基于ANFIS改进的大气腐蚀环境缺失数据填补方法

目的针对大气腐蚀中重要环境数据缺失的复杂问题,提出一种相关因素(Relevance Factors)和自适应神经模糊推理系统(Adaptive Neuro-Fuzzy Inference System)结合的方法(RF-ANFIS)对缺失数据进行填补。方法首先采用相关因素方法计算缺失数据和多项环境因素间的相关程度,筛选出相关系数较大的因子,然后应用ANFIS构建缺失数据与所选环境因子的关系模型。最后以二氧化硫数据为具体对象,采用北京2015年的气象数据对所建立的模型进行检验。结果经过改进的RF-ANFIS模型在最优情况下样本均方误差为0.696,在14个测试样本中有13个相对误差在20%以内,针对有限样本的数据分析中更为适用。结论该方法有效提高了大气腐蚀环境数据缺失的填补精度,对在数据缺失情况下预测大气腐蚀速率具有重要意义。     

水质自动监测站数据质控研究

随着环境监测技术的不断发展,水质自动监测以其高效、精确、连续监测等诸多优点,正逐步成为当今水环境监测的主流.然而中国水质自动监测技术起步较晚,且涉及的领域较多,很多关于水质自动站数据质控的相关标准不能及时出台,而且各个地方的河流分布、水文情况及水质自动站建设情况都不尽相同,这就需要对原有的手工监测标准进行比较和分析,同时结合对水质自动站运维管理的经验,找出适合目前水质自动站运行管理的质控手段.     

基于博弈论的环保数据质量影响因素分析研究

污染源产排污数据是一切环保工作的基础,确保真实可信具有十分重要的意义.为此,分析了影响污染源产排污数据质量的主要因素:受经济利益驱动的自身因素;受政绩利益驱动的管理因素和受地方保护驱动的政府因素等.在新环保法未得到充分贯彻,排污单位、环保主管部门和地方政府三者违法成本均低于守法成本的条件下,以博奕论为分析工具讨论了三种影响因素产生负面作用的必然性,并根据分析结果针对性地提出了提高数据质量的应对策略.     

A Nationwide Analysis of U.S. Army Corps of Engineers Reservoir Performance in Meeting Operational Targets

The United States (U.S.) Army Corps of Engineers operates reservoirs across the U.S. with 89% of reservoirs constructed prior to 1980. Many reservoirs have experienced changes in environmental conditions (e.g., climate and sediment yield) and societal conditions (e.g., water/energy demand and ecological flows) since construction. These changes may challenge the potential for reservoirs to meet their operational targets (OTs) (management goals). Historic daily reservoir data and OTs were collected for 233 reservoirs. Analyses were developed to identify when and where reservoirs may be systematically departing from OTs in terms of the frequency and magnitude of departure. Fifty‐six percent of reservoirs consistently met operating targets, 30% were borderline, and 13% experienced frequent and large magnitude departures. Fifty‐two percent of reservoirs with large departures were due to shortages and were located in the South Pacific and Southwestern divisions. This work provides a framework to identify reservoir performance in relation to management goals, a necessary step for moving toward adaptive management under changing conditions. All individual reservoir analyses are provided via an interactive data visualization tool: https://nicholasinstitute.duke.edu/reservoir-data .     

Cyberinfrastructure and Web Apps for Managing and Disseminating the National Water Model

Hydrologic modeling can be used to provide warnings before, and to support operations during and after floods. Recent technological advances have increased our ability to create hydrologic models over large areas. In the United States (U.S.), a new National Water Model (NWM) that generates hydrologic variables at a national scale was released in August 2016. This model represents a substantial step forward in our ability to predict hydrologic events in a consistent fashion across the entire U.S. Nevertheless, for these hydrologic results to be effectively communicated, they need to be put in context and be presented in a way that is straightforward and facilitates management‐related decisions. The large amounts of data produced by the NWM present one of the major challenges to fulfill this goal. We created a cyberinfrastructure to store NWM results, “accessibility” web applications to retrieve NWM results, and a REST API to access NWM results programmatically. To demonstrate the utility of this cyberinfrastructure, we created additional web apps that illustrate how to use our REST API and communicate hydrologic forecasts with the aid of dynamic flood maps. This work offers a starting point for the development of a more comprehensive toolset to validate the NWM while also improving the ability to access and visualize NWM forecasts, and develop additional national‐scale‐derived products such as flood maps.     

Evaluation of Satellite‐Derived Rainfall Data for Multiple Physio‐Climatic Regions in the Santiago River Basin,Mexico

Assessment of water resources requires reliable rainfall data, and rain gauge networks may not provide adequate spatial representation due to limited point measurements. The Tropical Rainfall Measuring Mission (TRMM) provides rainfall data at global scale, and has been used with good results. However, TRMM data are an indirect measurement of rainfall, and therefore must be validated for its proper use. In this work, a validation scheme was designed and implemented to compare the TRMM Version 7 (V7) monthly rainfall product at different time frames with data measured in two hydrologic subregions of the Santiago River Basin (SRB) in Mexico: Río Alto Santiago and Río Bajo Santiago (RBS). Additionally, three physio‐climatic regions provide an assessment of the interplay of topography, distance from coastal regions, and seasonal weather patterns on the correspondence between both datasets. The TRMM V7 rainfall product exhibited good agreement with the rain gauge data particularly for the RBS and for the whole SRB during wettest summer and autumn seasons. However, strong regional dependence was observed due to differences in climate and topography. Overall, in spite of some noted underestimations, the monthly TRMM V7 rainfall product was found to provide useful information that can be used to complement limited monitoring as is the case of RBS. An improved combined rainfall product could be generated and thus gaining the most benefits from both data sources.     

The influence of building renovation and rental housing on urban trees

Urban forest ecosystems are complex and vulnerable social–ecological systems. The relationship between urban forests and housing is particularly variable and uncertain. We examine the influence of building renovation and rental housing on public trees at the parcel and street-section scale in a residential neighbourhood in Toronto, Canada. We use empirical data describing multiple tree inventories and government open data describing building permit applications to test for effects on urban forest structure, tree mortality, and tree planting. We found that the presence and number of building permits significantly predicted mortality at both scales, while planting was positively correlated with building permits at the street-section scale only. Multi-unit parcels had significantly lower rates of planting than single-unit parcels and multi-unit housing was positively correlated with mortality at the street-section scale. These findings suggest that where concentrated changes in housing stock are occurring, substantial losses of trees and associated ecosystem services are possible.     

Disasters and ‘conditions of possibility’: rethinking causation through an analysis of earthquakes in Nepal

What causes a disaster's aftermath? Scholars have increasingly turned to historical approaches that link outcomes to pre-disaster sociopolitical dynamics. Disasters lead to ‘critical junctures’ that ‘trigger’ events that unfold in the wake of the initial phenomenon. This paper argues that the ‘critical junctures’ paradigm shares limitations with ‘path dependency’ theory from which it is derived, namely a tendency towards historicism—a functionalist teleology better able to explain continuity than change. As an alternative, this analysis draws on Michel Foucault's understanding of ‘conditions of possibility’ as a way of rethinking agency/causation, moving away from individual subjects, events, or even historical conditions towards, instead, the new, radically destabilised ‘epistemological field’ emerging in the disaster's aftermath. This paper examines a series of devastating earthquakes in Nepal to consider how post-disaster ‘epistemological fields’ present new ‘conditions of possibility’ within which new ideas, actions, and outcomes become thinkable and possible in ways that pre-disaster historical conditions could not have predicted.     

Historical spatial reconstruction of a spawning‐aggregation fishery

Aggregations of individual animals that form for breeding purposes are a critical ecological process for many species, yet these aggregations are inherently vulnerable to exploitation. Studies of the decline of exploited populations that form breeding aggregations tend to focus on catch rate and thus often overlook reductions in geographic range. We tested the hypothesis that catch rate and site occupancy of exploited fish‐spawning aggregations (FSAs) decline in synchrony over time. We used the Spanish mackerel (Scomberomorus commerson) spawning‐aggregation fishery in the Great Barrier Reef as a case study. Data were compiled from historical newspaper archives, fisher knowledge, and contemporary fishery logbooks to reconstruct catch rates and exploitation trends from the inception of the fishery. Our fine‐scale analysis of catch and effort data spanned 103 years (1911–2013) and revealed a spatial expansion of fishing effort. Effort shifted offshore at a rate of 9.4 nm/decade, and 2.9 newly targeted FSAs were reported/decade. Spatial expansion of effort masked the sequential exploitation, commercial extinction, and loss of 70% of exploited FSAs. After standardizing for improvements in technological innovations, average catch rates declined by 90.5% from 1934 to 2011 (from 119.4 to 11.41 fish/vessel/trip). Mean catch rate of Spanish mackerel and occupancy of exploited mackerel FSAs were not significantly related. Our study revealed a special kind of shifting spatial baseline in which a contraction in exploited FSAs occurred undetected. Knowledge of temporally and spatially explicit information on FSAs can be relevant for the conservation and management of FSA species.     

A spatially explicit estimate of the prewhaling abundance of the endangered North Atlantic right whale

The North Atlantic right whale (NARW) (Eubalaena glacialis) is one of the world's most threatened whales. It came close to extinction after nearly a millennium of exploitation and currently persists as a population of only approximately 500 individuals. Setting appropriate conservation targets for this species requires an understanding of its historical population size, as a baseline for measuring levels of depletion and progress toward recovery. This is made difficult by the scarcity of records over this species’ long whaling history. We sought to estimate the preexploitation population size of the North Atlantic right whale and understand how this species was distributed across its range. We used a spatially explicit data set on historical catches of North Pacific right whales (NPRWs) (Eubalaena japonica) to model the relationship between right whale relative density and the environment during the summer feeding season. Assuming the 2 right whale species select similar environments, we projected this model to the North Atlantic to predict how the relative abundance of NARWs varied across their range. We calibrated these relative abundances with estimates of the NPRW total prewhaling population size to obtain high and low estimates for the overall NARW population size prior to exploitation. The model predicted 9,075–21,328 right whales in the North Atlantic. The current NARW population is thus <6% of the historical North Atlantic carrying capacity and has enormous potential for recovery. According to the model, in June–September NARWs concentrated in 2 main feeding areas: east of the Grand Banks of Newfoundland and in the Norwegian Sea. These 2 areas may become important in the future as feeding grounds and may already be used more regularly by this endangered species than is thought.