Using GIS and radar imagery for assessment of Land Capability for arable agriculture in Rigo District, Central Province, Papua New Guinea

Accurate and reliable spatial data on soil types is very limited in Papua New Guinea. The onlyregional digital dataset available is the Papua New Guinea Resource Information System (PNGRIS). This dataset is an highly extrapolated modelled GIS dataset which provides only a coarse and broad interpolation of probable soils which does not provide realistic on ground soil representation.

Using a 10 m Digital Elevation Model (DEM) derived from GEOSAR Radar data a four class Topographic Position Index was generated using Land Facet Tools Extension for ArcGIS (Jenness et. al 2011). This extension divided the topography into Ridges, Upper Slopes,Gentle Slopes and Valleys. Using this classification ”Lower Slopes” of less than and equal to 10o were selected as potential suitable sites for intensified arable agriculture. Slopes above 10o gradient are more susceptible to rill and sheet erosion due to the intense high precipitation events experienced during the wet season.

This broad topographic based classification was further constrained by the bedrock geology underling the previously identified lower slopes. The soils best suited to intensive agricultural production were identified as those areas underlain by intermediate or mafic rocks e.g. basalt, gabbro and other igneous rocks high in ferro-magnesium minerals and iron oxides and their derivatives which provide the potential for the development of more productive and sustainable agricultural soils.

This broad first pass classification identified 41,533 hectares of land for potential agricultural expansion within the Rigo District in the coastal lowlands and foothills centred about the town of Kwikila some two hours by road southeast of the capital Port Moresby. Of the 41,533 ha identified some 10,632 ha is identified within PNGRIS as being prone to waterlogging and inundation of varying duration and severity. Most observed current agriculture land also falls within this first pass classification.