Combining Metric Aerial Photography and Near-infrared Videography to Define Within-field Soil Sampling Frameworks

Navigation: intro | 1. method | 2. fusion | 3. fusion | 4. fusion | 5. frameworks | 6. geostats | 7. geostats | 8. conclusions

[an error occurred while processing this directive]

Introduction

Spatial data collection can contribute up to 80% of the cost of deploying a Geographic Information System (GIS) based Decision Support System (DSS). The use of remotely sensed information, field survey using differential Global Positioning System (dGPS) and geostatistical interpolation methods maximises data quality for a given rate of sampling. This study investigates the combination of metric aerial photography and near-infrared (NIR) videography data to improve the design of field-survey sampling frameworks.

For the purposes of a soil survey there is the recognition that characterisation of within-field variability can be a problem for random, or grid-based sampling frameworks. To address this issue, sampling can be stratified using secondary sources of information such as land use plans, soil map units or remotely-sensed imagery. Remotely-sensed information is easily synchronised with field survey and from a light aircraft platform is available at a cost which is being reduced by developments in sensor and image processing technology.

This study presents a methodology for defining soil-sampling frameworks based on the integration of medium format colour photography and NIR videography. It outlines the methodology adopted for the preparation and integration of the image data and then details the approach taken to structuring the soil-sampling frameworks using models of within-field variability. Two sampling strategies are compared, one using the field as the sampling unit and the second using variability classes to sub-stratify the sampling. The effectiveness of the two strategies is compared using the prediction accuracy of geostatistically-derived soil property maps.

Follow links 1 to 8 to view the methods, results and interpretation for this study.