Economic analysis of site survey and productivity modelling for the selection of plantation areas

In designing appropriate site survey procedures, the critical question for forest managers is not only how accurately can yield be forecast, but rather what level of soils information is required to make sound on-the-ground decisions? Two issues are pertinent: (1) what intensity of survey is appropriate and (2) what level of information should be collected at each survey point. The resolution of these issues is one of cost-benefit: are the costs of a more intensive site assessment offset by better selection of the area to be planted? Two plantation areas, that had been planted before the current site selection and forest productivity prediction systems were developed, were selected for study. Predicted productivity surfaces for these plantation areas were made using different qualities and intensity of site survey data as inputs into the process-based site productivity model, PROMOD. These predictions were used to determine the area within the plantation boundaries that was in fact suitable for plantation development under current planting guidelines. Using high intensity site survey procedures (one sample plot per hectare drilled deeply with salinity and nutrient sample taken) PROMOD was able to produce an accurate productivity surface. However, even a low intensity survey (one sample plot per 5 ha drilled shallowly) prevented the worst land being selected for plantation development. Comparing the net present value of the plantation that would be established under the high intensity site survey procedure to that established in 1990 suggests that research and development into site survey and site productivity modelling may be worth $500 ha(-1) of plantation established. (C) 2002 Elsevier Science B.V. All lights reserved.