Evaluating operational reliability of smartphone-based measurement systems for forest volume estimation under measurement uncertainty

Abstract Smartphone applications offer considerable potential for rapid forest inventory, but their reliability for plot-level volume estimation remains uncertain. Therefore, this study evaluated how smartphone-derived diameter at breast height (DBH) and tree height measurements influence model-derived plot-level volume estimates in a peri-urban forest in central Nepal. A total of 668 trees across 75 plots were measured using conventional methods and smartphone applications on Android and iOS devices. Species-specific allometric equations developed for Nepal were used to estimate volume under two scenarios: (A) smartphone-derived DBH combined with manual height and (B) smartphone-derived DBH and height. Mixed-effects models were used to evaluate tree- and plot-level volume errors while accounting for volume-dependent heteroskedasticity. Smartphone DBH measurements were generally less variable than height measurements, but systematic DBH errors could still propagate into appreciable plot-level volume discrepancies. Under Scenario A, the Android configuration showed a plot-level bias of − 7.06% despite using manual height, whereas iPhone-based Arboreal Tree Height (ATH) and ForestScanner produced smaller biases ranging from − 0.94% to 4.34%. Incorporating smartphone-derived height generally increased errors, with Android showing the largest discrepancy under Scenario B (bias = − 16.54%, MAE = 53.87 m³ ha⁻¹). These findings demonstrate that even relatively modest systematic DBH errors can propagate into appreciable plot-level volume discrepancies, while errors in smartphone-derived height can further increase these discrepancies. Further validation across diverse forest and terrain conditions, devices, and operators is needed before broader operational application.