Fruit-based allometry of Strychnos madagascariensis and S. spinosa (Loganiaceae) in the savannah woodlands of the Umhlabuyalingana municipality, KwaZulu-Natal, South Africa

Savannah woodlands of South Africa are dominated by fruit tree species that have a potential for commercial harvest from the wild. However, allometric equations that enable the quantification of fruit biomass of trees are non-existent. The aim of this study was to develop fruit-based allometric equations for Strychnos madagascariensis and S. spinosa species. A total of 80 trees were selected by applying a stratified sampling approach according to four stem diameter classes during fruit ripening period. For each tree, the following parameters were measured: fruit biomass, diameter at breast height (DBH), canopy diameter, and total height. Six forms of the allometric models were fitted to the data using ordinary least squares method. The Akaike information criterion was used to select the best models and the Root Mean Squared Error (RMSE) was used to evaluate the quality of the predictions. DBH was the only appropriate variable in the prediction of the fruit biomass and explained 99.9% of the variation in fruit biomass. The simple linear regressions linking the DBH (in cm) to the total fresh fruit biomass (FB; in kg) were the best models and were expressed by (1) FB = 1.0243 xDBH(1.1841); and (2) FB = 1.0297 xDBH(1.1956); respectively for Strychnos madagascariensis and Strychnos spinosa. These equations provided realistic predictions of fresh fruit biomass. They induced on average a prediction error of 5.4 kg on the total fresh fruit biomass of a tree. Larger trees (DBH > 25 cm for S. madagascariensis; DBH > 11 cm for S. spinosa) were more susceptible to fruit biomass prediction errors than smaller trees. This study showed that simple linear regression basing on DBH is the best approach to estimate fresh fruit biomass of trees in savannah woodlands.