透過您的圖書館登入
IP:3.145.119.199
  • 期刊
  • OpenAccess

Models for Predicting Stem Diameter from Crown Diameter of Open Grown Trees in Sondu-Nyando River Catchment, Kenya

並列摘要


Information on stocks of trees on farm is scanty and in many cases lacking. Assessing the stocking density of trees on farms require models relating tree bole diameter with its crown diameter. However, bolecrown diameter models of open grown trees on farms is lacking in the Lake Victoria basin and indeed in Kenya. The focus of this study was to develop regression equations that would predict individual tree Diameter at Breast Height (DBH) from its crown diameter (Cd) for selected open grown tree species in Sondu-Nyando River catchment. Stratified random sampling was used and GPS readings, DBH and Cd, collected from 20 unequal sized sample plots, along selected agro ecological gradients. The models were developed using DBH and Cd measurements from 578 trees of five most dominant tree species (Cupressus lusitanica, Eucalyptus saligna, Grevillea robusta, Persea americana and Croton megalocarpus). The DBH - Cd models displayed good fit (R^2 > 0.586). Persea americana had the highest adjusted R^2 (0.875) and Eucalyptus saligna the lowest (R^2 = 0.586). F-test showed regression coefficients were significant in all the models. Residuals were more concentrated in lower diameters, implying a negative exponential DBH distribution. The Linear, Exponential and Power models performed well with the highest R^2 (up to 0.875, 0.676 and 0.655). The developed models are applicable across species groups and not across species.

延伸閱讀