Highlights A 10-year dataset of time-series MODIS imagery and in situ Chl- a concentration were curated for Lake Okeechobee. LSTM significantly outperformed KNN, SVR, and RF for Chl- a prediction and subsequent HAB detection. The optimal window length was found to be 13 days with a 4-day temporal resolution for the LSTM model. KNN, SVR, and RF models were not effective at utilizing the temporal dynamics of the input features. Abstract. Harmful algal blooms (HABs) in inland waterbodies are a global concern due to their negative impact on human, animal, and ecosystem health. Chlorophyll-a (Chl-a) concentration is an important water quality parameter for monitoring HABs. While statistical and machine learning (ML) models have been widely studied to predict Chl-a concentration and HABs based on single-time-point satellite data, this work assessed whether long short-term memory (LSTM) can improve both tasks by leveraging temporal features in time-series MODIS satellite images compared to three classical ML models, including k-nearest neighbor (KNN), support vector regression (SVR), and random forest (RF). A dataset of daily MODIS images and monthly in situ Chl-a concentration measurements from 2011 to 2020 was curated for Lake Okeechobee, Florida. A window size of 13 days with a temporal resolution of four days was found to produce the optimal performance for LSTM, which significantly outperformed KNN, SVR, and RF for Chl-a prediction with a root mean square error of 11.95 µg/L, a mean absolute error of 8.55 µg/L, and a R 2 value of 0.43. The superior performance of LSTM for Chl-a prediction was likely due to its ability to leverage the temporal dynamics in the features associated with HAB development. The Chl-a predictions were further used to determine HAB events, showing better accuracy and a significantly higher F1 score for LSTM over the other models. The study suggested that combining LSTM with high-temporal-resolution time-series data should be preferred over applying common ML models on time-series or single-time-point remote sensing data for Chl-a and HAB monitoring. Keywords: Cyanobacteria, LSTM, Machine Learning, Remote Sensing, Water Quality.
Roadside vegetation composition assessment is essential to maintain ecological stability, control invasive species, and ensure the adoption of environmental regulations in areas surrounding active roadside construction zones. Traditional monitoring methods involving visual inspections are time-consuming, labor-intensive, and not scalable. Remote sensing offers a valuable alternative to automating large-scale vegetation assessment tasks efficiently. The study compares the performance of proximal RGB imagery processed using deep learning (DL) techniques against the vegetation indices (VIs) extracted at higher altitudes, establishing a foundation to use the prior in performing vegetation analysis using unmanned aerial vehicles (UAVs) for broader scalability. A pixel-wise annotated dataset for eight roadside vegetation species was curated to evaluate the performance of multiple semantic segmentation models in this context. The best-performing MAnet DL achieved a mean intersection over union of 0.90, highlighting the model’s capability in composition assessment tasks. Additionally, in predicting the vegetation cover—the DL model achieved an R2 of 0.996, an MAE of 1.225, an RMSE of 1.761, and an MAPE of 3.003% and outperformed the top VI method of SAVI, which achieved an R2 of 0.491, an MAE of 20.830, an RMSE of 23.473, and an MAPE of 59.057%. The strong performance of DL models on proximal RGB imagery underscores the potential of UAV-mounted high-resolution RGB sensors for automated roadside vegetation monitoring and management tasks at construction sites.