Near-term forecasts of stream temperature using deep learning and data assimilation in support of management decisions

Deep learning (DL) models are increasingly used to make accurate hindcasts of management-relevant variables, but they are less commonly used in forecasting applications. Data assimilation (DA) can be used for forecasts to leverage real-time observations, where the difference between model predictions and observations today is used to adjust the model to make better predictions tomorrow. In this use case, we developed a process-guided DL and DA approach to make 7-day probabilistic forecasts of daily maximum water temperature in the Delaware River Basin in support of water management decisions. Our modeling system produced forecasts of daily maximum water temperature with an average root mean squared error (RMSE) from 1.1 to 1.4°C for 1-day-ahead and 1.4 to 1.9°C for 7-day-ahead forecasts across all sites. The DA algorithm marginally improved forecast performance when compared with forecasts produced using the process-guided DL model alone (0%–14% lower RMSE with the DA algorithm). Across all sites and lead times, 65%–82% of observations were within 90% forecast confidence intervals, which allowed managers to anticipate probability of exceedances of ecologically relevant thresholds and aid in decisions about releasing reservoir water downstream. The flexibility of DL models shows promise for forecasting other important environmental variables and aid in decision-making.     

Dissolved organic sulfur in streams draining forested catchments in southern China

Dissolved organic sulfur (DOS) is an important fraction for sulfur mobilization in ecosystem. In this work stream waters were sampled in 25 forested sites in southern China to study the dissolved sulfur fractions. Dissolved sulfur was fractionated into dissolved organic sulfur (DOS) and inorganic sulfate (SO₄^2-) for 95 stream water samples. The results showed that the concentration of DOS ranged from 0 to 13.1 mg/L (average 1.3 mg/L) in all the streams. High concentrations of DOS in stream waters were found in the sites with high concentrations of sulfate. DOS constituted less than 60.1% of dissolved sulfur (average 17.9%). Statistical analysis showed that DOS concentration was correlated with SO₄^2- in streams waters and total sulfur in surface layer soils. The results also showed that DOS concentration in stream waters had a seasonal variation, but no trends were found with it. The implication was that the long term sulfur deposition had led the increase of the concentration and fraction of DOS in stream waters in acid rain prevailing regions     

Detention Outlet Retrofit Improves the Functionality of Existing Detention Basins by Reducing Erosive Flows in Receiving Channels

By discharging excess stormwater at rates that more frequently exceed the critical flow for stream erosion, conventional detention basins often contribute to increased channel instability in urban and suburban systems that can be detrimental to aquatic habitat and water quality, as well as adjacent property and infrastructure. However, these ubiquitous assets, valued at approximately $600,000 per km² in a representative suburban watershed, are ideal candidates to aid in reversing such cycles of channel degradation because improving their functionality would not necessarily require property acquisition or heavy construction. The objective of this research was to develop a simple, cost‐effective device that could be installed in detention basin outlets to reduce the erosive power of the relatively frequent storm events (~ < two‐year recurrence) and provide a passive bypass to maintain flood control performance during infrequent storms (such as the 100‐year recurrence). Results from a pilot installation show that the Detain H₂O device reduced the cumulative sediment transport capacity of the preretrofit condition by greater than 40%, and contributed to reduced flashiness and prolonged baseflows in receiving streams. When scaling the strategy across a watershed, these results suggest that potential gains in water quality and stream channel stability could be achieved at costs that are orders of magnitude less than comparable benefits from newly constructed stormwater control measures.     

Diverging response patterns of terrestrial and aquatic species to hydromorphological restoration

Although experiences with ecological restoration continue to accumulate, the effectiveness of restoration for biota remains debated. We complemented a traditional taxonomic analysis approach with information on 56 species traits to uncover the responses of 3 aquatic (fish, macroinvertebrates, macrophytes) and 2 terrestrial (carabid beetles, floodplain vegetation) biotic groups to 43 hydromorphological river restoration projects in Germany. All taxonomic groups responded positively to restoration, as shown by increased taxonomic richness (10–164%) and trait diversity (habitat, dispersal and mobility, size, form, life history, and feeding groups) (15–120%). Responses, however, were stronger for terrestrial than aquatic biota, and, contrary to our expectation, taxonomic responses were stronger than those of traits. Nevertheless, trait analysis provided mechanistic insights into the drivers of community change following restoration. Trait analysis for terrestrial biota indicated restoration success was likely enhanced by lateral connectivity and reestablishment of dynamic processes in the floodplain. The weaker response of aquatic biota suggests recovery was hindered by the persistence of stressors in the aquatic environment, such as degraded water quality, dispersal constraints, and insufficient hydromorphological change. Therefore, river restoration requires combined local- and regional-scale approaches to maximize the response of both aquatic and terrestrial organisms. Due to the contrasting responses of aquatic and terrestrial biota, the planning and assessment of river restoration outcomes should consider effects on both components of riverine landscapes.     

巢湖流域河流鱼类群落的时空分布

基于2013年4月和10月对巢湖流域66个河道样点的调查数据,初步研究了巢湖流域河流鱼类群落的时空变化特征。主要研究结果显示,鱼类多样性无显著性的水系间、生态分区间的变化,但随季节和河流级别显著变化:10月份的个体数显著高于4月份,2级河流的物种数和个体数均显著大于1级、3级和4级河流。鱼类群落结构的季节动态显著,随生态分区显著变化,但不受水系、河流级别的显著影响。在二级生态分区水平上,仅西南森林生态亚区的鱼类群落结构与其他5个生态亚区的显著差异;宽鳍鱲、吻虾虎鱼等在西南森林生态亚区具有更高多度,而鲫、鰐、鲤等物种在其他生态亚区多度更高。     

Hot,Salty Water: A Confluence of Issues in Managing Stormwater Runoff for Urban Streams

Research increasingly highlights cause and effect relationships between urbanization and stream conditions are complex and highly variable across physical and biological regions. Research also demonstrates stormwater runoff is a key causal agent in altering stream conditions in urban settings. More specifically, thermal pollution and high salt levels are two consequences of urbanization and subsequent runoff. This study describes a demonstration model populated with data from a high gradient headwaters stream. The model was designed to explain surface water‐groundwater dynamics related to salinity and thermal pollution. Modeled scenarios show long‐term additive impacts from salt application and suggest reducing flow rates, as stormwater management practices are typically designed to do, have the potential to greatly reduce salt concentrations and simultaneously reduce thermal pollution. This demonstration model offers planners and managers reason to be confident that stormwater management efforts can have positive impacts.     

Recent Changes in Stream Flashiness and Flooding,and Effects of Flood Management in North Carolina and Virginia

The southeastern United States has undergone anthropogenic changes in landscape structure, with the potential to increase (e.g., urbanization) and decrease (e.g., reservoir construction) stream flashiness and flooding. Assessment of the outcome of such change can provide insight into the efficacy of current strategies and policies to manage water resources. We (1) examined trends in precipitation, floods, and stream flashiness and (2) assessed the relative influence of land cover and flow‐regulating features (e.g., best management practices and artificial water bodies) on stream flashiness from 1991 to 2013. We found mean annual precipitation decreased, which coincided with decreasing trends in floods. In contrast, stream flashiness, overall, showed an increasing trend during the period of study. However, upon closer examination, 20 watersheds showed stable stream flashiness, whereas 5 increased and 6 decreased in flashiness. Urban watersheds were among those that increased or decreased in flashiness. Watersheds that increased in stream flashiness gained more urban cover, lost more forested cover and had fewer best management practices installed than urban watersheds that decreased in stream flashiness. We found best management practices are more effective than artificial water bodies in regulating flashy floods. Flashiness index is a valuable and straightforward metric to characterize changes in streamflow and help to assess the efficacy of management interventions.     

温榆河氨氮污染控制措施的效果模拟

以城市退水为主要补给水源的北京市温榆河呈现以氨氮(18.68 mg·L-1)污染为主的特征.针对温榆河的水质改善需求,本研究以氨氮为目标污染物,采用QUAL2K河流水质模型和情景分析方法,对温榆河污染防治与水质改善措施的效果进行量化评估.模拟结果表明,提高污水厂的处理量和排放标准是改善温榆河水质的关键,可降低温榆河北关闸断面氨氮浓度29.66％;建设旁路离线河流净化器可削减5.57％的氨氮污染,是温榆河水质改善的有效措施.本研究提出的包括改善上游来水水质、干流截污、支流污染控制、污水厂提标改造、建设旁路离线河流净化器和生态河道等措施在内的综合实施方案,可使得温榆河的水质全程达到地表水V类标准.     

大型底栖动物生物评价指数比较与应用

采用不同生物指数评价河流健康,结果往往有差异,如何识别各种生物指数的关系和适用性成为亟待解决的问题.基于太子河流域大型底栖动物群落研究,比较了5种大型底栖动物生物指数评价结果的差异.结果表明,5种指数具有极显著的相关性,但由于健康等级划分标准差异,造成不同指数的健康评价等级存在一定差异.同时分析了不同指数对不同类型人为活动影响的敏感性,研究其在河流健康评价中的适用性.结果表明在太子河流域中,BI指数与土地利用类型和溶氧具有较好的相关关系,对这两类的干扰具有较好的指示作用;FBI指数对酸污染和氨氮污染具有较强的指示作用;ASPT指数与耗氧相关的水质指标有较强的负相关;B-IBI指数与总氮呈显著的负相关关系,可较好指示氮污染,而且B-IBI指数与其他类型人为干扰活动均呈显著的相关关系,对土地利用和水质污染也具有较好的指示作用.综上所述,大型底栖动物BI指数和ASPT指数应该分别适用于评价流域土地利用和水化学指标对河流生态系统的影响,而B-IBI可用于评估多种类型人为干扰活动.     

超高速瞬态测试系统软硬件构架设计研究

目的研究采样频率为2~10 MHz的64通道超高速同步瞬态测试系统的设计技术,实现两类典型超高速瞬态测试系统的硬件架构设计与软件架构设计。方法一类采用PXI-Express高带宽总线和高速RAID磁盘阵列架构构建持续流盘存储的连续高速测试系统,另一类是采用大容量板载数据缓存和PXI总线事后下载传输数据的架构构造高速测试系统。在高性能测试软件设计方面,主要应用生产者/消费者结构与有限状态机相结合的软件架构进行高性能测试系统软件设计。结果目前64通道下基于持续流盘架构的测试系统受数据记录的速度限制系统最高采样频率仅达2.5 MHz,而基于板载缓存数据与PXI总线事后下载数据架构的测试系统最高采样频率可达10 MHz,测试时长可达5 s。结论当前两类架构的测试系统均可满足超高速瞬态测试需求,设计时需根据需求的最高采样频率决定使用的架构形式。