Land capability assessment and classification for sustainable pond-based aquaculture systems

The project was officially launched in November 2005 in Makassar, Indonesia following the foundation of a National Steering Committee in Jakarta to coordinate the extension components of this project and FIS/2003/027. Prior to the official launch, the project leaders and collaborating scientists from RICA, GMU, AIMS and UNSW met in Sydney to develop the research and extension strategy and build a more effective collaborative network between the partner agencies. The team also commenced reviewing existing, relevant datasets from the former project (FIS/97/22). Data collected from FIS/97/22 is suitable for the current project and has been compiled and converted into raw files for GIS analyses. A fieldwork plan was implemented in January 2006. Additional soil and water quality and farm production data were collected for Barru, Wajo and Luwu to build on the database of the previous project. The elevation and coordinates of past and new sampling locations have been surveyed into a network of benchmarks to build terrain and GIS models of each study site. Hydrological assessment of tidal conditions and surface hydrology has also been conducted. Hydrological models have been applied to data to predict tides and characterise tidal regimes. Primary and secondary soil, topographic, climatic and hydrological data from Australian shrimp farming areas commenced and will be supplemented by additional primary data collection in late 2006 to early 2007.

The project has updated the computer mapping facilities and soil and water testing laboratory at RICA. GMU academics trained RICA mapping staff in Image Analysis in April 2006 and have assisted RICA by building expertise in Remote Sensing to compliment the centre's strengths in GIS modelling and mapping. GMU, RICA and UNSW staff have also conducted joint fieldwork to produce preliminary mapping units for the land classification scheme. The mapping laboratory at RICA will be expanded in mid 2006 through Government of Indonesia funding to accommodate an increase in project staff and new mapping resources jointly funded by ACIAR and the Government of Indonesia. The expansion of the laboratory will maintain its leadership in coastal aquaculture mapping in Indonesia.

Two postgraduate students are conducting their Masters by Research programs under the project. Mr Tarunamulia, an Ausaid-funded student and a Research Assistant under another ACIAR project, is working on the application of fuzzy logic to site assessment and land classification. Ms Claudette Rechtorik commenced employment as a part time Research Assistant in November 2005 and is conducting her Masters by Research program on Australian farms.

A review of research progress in May 2006 showed that the study sites in South Sulawesi will not provide sufficient data for the development of a combined coastal aquaculture classification scheme due to the low intensity of sea cage culture. To address this issue FIS/2002/076 and the related seacage project (FIS/2003/027) will be including sea cage and land-based sites in Lampung, Sumatra where both industries occur at the same location and at higher intensity than in South Sulawesi. The review of project data also showed that sandy soils, which are a problem elsewhere in Indonesia, are not sufficiently common in South Sulawesi. Accordingly, data from a recently launched ACIAR project in Aceh (FIS/2005/009) will be incorporated into the current study. Study sites in Aceh include sandy soils, acid sulfate clays and acid sulfate sands.

The Australian and Indonesian teams have designed and commenced implementing farm surveys to collect data on farm management, production and environmental conditions in Barru, Sinjai, Wajo, Luwu and Lampung in Indonesia and northern NSW and South East QLD in Australia. Farms in Far North Queensland will be surveyed in the next 6 months. The Australian Prawn Farmers Association is facilitating the survey work. Socioeconomic data from farmer interviews in both countries are being used to: investigate environmental risk perception and how it influences on farm management and site selection; identify socioeconomic and policy issues that influence site selection, and; to develop more effective dissemination strategies. The farm surveys are continuing through to the beginning of 2007 due to the incorporation of the Lampung and Aceh study sites. Primary data from the former project (FIS/97/22) for both countries have been collated into a database and are being analysed and converted into GIS data files for integration with socio-economic data from the farm surveys.

A major review of Indonesian and Australian aquaculture policy and planning tools has commenced to evaluate the current approaches to environmental decision making and to determine levels of scientific and technical information available to farmers and government officers.

The project is now in its second year and has developed draft site selection criteria for land-based farming and has commenced creating GIS models to map areas according to suitability for extensive, semi-intensive and intensive farming. Environmental factors relevant to the development of the Australian prawn farming industry have also been identified and, by the end of 2007, can be incorporated into existing policy and revised development guidelines to improve decision making for both farmers and government agencies. Findings from the Indonesian project and the development of site assessment methods will be directly applicable to intensive farming in Australia demonstrating benefits to the Australian industry from the collaborative research. The draft site selection criteria have been divided into two formats. The first format presents more technical criteria that target the staff of the Ministry of Marine Affairs and Fisheries and Australian Government Departments who have access to research support services. These more technical criteria rely on numerical data that can be analysed and interpreted using guidelines, to assess land according to physical properties (eg soil type, landform, erosion, elevation and slope), hydrological and meteorological (eg tidal conditions, rainfall), sociocultural (eg effects on or from adjacent landuse, proximity to areas of cultural significance or conservation areas, areas protected by other policies or environmental planning instruments) and accessibility to infrastructure. Draft guidelines to support the decision making process are now under development. The second format involves the presentation of site selection criteria in less technical, farm level guidelines on site assessment relevant to both Australian and Indonesian farmers, and also to farmers in South East Asia and South Asia (with modification to the supporting guidelines). These guidelines use less text and do not heavily rely on scientific data. The guidelines present simple and cost effective methods to evaluate land according to the physical, chemical and biological characteristics of a farm or an area of interest. Both formats will be evaluated by the Local Advisory Committee in Indonesia and stakeholders at an extension workshop and field visits scheduled for November 2007.

The project team has developed draft GIS models to map areas for land-based aquaculture using South Sulawesi study sites and selected areas of Aceh drawing on data from FIS/2005/009. Only preliminary analysis of data from Lampung, Sumatra has been undertaken but will be completed by November 2007 when all data collection and analyses will be completed. The draft GIS models have been applied to 1:100,000, 1:50,000 and 1:10,000 scales to evaluate resolution and accuracy limitations. At 1:100,000 and 1:50,000, coarse scale mapping is proving to be useful as a scoping tool. At these two scales, the Ministry of Marine Affairs and Fisheries can use the mapping classes to identify areas that are suitable or unsuitable for land-based aquaculture but a further breakdown into different levels of suitability is not possible at 1:100,000. At 1:50,000 a breakdown of the suitability classes is hamstrung by data availability. By contrast the 1:10,000 draft maps demonstrate a capacity to breakdown the suitability classes into:

highly suitable - suitable for all forms of land-based aquaculture but best used for intensive farming
moderately suitable - suitable for extensive to semi-intensive farming
marginally suitable - best suited for extensive farming or polyculture

These draft classes will be reviewed further in November 2007 with stakeholder inputs and groundtruthed before a final set of classes are applied. Each suitability class is currently being subdivided further so that sub-classes specify farm management requirements to ensure that the carrying capacity of the land is not exceeded and prevent environmental and social impacts. Project FIS/2002/076 is also developing similar GIS models for seacage farming in collaboration with Project FIS/2003/027 to develop an overall coastal classification scheme for aquaculture.

The socioeconomic studies have shown that there is a significant need for integration of information on disease prevention and management with site assessment. Accordingly, the current project is collaborating on ACIAR Project FIS/200/061 to develop complimentary extension materials and to target a common audience. In addition, the two projects have designed joint studies to investigate associations between patterns of disease outbreaks with environmental risk factors (eg. soil type).

The draft guidelines, maps and related extension materials will undergo stakeholder review in November 2007 and published by May 2008. In 2008 the project team will focus on the production and promotion of extension materials and will run workshops to facilitate uptake. All extension activities involve inputs and guidance from the National Steering Committee and Local Advisory Committee in Indonesia.

The project focussed on the completion of fieldwork, data analyses, production of draft site selection criteria and associated draft land capability maps. The project also co-developed mapping models for seacage farming with FIS/2003/027 (seacage project).

Fieldwork was completed at all South Sulawesi-based locations including the ground truthing of draft maps for land- and sea-based aquaculture. Sites on the east and west coast of South Sulawesi were used to compare different soil, geomorphic and hydrological settings and various production systems (shrimp monoculture, polyculture, seaweed culture and fish monoculture). Data from FIS/2005/009 in Aceh were used to include sandy soil environments in the data analyses and subsequent development of the site selection criteria and mapping models. Study sites in Lampung were included under a project variation to enable the land-based and seacage projects to conduct joint research where seacage and land-based farming operate intensively in close vicinity. Environmental, socioeconomic and production data were analysed using multivariate methods to develop site selection criteria for a range of extensive farming practices. The draft site selection criteria target farm-level decision making. The criteria are based on low cost technologies to evaluate the physical and human constraints on brackishwater aquaculture development, and decision making processes to identify issues that require technical intervention from government agencies. Draft site selection criteria guidelines are close to completion and will be reviewed by stakeholders in 2008.

Site selection criteria were developed further and translated into GIS-based mapping criteria to underpin the development of mapping models for spatial planning purposes. Field and remote sensed data were combined in the analyses to refine techniques for land evaluation over large areas, and to improve the predictive accuracy of the image classification used in the mapping. The draft GIS models were first tested in Pinrang regency and more recently applied to sites in Lampung and the east coast of South Sulawesi. Ground-truthing was used to test the accuracy of mapping boundaries and land classification. Data from the Australian study sites are currently being analysed to revise existing site selection criteria and provide inputs to policy development in NSW and Queensland. Site selection criteria from the Indonesian component are being revised in the context of the Australian data set and the intensity of the farming systems on the east coast of Australia. The analysis of the Australian data set also involves multivariate approaches.

Mapping models for Indonesia are based on three spatial scales. At 1:100,000 the maps differentiate between suitable and unsuitable classes for regional-scale planning. Selected areas have been mapped at 1:50,000 and the project is currently testing the accuracy of suitability sub-classes at this scale. At the 1:10,000 scale all maps are sufficiently detailed to divide the suitability classes into well-defined sub classes which include specific information on farming methods, the type of culture systems best suited to the location and the maximum intensity of the farming systems. Draft guidelines for map interpretation are being produced and will be reviewed by government agencies in July/August 2008.

The land-based and seacage project have continued to operate under the auspices of the National Steering Committee (NSC) and a Local Advisory Committee (LAC) in South Sulawesi. The two committees have successfully fostered information exchange between the research agencies and stakeholders leading to more effective extension and adoption. The spatial data and mapping outputs of the project have also been adopted by the Agency for Regional Development and Zoning and the Bureau of Rural Settlement to facilitate broader landuse planning and to minimise encroachment of brackishwater areas by other landuse. The two committees have enhanced interagency collaboration and communication between the projects and farmers. A joint extension workshop, held in November 2007, enabled stakeholders, researchers and the LAC to further develop the dissemination and adoption strategies of the land-based and seacage projects. Stakeholders were also involved in small workshops to identify appropriate extension materials. A print publishing process was also developed to improve in-country production of extension materials for both projects. The land-based team visited farmer groups after the workshop to plan community extension and adoption programs. A farmer workshop was held in Luwu in December 2007 and more district-based workshops are planned for 2008.

The project has also developed stronger links to FIS/2000/061 (disease management project). The land-based project has characterised and mapped soils at the disease project's experimental ponds in Barru and Pinrang, and in 2008 will also undertake a social survey to develop a better understanding of farmer risk perception and how it affects the uptake of technology. The two projects are also integrating information on pond management to minimise the effects of soil and water quality on disease outbreaks.