基于深度神经网络算法的水体透明度反演方法

水体透明度能够直观反映湖泊水质状态,掌握长时间大尺度湖泊水体透明度是控制和改善湖泊水生态环境的关键。由于滇池的水质原位监测工作起步较晚,导致长时间序列的历史湖泊水体透明度数据的缺失。为此,以滇池为研究区,以深度神经网络算法为理论基础,以原位监测和MODIS遥感影像为数据,对2001年1月1日—2018年12月31日滇池水体透明度进行反演,并利用地理空间分析方法探讨了滇池湖泊水体透明度时空变化特征。研究结果表明:(1)提出的反演模型具有较好的性能(RMSE=0.1359,MAE=0.1134),能够客观反映湖泊水体透明度状况;(2)时间变化特征分析结果表明,滇池水体透明度总体呈现下降趋势,综合变化率为-0.08 m/10 a;(3)空间变化特征分析结果表明,水体透明度较高的区域下降率较大,水体透明度较低的区域变化趋势相对稳定,距离城区及居民区较近的水体透明度相对较低;人类活动将成为影响滇池水体透明度变化的重要因素,同时也是造成滇池水体污染的主要因素。     

基于小波理论的干旱区内陆湖泊叶绿素a的TM影像遥感反演

叶绿素a(Chl-a)是衡量湖泊富营养化的重要指标,利用遥感技术动态监测面积较大的湖区水体中Chl-a浓度对了解湖区水质具有重要意义。以内蒙古乌梁素海为例,提出利用TM影像中的水体实测光谱进行小波去噪和光谱信号重构,并结合水质采样实测数据进行神经网络拟合,建立光谱反射率比值与Chl-a浓度的反演模型的方法。结果显示:小波理论和神经网络相结合的模型可以适用于估算乌梁素海Chl-a浓度,去噪后Chl-a浓度与光谱信号的相关系数(-0.575)较去噪前(-0.417)明显增强,去噪后的采样点光谱信号与Chl-a浓度之间表现出比原始信号更强的负相关性,证明了去噪后的观测值可进一步减弱随机误差的干扰和去除噪声,使观测数据更加逼近Chl-a浓度的真实情况,图像去噪重构结果显示重构后的光谱范围较之前有所缩窄,部分信号点得到了增强,但基本剖面结构并没有产生较大变化,反演模型的平均相对误差为0.142,与其他研究相比差别不大。反演得出的乌梁素海Chl-a浓度分布反映了污染源的分布,同时说明了乌梁素海Chl-a浓度在时空分布上呈现一定的差异,表现为丰水期呈现浅水区Chl-a浓度值高于湖心区,来水区高于其他湖区的分布趋势,枯水期乌梁素海中部呈现由西向东Chl-a浓度逐步降低的分布规律,西部呈均一化分布。反演模型基本可以满足实际预测的需要。但模型在具体应用中在影像数据采集、数据量及算法方面还有很大的改进空间,该方法的提出为干旱区大型内陆水体富营养化的实时定量遥感监测提供了新的解决方案。     

Remote sensing for lake research and monitoring – Recent advances

Lakes are important ecosystems providing various ecosystem services. Stressors such as eutrophication or climate change, however, threaten their ecological functions. National and international legislations address these threats and claim consistent, long-term monitoring schemes. Remote sensing data and products provide synoptic, spatio-temporal views and their integration can lead to a better understanding of lake ecology and water quality. Remote sensing therefore gains increasing awareness for analysing water bodies. Various empirical and semi-analytical algorithms exist to derive remote sensing indicators as proxies for climate change or ecological response variables. Nevertheless, most monitoring networks lack an integration of remote sensing data. This review article therefore provides a comprehensive overview how remote sensing can support lake research and monitoring. We focus on remote sensing indicators of lake properties, i.e. water transparency (suspended particulate matter, coloured dissolved organic matter, Secchi disc depth, diffuse attenuation coefficient, turbidity), biota (phytoplankton, cyanobacteria, submerged and emerged aquatic vegetation), bathymetry, water temperature (surface temperature) and ice phenology (ice cover, ice-on, ice-out). After a brief background introducing principles of lake remote sensing we give a review on available sensors and methods. We categorise case studies on remote sensing indicators with respect to lake properties and processes. We discuss existing challenges and benefits of integrating remote sensing into lake monitoring and ecological research including data availability, ready-to-use tools and accuracies.     

洱海动态水环境容量模拟研究

基于环境流体动力学模型EFDC (Environmental Fluid Dynamic Code),建立了洱海湖泊及湖湾的三维水动力水质模型。并利用2001－2010 年的气象水文数据对洱海的水动力水质状况进行了连续10 年的数值模拟,模拟结果与实测结果吻合度较好,表明水动力水质模型计算结果较为合理。应用建立的数值模型分析了洱海的水流、水位和水质的时空变化特征,并在此基础上计算分析主要水质因子的动态水环境容量。结果表明,若以地表水二类水标准为规划目标,洱海总氮(TN)、总磷(TP)、COD_Mn和BOD₅ 的平均环境容量分别为1 150、72、12 780 和10 093 吨/年。洱海三维水动力水质模型的建立,为洱海的水环境保护工作提供了一个重要的管理和分析技术平台,对于洱海地区的污染物总量控制具有重要的意义。     

应用多种生物类群评价白洋淀水环境质量变化

白洋淀是华北平原最大的浅水湖泊。近几十年来,白洋淀上游耗水增加,天然入淀水量急剧减少,使白洋淀水位明显降低;同时,随着淀泊周边农村生活污水的排入,白洋淀发生了严重的富营养化。2017年雄安新区设立,白洋淀环境治理工作的重要性空前提高。近几年通过上游水库和跨流域生态补水,白洋淀水位明显抬升,同时淀中和淀边农村实施了生活污水收集。为了解白洋淀当前水环境状况,对前期补水和治污的效果进行评估,2019年8月对白洋淀水体理化指标及水生生物进行调查,采用水质、浮游植物、浮游动物、底栖动物指标评价白洋淀水环境质量,并与2010年的调查数据进行对比,辅以沉水植物指标评价水体生态状况。结果表明: 白洋淀2019年总磷达到优于Ⅲ类水平,总氮优于Ⅱ类水平,水环境状况比2010年有较大改善。其中,总磷浓度下降了88.6%,总氮浓度下降了83.9%,叶绿素a浓度下降了47.8%,透明度增大了43.4%;浮游动物、底栖动物等生物多样性明显提高,浮游植物密度显著降低,优势种从普遍的重富营养型转变为富营养型;沉水植物清洁型种类分布范围缩小。这说明沉水植物物种评价法不适于白洋淀水环境质量评价,浮游动物多样性评价方法也不适用于当前白洋淀水环境质量评价。早期,由于污染源集中在西部,白洋淀西部水质差于东部;现在,由于自西、北、南多方向补水,淀区的水动力条件发生显著改变,加之西部实施治污工程,淀区水质呈现空间均一化的特点。     

云南省抚仙湖和杞麓湖流域土地利用变化对水质的影响

基于遥感技术和相关性分析方法对云南省2个典型高原湖泊抚仙湖和杞麓湖流域土地利用与湖泊水质关系进行了研究.结果表明:1989-2005年2个流域森林(灌丛)和建筑用地大量增长,荒地和农田有所减少,杞麓湖的污染程度比抚仙湖严重,2个湖泊的水质都呈下降趋势;湖滨200 m左右的缓冲带变化明显,且比流域其它土地利用变化对水质影响更大;城镇化和密集的农业耕作产生的面源污染是影响湖泊水质状况持续下降的重要因素,2个湖泊流域的土地利用结构不同也导致二者之间水质的差异.     

Spatial decision support system for assessing lake pollution hazard: southeastern pampean shallow lakes (Argentina) as a case study

This study gives an account of the implementation of a decision support system as a logical framework for assessing lake pollution hazard. The use of this system is demonstrated with an example from two lake watersheds, each one with different land-use, soil and topographic characteristics and also management regulations for natural resource protection. Lake pollution hazard is assessed as a function of two primary topics: hydrologic and soil conditions. This model shows the state of each evaluated subwatershed with respect to its potential contribution to lake pollution hazard based mainly on easily measurable and commonly used parameters: drainage density, hydraulic gradient, water table depth, soil media, land-cover and topography. Mapped outputs from the logic model allowed the identification of several subwatersheds in each lake system as the main water resource protection areas. La Brava Lake Watershed results reasonably fit the natural conditions of the watershed, identifying those subwatersheds with a moderate to high drainage density, high hydraulic gradient and extensive agricultural activities. Regarding Los Padres Lake Watershed, moderate to high drainage density, a high to a very high hydraulic gradient, shallow water depth, silt-loam soil texture and intensive agricultural and residential lands, contributed to a high lake pollution hazard. Results highlighted differences between both freshwater systems, being Los Padres Lake Watershed a more vulnerable ecosystem, requiring decision maker’s intervention based on accurate and up-to-date information.     

Monitoring ecological status of wetlands using linked fuzzy inference system- remote sensing analysis

The present study develops an applicable model to simulate the ecological status of saltwater lakes in which depth and total dissolved solids are selected as the effective factors on aquatic habitats. First, spectral images of the operational land imager of Landsat 8 were used to simulate distribution of depth and total dissolved solids by applying two feed forward neural networks. Next, a Mamdani fuzzy inference system was used to develop habitat suitability rules of Artemia and Flamingo as the selected target species in which expert opinions were considered. Finally, habitat suitability maps of target species were generated by linking distribution maps of selected effective parameters and fuzzy inference system. Based on the results in the Urmia lake as a case study, the Nash–Sutcliffe efficiency coefficients of depth and total dissolved solids are 0.88 and 0.5 which indicates the proposed method for simulating distribution of these parameters is reliable. Average depth in the simulated date is 227 cm, while average simulated total dissolved solids is 264 g per litre. Simulation of habitat suitability maps demonstrated that average habitat suitability of Artemia is less than 30% in the most areas of the lake. Moreover, average habitat suitability of the Flamingo is less than 10% which implies the ecological status of the lake is critical and ecological restoration is necessary. The main advantage of the proposed method is to develop a framework for combining the expert opinions with remote sensing data processing to generate habitat suitability maps in lakes.     

基于形态结构特征的洞庭湖湖泊健康评价

湖泊形态结构特征与湖泊健康水平密切相关,是评价湖泊生态系统、服务功能是否健康的基础。以历史文献、图件及1986—2010年水沙实测数据等资料为依据,选取洲滩出露频率、水动力条件和几何形态等3类共8个湖泊形态结构特征指标,采用突变级数法对典型年洞庭湖湖泊健康状况进行评价。结果表明:(1)东洞庭湖湖泊健康状况相对较好(隶属度平均值为0.7486),但波动较大(0.6775—0.7965);南洞庭湖次之(0.7311);西洞庭湖最差(0.7269),其健康相对隶属度波动最小(0.7026—0.7487);(2)近20a来,洞庭湖湖泊健康状况呈现"先恶化、后改善"的趋势。1998年其健康隶属度最低(0.6912),2004年值最高(0.7711),其余典型年的值介于二者之间。这充分说明"4350"工程实施、三峡水库蓄水运行等有效地改善了洞庭湖湖泊形态结构,使其朝健康方向发展。     

General Description of Partly Meromictic Hypertrophic Lake Verevi,its Ecological Status,Changes during the Past Eight Decades,and Restoration Problems

The present study describes generally the ecosystem of Lake Verevi while more detailed approaches are presented in the same issue. The main task of the article is to estimate long-term changes and find the best method for the restoration of good ecological status. Lake Verevi (surface 12.6 ha, mean depth 3.6 m, maximum depth 11 m, drainage area 1.1 km², water exchange 0.63-times per year) is a hypertrophic hard-water lake located in town Elva (6400 inhabitants). Long-term complex limnological investigations have taken place since 1929. The lake has been contaminated by irregular discharge of urban wastewaters from oxidation ponds since 1978, flood from streets, and infiltrated waters from the surrounding farms. The so-called spring meromixis occurred due to extremely warm springs in recent years. The index value of buffer capacity of Lake Verevi calculated from natural conditions is on the medium level. Water properties were analysed according to the requirements of the EU Water Framework Directive. According to the classification, water quality as a long-term average of surface layers is moderate-good, but the water quality of bottom layers is bad. Values in deeper layers usually exceed 20–30 times the calculated reference values by Vighi and Chiaudani’s model. Naturally, at the beginning of the 20th century the limnological type of the lake was moderately eutrophic. During the 1980s and 1990s the ecosystem was out of balance by abiotic characteristics as well as by plankton indicators. Rapid fluctuations of species composition and abundance can be found in recent years. Seasonal variations are considerable and species composition differs remarkably also in the water column. The dominating macrophyte species vary from year to year. Since the annual amount of precipitation from the atmosphere onto the lake surface is several times higher, the impact of swimmers could be considered irrelevant. Some restoration methods were discussed. The first step, stopping external pollution, was completed by damming the inlet. Drainage (siphoning) of the hypolimnetic water is discussed. Secondary pollution occurs because Fe:P values are below the threshold. The authors propose to use phosphorus precipitation and hypolimnetic aeration instead of siphoning.     

青海湖刚毛藻分布特征变化及成因分析

研究通过现场调查、资料收集和遥感影像解译, 分析了青海湖水质及刚毛藻分布特征的变化。研究结果表明, 2011—2019年间青海湖主要水质指标保持动态稳定, 水质未呈现显著变化趋势。2019年现场调查发现, 近岸浅水区域存在数量可观的刚毛藻分布, 其中以鸟岛和布哈河口周边区域刚毛藻生物量最大, 8月的平均生物量达到5213.4 g/m2。遥感影像分析显示, 刚毛藻主要分布在青海湖西部及西北部湖湾及入湖河口附近, 1987—2019年间刚毛藻覆盖面积呈先下降后上升趋势。分析表明, 近年来刚毛藻生物量增加和覆盖面积扩大可能与青海湖水位不断上涨以及新生浅水淹没区范围的扩大有关。目前, 刚毛藻分布的这种变化对青海湖水生态环境的影响尚不明确, 后续需进一步关注气候变化与人类活动双重影响下青海湖的水生态系统健康问题。     

太湖春季水体固有光学特性及其对遥感反射率变化的影响

吸收特性和后向散射特性是水体重要的光学特性,同时也是建立生物光学模型的基本参数。利用2009年4月太湖春季实测数据,结合生物光学模型推导了太湖春季水体颗粒物后向散射系数,在此基础上分析了太湖春季水体的吸收特性和后向散射特性,并利用经验正交分解方法对遥感反射率变化的影响因子进行了分析。结果表明：（1）非色素颗粒物是影响太湖春季水体吸收特性的主导因子,色素颗粒物和CDOM对总吸收（不包含纯水）的贡献相对较小,且色素颗粒物在梅梁湾湖区的包裹效应明显大于其他湖区。（2）颗粒物后向散射系数与总悬浮物和无机悬浮物具有很强的相关性（相关系数均在0.88以上）,与有机悬浮物的相关性相对较弱（相关系数均在0.73以下）,且水体中多次散射对水面总辐亮度有较大的贡献,平均贡献率高达93.46%。（3）利用经验正交分解方法将遥感反射率变化光谱分解成3个正交因子,3个正交因子总共解释了约99%的遥感反射率变化信息,其中,第一正交因子解释了93%的变化信息,第二和第三正交因子分别解释了5%和1%的变化信息。通过对各正交因子与水体不同组分的吸收和后向散射系数进行相关性分析得出,颗粒物的后向散射对水面反射光谱的形成具有非常重要的影响,太湖春季水体遥感反射率的变化主要取决于无机颗粒物的吸收和后向散射,有机颗粒物对遥感反射率的变化影响较小。     

污染湖泊生态系统服务净价值评估——以滇池为例

水污染是我国湖泊生态系统面临的主要环境问题,评估污染条件下湖泊生态系统服务价值对认识和恢复污染湖泊生态功能具有重要意义。针对当前湖泊生态系统服务价值评估方法中未体现水污染负面影响的问题,在传统生态系统服务功能价值评估框架的基础上,将水污染造成的经济损失内化到生态系统服务功能价值中,提出了针对污染湖泊的生态系统服务功能价值评估方法。应用该方法核算了滇池生态系统服务净价值。结果表明:2006—2015年滇池年均生态系统服务功能价值为386.55亿元/a,水污染改变了滇池的价值结构,供水价值丧失,气候调节、旅游和蓄水等成为滇池生态系统服务功能价值的主体;滇池水污染对流域内外经济社会发展造成的损失达到15.23亿元/a,主要经济损失体现在生态补水费用和蓝藻去除费用;综合湖泊生态系统服务的正向价值和水污染的负向损失,滇池的净生态系统服务价值为371.33亿元/a,占流域GDP的11.7%。将水污染损失纳入湖泊生态系统服务功能价值评估中增强了评价结果的合理性,能够更加真实地反映湖泊生态系统对人类经济社会发展的综合影响。     

DO PHYSICAL FACTORS REGULATE PHYTOPLANKTON DISTRIBUTION PATTERNS IN LARGE, SHALLOW LAKES?

Factors that regulate phytoplankton dynamics in shallow, productive lakes are poorly understood, due to their predisposition for frequent algal blooms and sediment resuspension events. In Lake Apopka, greatest phytoplankton biomass reflects wind-induced resuspension of algae (meroplankton) that exists on the aphotic lake bottom in a layer approximately 5 cm thick; this assemblage is dominated by diatoms (>60% of total biomass) that can occur in resting stages. Once exposed to moderate light, meroplankton are capable of growth and photosynthetic rates comparable with surface populations. In Lake Okeechobee, remote sensing was used to assess the basin-wide distribution of suspended particles. Satellite reflectance values agreed well with in situ particle densities at 20 in-lake stations (average r²; LANDSAT = 0.81, AVHRR = 0.53), and maps of algal blooms (r² = 0.79, p ≤ 0.01). The greatest chlorophyll concentrations occurred in the vicinity of tributary nutrient inputs at the lake's perimeter, while turbidity increased towards the center of the lake reflecting predominant water circulation patterns. These results underscore the importance of physical-biological interactions in lakes.     

山东麻大湖污染生态恢复模式与应用

麻大湖位于山东省博兴、桓台、高青三县交界处 ,系由乌河、朱龙河、孝妇河等 7条河流汇水而成的内陆淡水湖泊。该湖自 196 6年始遭受淄博市工业废水严重污染 ,自然生态系统极度破坏 ,经济损失巨大。 2 0世纪 70年代中期 ,沿湖群众因受污染之苦而人民来信来访不断。直到 90年代初 ,该湖污染一直受到国务院和山东省政府的关注。在各级环保部门不懈努力下 ,该湖污染几经治理 ,取得一定成效 ,但仍没有使其污染生态得以恢复。为此 ,笔者于1994 - 1995年对麻大湖污染生态恢复模式进行了研究 ,并为滨州地区行署决定开发利用麻大湖提供了决策依据。…     

Water quality of rivers in the drainage basin of Lake Peipsi

Water quality and general regularities that have taken place in the water quality of rivers in the drainage basin of Lake Peipsi have been studied. The lake is located in the basin of the Narva River. Several changes of the water quality of the lake in 1970–1990, caused by increasing human impact, have been observed. The water quality of Lake Peipsi depends on its pollution load whereby the main part of the pollution reaches the lake via rivers. In 1985–1989 an extensive research programme of the lake and its drainage basin was carried out. In this paper a part of these results as well as recent changes in the water quality of the rivers of the Estonian side are studied. In the last years an improvement of the general state of rivers in the drainage basin of Lake Peipsi has taken place; the content of organic substances and nutrients in the inflows of Lake Peipsi and in the Narva River (lake outflow) is decreasing.     

基于RS和GIS的南四湖生态风险评价

以南四湖为研究对象,基于遥感和地理信息系统技术选择干旱、洪涝和水体污染为生态风险源,用景观格局指数评价生态系统的地位和受体的易损性,并依据生态风险值的度量的原理,通过计算综合风险概率及生态综合损失度得到南四湖的生态风险值。研究结果显示南四湖区域生态风险存在上下级湖分异与湖滨带状分异两大空间规律,体现了南四湖区域生态风险的空间异质性。     

可鲁克湖流域生态环境质量诊断

为查清可鲁克湖流域生态环境质量及其演化特征,依据环境生态学相关理论,运用综合调查、遥感解译与反演、统计分析等技术手段,选取自然地理、气象、土地利用/覆盖和社会经济4个方面的15个因子,采用因子分析和熵值法计算指标权重,建立流域土壤质量模型和生态环境质量诊断模型,分析2000、2005、2010和2015年可鲁克湖流域土壤与生态环境质量的变化特征。结果表明: 2000、2005、2010和2015年,可鲁克湖流域生态环境质量均值依次为21、47、54和72,呈稳定上升趋势,生态环境质量等级由较差转为良好,土壤质量整体处于中等水平;空间上,北部山区和流域下游湿地及河流周边区域的生态环境质量明显转好。流域生态环境质量变化是人类活动与自然因素共同作用的结果,土壤质量与湖泊面积是指示流域生态环境的关键因子,可鲁克湖最小生态需水量是维持流域生态环境良性发展的基本保障。     

河湖生态系统生态用水优化研究——以滇池流域为例

生态需水是河流与湖泊生态系统健康的重要基础。湖泊流域的河流与湖泊生态系统之间存在密切的水量联系,目前对流域内生态用水的研究多为单一生态系统生态需水简单相加,忽略了河流和湖泊之间复杂的水量联系。基于河湖复合生态系统之间的水量联系构建了河湖生态系统生态用水优化模型,并以滇池为例分析了河湖生态系统生态用水规律。结果表明：湖泊流域中单一河流或湖泊生态需水计算结果不能满足复合生态系统的生态用水要求,需要综合考虑河流和湖泊之间的水量联系;在当前水质状况下,牛栏江每年的调水量不能满足滇池流域的生态用水要求;滇池流域水体污染对流域内生态用水影响较大,随着水体污染程度的下降,流域生态用水量和调水量呈指数下降,河流生态用水呈线性下降。     

Lake Ladoga: chemical pollution and biochemical self-purification

In order to evaluate the current state of Lake Ladoga a comprehensive investigation of its pollution by metals, oil products, phenols, anionic detergents and 3,4-benzopyrene was carried out in July 1993. The results indicate a considerable pollution of the lake waters by metals and phenols. High degree of pollution of sediments by oil products and 3,4-benzopyrene was detected in the northern archipelago and bays (Impilahti, Hiidenselka, Pitkäranta) and in the area of Priozersk. A new index (I _t ) is suggested for the estimation of biochemical self-purification capacity of freshwaters. Values of the index obtained for different parts of Lake Ladoga led us to the conclusion that at the present time the lake's capability for biochemical self-purification is relatively low. This situation is caused by strong anthropogenic impact on the lake. The approximate pollutant loads to Lake Ladoga were estimated on the basis of our 1991–1993 monitoring survey and from literature data.