Agriculture, forestry, range resources

In the area of crop specie identification, it has been found that temporal data analysis, preliminary stratification, and unequal probability analysis were several of the factors that contributed to high identification accuracies. Single data set accuracies on fields of greater than 80,000 sq m (20 acres) are in the 70- to 90-percent range; however, with the use of temporal data, accuracies of 95 percent have been reported. Identification accuracy drops off significantly on areas of less than 80,000 sq m (20 acres) as does measurement accuracy. Forest stratification into coniferous and deciduous areas has been accomplished to a 90- to 95-percent accuracy level. Using multistage sampling techniques, the timber volume of a national forest district has been estimated to a confidence level and standard deviation acceptable to the Forest Service at a very favorable cost-benefit time ratio. Range specie/plant community vegetation mapping has been accomplished at various levels of success (69- to 90-percent accuracy). However, several investigators have obtained encouraging initial results in range biomass (forage production) estimation and range readiness predictions. Soil association map correction and soil association mapping in new area appear to have been proven feasible on large areas; however, testing in a complex soil area should be undertaken.