as shown in Table 1), soil type, and soil mapping unit, so that the flow of soil profile data from the soil attribute tables to linkage with the database tables of the digital soil maps can be well understood. The tables are arranged such that all relationships are "one-to-many," thereby facilitating data flow and database queries in GIS (Baja, 2009). GIS database consists of two sets: spatial data and attribute information attached to spatial data.

Programming Tools
SUFIG-Wilkom is constructed by utilizing a programming tools that allow for customizing the existing application programs in GIS. With such tools, the capability and functionality of SUFIG-Wilkom can be improved considerably or even expanded up better than the standard package modules (see for examples, Andreu et al., 1996;Oliver et al., 2012). All the functionality contained in SUFIG-Wilkom can be controlled by a program that is created using the Visual Basic programming tools. Visual Basic is a system that is more powerful and more compatible with other applications including ArcObject features that allow for creation of a stand-alone application.

Menu and Toolbar
Menu is a list of commands that is found at the top of SUFIG-Wilkom, while toolbar is a set of tools that are grouped according to their respective functions. This study utilises four main menus that are prepared for use in current research or future development. These main menus consist of ( Figure 5): • Commodity_Selection  Figure 9, an sion y has produce ral commodity ely in GIS to on related to la f agriculture c nd fish pond s and criteria, ment. A user fr ferent level of od mapping acc like other G on the movem uses the Zoom ouse button th to move the m ouse button. Cl oducts Wilkom are tw ap products. M vailable map l raphs extracte ub-menu, for e level for a sp nd its attribute Figure 9 ed a user frien y) allocation, on guide users t and use, and commodities,  Vol. 11, No. 5;2017