Long-run forecasting surface and groundwater dynamics from intermittent observation data: An evaluation for 50 years.

The accurate prediction of water dynamics is critical for operational water resource management. In this study, we propose a novel approach to perform long-term forecasts of daily water dynamics, including river levels, river discharges, and groundwater levels, with a lead time of 7-30 days. The approach is based on the state-of-the-art neural network, bidirectional long short-term memory (BiLSTM), to enhance the accuracy and consistency of dynamic predictions. The operation of this forecasting system relies on an in-situ database observed for over 50 years with records gauging in 19 rivers, the karst aquifer, the English Channel, and the meteorological network in Normandy, France. To address the problem of missing measurements and gauge installations over time, we developed an adaptive scheme in which the neural network is regularly adjusted and re-trained in response to changing inputs during a long operation. Advances in BiLSTM with extensive learning past-to-future and future-to-past further help to avoid time-lag calibration that simplifies data processing. The proposed approach provides high accuracy and consistent prediction for the three water dynamics within a similar accuracy range as an on-site observation, with approximately 3 % error in the measurement range for the 7 day-ahead predictions and 6 % error for the 30 d-ahead predictions. The system also effectively fills the gap in actual measurements and detects anomalies at gauges that can last for years. Working with multiple dynamics not only proves that the data-driven model is a unified approach but also reveals the impact of the physical background of the dynamics on the performance of their predictions. Groundwater undergoes a slow filtration process following a low-frequency fluctuation, favoring long-term prediction, which differs from other higher-frequency river dynamics. The physical nature drives the predictive performance even when using a data-driven model.

Spatial and temporal variability of water quality of an urbanized river in Algeria: the case of Soummam Wadi.

Spatial and temporal variations of water quality were investigated at four sites of an urbanized river in Algeria during a period of low water level in the years 2002, 2003, and 2004. Physical-chemical parameters (temperature, pH, conductivity, dissolved oxygen, suspended matter, chemical oxygen demand [COD], and 5-day biochemical oxygen demand [BOD5]) were measured. The Soummam River showed a strong pollutant load, which was organic in origin and expressed by mean concentrations in suspended matter, COD and BOD5 exceeding 150, 100, and 50 mg/L, respectively. The spatial variation highlighted two areas--(1) the first one gathers the upstream and central sites of the river, and (2) the second one is found downstream. In the downstream area, the pollutant load is almost twice as high as in the first area and, the percent saturation of dissolved oxygen is relatively weak (< 55%). This load is the result of the significant volume of urban and industrial emissions in the river, the high temperature during low-water-level periods, and flood events, which occurred just before the period of low water level. The Soummam River was classified according to the criteria of appreciation of surface water and was found to be extremely polluted. This work is one of the first studies on the quality of rivers in Algeria. This research will be useful as a first step for future works in North Africa and will add to knowledge on the water quality in the Mediterranean Basin.

Proton NMR spectroscopy assignment of D-glucose residues in highly acetylated starch.

1H NMR spectroscopy assignments have been obtained for starch acetates using COSY and HOHAHA experiments by comparison with the spectra of amylose triacetate and of peracetylated malto-oligosaccharides (maltotriose, maltotetraose, maltoheptaose). These assignments are valuable for the location and evaluation of the substitution pattern in modified starches. The bulk of the 1H NMR spectra of highly acetylated starch strongly resembles the spectrum of amylose triacetate in which all protons are identified and display distinct chemical shifts. The resolving power of the HOHAHA experiment allowed the distinction of minor spin systems. Beside these strong signals pertaining to an average 2.3.6-tri-O-acetyl-alpha-(1-->4) linked D-glucopyranose unit in an infinite chain, the combination of COSY and HOHAHA experiments allowed the identification of these systems to the terminal, n-1, n-2, and to partially acetylated glucopyranosyl units. As an example, two different preparations of starch acetates with degrees of substitution 2.74 and 2.63 were examined. In one case, NMR demonstrates that the defects of acetylation are random on the polymeric chain (with corresponding signals for unacylated secondary hydroxyl positions at delta 3.61 and 3.40) while in another case, these signals are not detectable, probably due to the presence of clusters of non-acetylated residue forming solid-like zones.