Assessing forestland area based on canopy cover in a semi-arid region: a case study

Summary Forestland area estimates are the first principle of most ecological assessments across spatial scales. The definition of forest used by the United Nations Food and Agriculture Organization is based on a 10 per cent canopy cover threshold while the definition of forest used by the US Forest Service Forest Inventory and Analysis programme in western woodlands is based on a 5 per cent canopy cover threshold. Most of the eastern US forests have the ability to reach 10 per cent cover; however, in semi-arid areas such as western Texas, there is considerable area of forestland with less than 10 per cent canopy cover. The main objective of this research was to provide an estimate of forestland area in west Texas based on a 10 per cent canopy cover threshold. To accomplish this, we developed and compared three classification models that discriminate between inventory plots with less than 10 per cent cover and those with at least 10 per cent cover. We found that ~3.9 million ha (17 per cent) of forestland area (based on 5 per cent cover) did not meet the 10 per cent canopy cover threshold.