Trust-based anomaly detection of dendrometer and environmental parameter measurements of eucalypts in South Africa and Portugal

Environmental monitoring systems for forest management face significant challenges in ensuring data reliability and detecting anomalies across heterogeneous sensor networks. Traditional approaches struggle with sensor degradation, intermittent connectivity, and environmental variability, particularly in remote eucalyptus plantations where infrastructure is limited and conditions are dynamic. These challenges are compounded by the critical nature of timely and accurate data for sustainable forestry decisions, where short-term gaps reduce model reliability and hinder the detection of stress patterns. Existing anomaly detection approaches in this domain either focus solely on reconstruction-based detection or rely on outlier scoring without incorporating trust evaluation, leading to limited robustness against sensor faults, low temporal stability, and poor generalisation across heterogeneous sensor networks. This paper introduces a novel hybrid system that integrates machine learning-based trust management with an enhanced autoencoder architecture for improved anomaly detection called Trust and Reconstruction-based Anomaly Detection (TRAD). TRAD leverages a unique dataset from sensor networks deployed across South Africa and Portugal. The architecture combines a feedforward neural network (FFN) with time decay capabilities and structured dimensionality reduction, enabling the simultaneous evaluation of trust and reconstruction-based fault identification. Experimental results demonstrate superior performance over traditional methods, achieving F1-scores up to 89% with strong ROC values exceeding 0.93 and a Temporal Stability Index (TSI) of 0.95 under diverse conditions. The system effectively handles simulated sensor degradation, missing data, and environmental variation. This work contributes to both the monitoring literature and machine learning applications in resource-constrained settings, with direct implications for sustainable forestry management and precision agriculture.