An overview of patterns and dynamics of suspended sediment transport in an agroforest headwater system in humid climate: Results from a long-term monitoring

Predicting suspended sediment load in Peninsular Malaysia using support vector machine and deep learning algorithms

High loads of suspended sediments in rivers are known to cause detrimental effects to potable water sources, river water quality, irrigation activities, and dam or reservoir operations. For this reason, the study of suspended sediment load (SSL) prediction is important for monitoring and damage mitigation purposes. The present study tests and develops machine learning (ML) models, based on the support vector machine (SVM), artificial neural network (ANN) and long short-term memory (LSTM) algorithms, to predict SSL based on 11 different river data sets comprising of streamflow (SF) and SSL data obtained from the Malaysian Department of Irrigation and Drainage. The main objective of the present study is to propose a single model that is capable of accurately predicting SSLs for any river data set within Peninsular Malaysia. The ANN3 model, based on the ANN algorithm and input scenario 3 (inputs consisting of current-day SF, previous-day SF, and previous-day SSL), is determined as the best model in the present study as it produced the best predictive performance for 5 out of 11 of the tested data sets and obtained the highest average RM with a score of 2.64 when compared to the other tested models, indicating that it has the highest reliability to produce relatively high-accuracy SSL predictions for different data sets. Therefore, the ANN3 model is proposed as a universal model for the prediction of SSL within Peninsular Malaysia.

20-Years of hindsight into hydrological dynamics of a mountain forest catchment in the Central Spanish Pyrenees

Mediterranean mountain forests play a significant role in hydrological regulation. In this study, hydrological dynamics was examined at different temporal scales in a small mountain forest catchment in the Central Spanish Pyrenees (San Salvador), based on a 20-year dataset (1999-2019). Mean annual runoff coefficient is 0.21, and ranged from 0.02 to 0.58. The catchment has a bi-modal hydrological behavior with two hydrological periods: a dry-period between July and December, and a wet-period between January and June. During the study period, only 108 floods were recorded, suggesting a low responsiveness of the catchment, with a high variable response. Spearman correlation analysis and stepwise multivariate regression suggest that the hydrological response in the San Salvador catchment is mainly depending on water table, with antecedent moisture conditions and rainfall depth as secondary factors. Seasonal differences were also observed: during dry season, the response was mainly related to rainfall depth and rainfall intensity; in contrast in wet season, the response was mainly related to antecedent conditions (previous rainfall and base flow). Thus, the already challenging water resources management in the Mediterranean basin is magnified by the key function of forests as natural modulators of water cycle. Consequently, the study of natural forested catchments is needed and long-datasets have to be analysed to understand the role of natural Mediterranean forest in the hydrological dynamics and its evolution and adaptation in a context of Global Change.

Strategies for identifying pollution sources in a headwater catchment based on multi-scale water quality monitoring

Rural headwater catchments are important to describe the connectivity of pollution sources to water bodies. Strategies to optimize water quality monitoring networks, as parameter definition, sampling, and statistical approach, have been widely discussed. The objectives of this study were to describe the spatial and temporal dynamics (intra- and inter-events) of water quality and to establish its implications for environmental monitoring programs. The monitoring was carried out in a rural headwater catchment (1.2 km²) with shallow soils, high slopes, and intense agricultural activity in Southern Brazil. To better describe the impact of agriculture on water resources, the monitoring strategy was based on definition of the best set of parameters and different sampling frequency to incorporate intra- and inter-event variability and statistical analysis approach. We also analyzed parameters in different sub-basins with physiographic traits. Three hydrological compartments were analyzed: surface flow, groundwater, and base flow. Physico-chemical parameters, the concentration of elements associated with agricultural activity, and biological parameters were evaluated. Total phosphorus and turbidity were the parameters most affected by agricultural activity. They reflected on the inter- and intra-events, the impacts of soil and water degradation by agricultural activity, and the precarious rural sanitation conditions. Spatiotemporal variability of the parameters characterizes the different mechanisms for transferring pollutants from diffuse sources to water bodies. Spatial and temporal patterns in water quality changes were used to discuss environmental monitoring strategies, such as parameter and sampling frequency definition, to improve soil and water conservation programs at the catchment scale.