Environmental monitoring is essential to assess and, if possible, anticipate the consequences of various marine economic developments. This study describes progress in environmental monitoring by developing and applying a eutrophication index (EI) for marine protected areas (MPAs). The EI combines available data, such as biological oxygen demands, dissolved inorganic nitrogen and phosphorus, and chlorophyll-a, with the weighting factors calculated from principal component analysis to assess environmental quality. Its effectiveness was tested using nearly three decades of environmental data (since 1996) from the Nha Trang MPA in Vietnam. The EI revealed clear trends in environmental quality. In the period 1996–2006, environmental conditions deteriorated, negatively impacting aquaculture. In the later period, 2007–2024, improved environmental protection policies, technological developments, expanding tourism, and heightened public awareness contributed to a reversal of this trend. During the earlier period, the EI indicated poor environmental quality (Level V), while in the later years, it improved significantly, approaching Level II. This study also identified the spatial eutrophication patterns and helped to determine the causes of specific eutrophication levels. These included port development, aquaculture activities, and domestic waste discharge. These findings highlight the close relationship between environmental quality and economic activities in the bay. Overall, the new EI and its sensitivity maps enhance environmental monitoring capabilities. They provide valuable tools for decision-makers, aiding in the strategic planning of marine economic development, ecosystem protection, and sustainable resource use. The approach supports long-term environmental stewardship and more informed, adaptive management of coastal and marine areas.

Coastal water quality monitoring is critical for ecosystem health and regulatory compliance under Vietnam’s QCVN 10:2023/BTMT standards. Thus, this study addressed the limitations of traditional predicting models by proposing a hybrid framework that integrate Hidden Markov Models (HMM, dual-branch Long Short-Term Memory (LSTM) networks, and attention mechanisms for multi class water quality classification. The model was developed using 1,305 daily observations from 2020-2023 of five water parameters – TSS, pH, TPH, Total Coliform, and DO – at the Sao Den-Ben Dinh station positioned in the east coast of Ho Chi Minh City. The hybrid architecture achieved a moderate accuracy of 74.87% and F1-score of 64.11%, representing a statistically significant improvement over a standalone HMM (44%), LSTM (68.2%), and ARIMA (70,77%). Attention weights identified a one-to-three-day lead time critical for predicting pollutions events. However, a significant limitation remains in a minority class detection, where the “Poor” category reached only 14% recall due to a severe 7:5:1 data imbalance. Generally, this framework provides an interpretable, regulatory-aligned decision-support tool for early-warning systems, offering a theoretically principled approach to capturing multi-scale temporal dynamics in coastal environments.

This research develops a vigorous approach to estimate Chlorophyll-a (Chl-a) concentration in the dynamic, optically complex waters (or Case 2 water including coastal waters, estuaries and inland water bodies) of Ganh Rai Bay, Vietnam by leveraging time series harmonized Landsat and Sentinel-2 (TM-HLS) imagery. One of the fundamental challenges while conducting this method to compute Chl-a signal is noise caused by suspended sediments (TSS) and coloured dissolved organic matter (CDOM). Thus, the study utilizes spectral features (such as Blue/Green and Red/Green ratios) to correct TSS and CDOM interference, a critical step for Case 2 waters. Otherwise, by learning spectral correlation between different bands of the TM-HLS products and in-situ Chl-a lab-computed data, several log-transformed algorithms of selected bands were investigated to examine their efficiency in estimating Chl-a content. Results showed that the log-level regression model was the most effective which yielded a high coefficient of determination (R2 = 0.888) and a minimal standard error (SE = 0.06). Furthermore, the spatial distribution analysis, utilizing the log-level model, revealed that the Chl-a concentration was highly variable in coastal area (13-15mg/m3) due to river discharge and the semi-diurnal tidal regime, but more stable offshore. The HLS data set is confirmed to be effective for continuous spatiotemporal water quality assessment.