基于随机森林的内陆湖泊水体有色可溶性有机物(CDOM)浓度遥感估算

水体中的有色可溶性有机物(CDOM)是湖泊生态系统中氮、磷等有机营养物质的重要来源,利用卫星遥感数据反演内陆水体中CDOM浓度一直是个挑战.因此本文基于滇池2009年9月、2017年4月以及太湖2016年7月的现场原位观测和室内实验,在分析水体固有光学特性的基础上,引入机器学习算法,建立了基于哨兵-3A OLCI传感器的我国内陆湖泊水体CDOM浓度随机森林反演模型.利用独立的验证数据集对所构建的随机森林模型及常用的波段比值模型、一阶微分模型、半分析模型、BP神经网络模型等的反演精度进行评价.结果表明:随机森林模型的均方根误差为0.14 m-1,平均相对误差为21%,与反演效果相对较好的BP神经网络模型相比,均方根误差降低了50%,平均相对误差降低了38%,反演精度得到了显著的提高.根据随机森林算法的特征重要性参数提供的各自变量影响力结果,发现B11(709 nm)和B6(560 nm)波段贡献率最大,是反演CDOM的敏感波段.最后将随机森林模型应用到滇池2017年4月12日、太湖2017年5月18日的哨兵-3A OLCI影像上,得到滇池、太湖水体CDOM浓度分布图.滇池CDOM浓度的分布特征大致符合东北、西南高,中西部低的趋势,且河口处的CDOM浓度高于湖泊水体,表明径流的输入给滇池水体带来了大量的CDOM.太湖CDOM浓度的分布特征大致符合西部高,湖心区和东部低的趋势.太湖西部以及北部梅梁湾受入湖河流影响较大,CDOM浓度较高,太湖开敞区远离河口处,受外源河流的影响逐渐减小,且由于湖水的不断稀释,CDOM浓度不断降低.太湖东部水生植物很多,湖水较为清澈,CDOM浓度较低.     

太湖梅梁湾有色可溶性有机物的空间分布及光学行为

2004年3月对太湖梅梁湾有色可溶性有机物(CDOM)的吸收和荧光等光学行为进行研究,并由此探讨了CDOM的空间分布．结果表明,溶解性有机碳(DOC)的浓度在10．48-19．72 mg／L间变化,其均值为13．20±2．79 mg／L;CDOM在280 nm,355 nm和440 nm的吸收系数分别为18．73-31．91 m-1(平均值23．19±4．36 m-1)、4．63-7．14 m-1(平均值5．76±0．91 m-1)、1．45-2．99 m-1(平均值1．92±0．40 m-1);355 nm波长处CDOM的比吸收系数为0．34-0．57 L／(mg·m),平均值0．44±0．06 L/(mg·m);表征CDOM分子大小的比值a(250)／a(365)变化范围为5．05-7．55;355 nm的激发波长、450 nm的发射波长处的荧光值的变化范围0．79-3 04 nm-1(平均值1．69±0．77 nm-1)．CDOM吸收系数、DOC浓度、荧光强度的分析显示CDOM浓度呈现从河口往湾内、湾口递减的趋势．CDOM吸收与DOC浓度的相关性随波长的降低而增加,在短波部分存在明显的正相关．355 nm处的荧光值、DOC浓度与CDOM吸收系分别存在如下显著性正相关关系:Fn(355)=0．692(±0．135)a(355)-2．297(±0．786),a(355)=0．233(±0．061)DOC 2．690(±0．816)．280 -500 nm、280-360 nm、360-440 nm指数函数斜率S值分别为13．86±0．91、18．54±1．11、12．93±0．92μm-1,S值与比吸收系数之间存在显著的负线性相关关系,而与a(25)／a(365)值则存在显著的正线性关系．比吸收系数越大,a(250)／a(365)值和S值就越小,对应的CDOM分子量就越大,腐质酸的比例就越高．     

引黄水源水库平-丰水期有色可溶性有机物(CDOM)的来源、组成及差异分析

为探究引黄水源水库——门楼水库平水期和丰水期有色可溶性有机物(CDOM)的组成特征、来源及差异,运用紫外—可见光谱技术(UV-vis)和三维荧光光谱(EEMs)技术,结合平行因子分析法(PARAFAC)分析2022年5月(平水期)和2022年7月(丰水期)有色可溶性有机物含量及组分变化。研究结果表明:PARAFAC识别出2类荧光组分,分别是C1(Ex=355 nm,Em=476 nm,类腐殖质组分)和组分C2(Ex=225 nm,Em=320 nm,类蛋白组分);丰水期CDOM组分荧光强度显著高于平水期。CDOM光谱参数表明,门楼水库水体处于中营养状态,水体CDOM受新生内源和外源输入共同影响,以自生源为主;水库CDOM具有相对分子量小、腐殖化程度较弱的特点。丰水期水库富营养化水平和CDOM相对浓度低于平水期;丰水期CDOM疏水性组分比例和芳香化程度高于平水期。水质理化指标、CDOM组分和光谱参数相关性分析结果表明SUVA₂₆₀和SUVA₂₈₀与DOC呈显著负相关,说明紫外—可见光谱参数在一定程度上可以用来估算DOC的浓度;Chl.a浓度作为...     

亚热带河口区水库DOC和DIC浓度时空变化特征

沿海水库汇聚并埋藏着大量的碳,是全球碳循环的重要区域.水体溶解有机碳(DOC)和溶解无机碳(DIC)的生物地球化学行为是水库碳循环研究的重要组成部分,对其系统生物过程和生态环境变化具有重要的影响.为了解亚热带河口区文武砂水库表层水体DOC和DIC的时空分布特征,本研究于2018年11月、2019年3月和6月分别对库区表...     

淮河流域周村水库夏季CDOM吸收光谱特征、空间分布及其来源分析

基于2015年8月采集的24个淮河流域以周村水源水库为代表的表层水样的有色溶解性有机物(CDOM)吸收系数数据,研究了CDOM吸收光谱的空间分布特征,考察了CDOM的吸收系数与水质参数的相关关系,同时探讨了周村水库夏季CDOM的潜在来源.结果显示:依据CDOM的吸收光谱空间分布特性及采样点分布特征,周村水库分为入库口、过渡区和主库区3个特征水域;CDOM的吸收系数沿入库口到主库区依次递减,S值呈现相反的趋势;分析发现S240~500与a(355)和a*(355)呈极显著负相关(R~2=0.98、0.88);CDOM吸收系数a(355)与溶解性有机碳(DOC)浓度具有良好的线性相关,有利于建立DOC遥感反演模型;同时,CDOM吸收系数a(355)与a_(ph)(440)存在极显著正线性相关,表明浮游植物的新陈代谢及其降解产物是夏季周村水库CDOM的潜在来源.综上,通过对夏季周村水库水体CDOM的研究,丰富了关于水源水体CDOM的调查资料,可为日后水库的管理提供技术支撑.     

太湖有色可溶性有机物(CDOM)对COD及BOD5的指示意义

化学需氧量(COD)、五日生化需氧量(BOD_5)及溶解性有机碳(DOC)是指示湖泊水质的重要指标,然而上述指标测定通常耗费大量时间、试剂及人力物力且排放大量有害废液.有色可溶性有机物(CDOM)是溶解性有机物(DOM)中可以强烈吸收光谱中的紫外光和可见光的部分,数据测定耗时短、方便快捷,且样品处理过程环境友好,能在很大程度上反映湖泊水质.本研究基于2016年2、5和8月在太湖均匀布设的32个采样点进行样品采集,运用光谱吸收与三维荧光-平行因子分析(EEMs-PARAFAC)探究太湖CDOM的光谱吸收和荧光组分,探讨CDOM光谱指标对湖泊BOD_5、COD及DOC浓度等湖泊环境质量指标的可替代性.结果表明:(1)运用EEMs-PARAFAC方法解析出3种荧光组分:类腐殖酸C1、类酪氨酸C2和类色氨酸C3.(2) COD和BOD_5和DOC在空间上呈现出相似的分布趋势,不同水期的最高值均出现在竺山湾和梅梁湾,由西北湖区至中部敞水区、东南湖湾递减.(3)在不同水期,COD、BOD_5、DOC浓度和C1组分均表现为丰水期极显著大于枯水期和平水期,a_(254)在丰、平、枯水期间无显著性差异,最大值出现在丰水期;C2与C3组分均在枯水期和平水期极显著大于丰水期.(4)在不同水文时期,COD、BOD_5和DOC浓度均与a_(254)、类腐殖酸C1呈显著正相关,丰水期太湖COD、BOD_5和DOC浓度与CDOM光谱指标的线性相关性要优于枯水期和平水期.(5) CDOM光谱指标在不同水文时期均能很好地替代COD、BOD_5和DOC等作为反映太湖水体中有机物污染程度及湖泊水质的指标.     

基于页岩线性岩石物理模型的TOC反演方法研究

总有机碳含量(TOC)是评价页岩有机质丰度和生烃能力的重要指标.常规TOC预测方法是通过对弹性参数体(例如密度反演体)进行线性拟合得到的,该方法具有地区经验性,并且缺乏有效的理论支撑.针对此问题,本文以岩石物理驱动的物性参数反演作为技术手段,首先对典型页岩的矿物组分以及孔隙流体特征进行分析,得出影响页岩弹性参数变化的四...     

Distribution,characteristics and potential impacts of chromophoric dissolved organic matter (CDOM) in Hudson Strait and Hudson Bay,Canada

The characteristics of chromophoric dissolved organic matter (CDOM) were studied in Hudson Bay and Hudson Strait in the Canadian Arctic. Hudson Bay receives a disproportionately large influx of river runoff. With high dissolved organic matter (DOM) concentrations in Arctic rivers the influence of CDOM on coastal and ocean systems can be significant, yet the distribution, characteristics and potential consequences of CDOM in these waters remain unknown. We collected 470 discrete water samples in offshore, coastal, estuarine and river waters in the region during September and October 2005. Mixing of CDOM appeared conservative with salinity, although regional differences exist due to variable DOM composition in the rivers discharging to the Bay and the presence of sea-ice melt, which has low CDOM concentrations and low salinity. There were higher concentrations of CDOM in Hudson Bay, especially in coastal waters with salinities <28$<28$, due to river runoff. Using CDOM composition of water masses as a tracer for the freshwater components revealed that river runoff is largely constrained to nearshore waters in Hudson Bay, while sea-ice melt is distributed more evenly in the Bay. Strong inshore–offshore gradients in the bio-optical properties of the surface waters in the Hudson Bay cause large variation in penetration of ultraviolet radiation and the photic depth within the bay, potentially controlling the vertical distribution of biomass and occurrence of deep chlorophyll maxima which are prevalent only in the more transparent offshore waters of the bay. The CDOM distribution and associated photoprocesses may influence the thermodynamics and stratification of the coastal waters, through trapping of radiant heating within the top few meters of the water column. Photoproduction of biologically labile substrates from CDOM could potentially stimulate the growth of biomass in Hudson Bay coastal waters. Further studies are needed to investigate the importance of terrestrial DOM in the Hudson Bay region, and the impact of hydroelectric development and climate change on these processes.     

Response characteristics of DOC flushing in an alpine catchment

The spatial distribution of source areas and associated residence times of water in the catchment are significant factors controlling the annual cycles of dissolved organic carbon (DOC) concentration in Deer Creek (Summit County, Colorado). During spring snowmelt (April–August 1992), stream DOC concentrations increased with the rising limb of the hydrograph, peaked before maximum discharge, then declined rapidly as melting continued. We investigated catchment sources of DOC to streamflow, measuring DOC in tension lysimeters, groundwater wells, snow and streamflow. Lysimeter data indicate that near-surface soil horizons are a primary contributor of DOC to streamflow during spring snowmelt. Concentrations of DOC in the lysimeters decrease rapidly during the melt period, supporting the hypothesis that hydrological flushing of catchment soils is the primary mechanism affecting the temporal variation of DOC in Deer Creek. Time constants of DOC flushing, characterizing the exponential decay of DOC concentration in the upper soil horizon, ranged from 10 to 30 days for the 10 lysimeter sites. Differences in the rate of flushing are influenced by topographical position, with near-stream riparian soils flushed more quickly than soils located further upslope. Variation in the amount of distribution of accumulated snow, and asynchronous melting of the snowpack across the landscape, staggered the onset of the spring flush throughout the catchment, prolonging the period of increased concentrations of DOC in the stream. Streamflow integrates the catchment-scale flushing responses, yielding a time constant associated with the recession of DOC in the stream channel (84 days) that is significantly longer than the time constants observed for particular locations in the upper soil. © 1997 John Wiley & Sons, Ltd.     

有色可溶性有机物(CDOM)吸收作为湖库化学需氧量监测替代指标的探讨

化学需氧量是衡量水体中有机物量及污染程度的综合性指标,也是我国《地表水环境质量标准(GB 3838—2002)》的重要评价指标.然而,由于测定过程缓慢和使用了有毒有害试剂易于形成二次污染,现行标准的高锰酸钾和重铬酸钾化学需氧量测定方法无法做到环境友好,也不能反映当前快速和实时监测的技术需求.因此,迫切需要发展操作简便、快速高效、灵敏可靠、环境友好和环保绿色的化学需氧量替代检测方法.本文首先从文献计量学视角比较我国与世界上发达国家化学需氧量研究主题论文发文量,剖析了我国发展化学需氧量替代检测方法的迫切性.基于全国大范围65个湖库706个样本有色可溶性有机物吸收系数、化学需氧量和其他水质参数同步调查数据,构建广覆盖范围的有色可溶性有机物特征波长吸收系数和化学需氧量间高精度线性关系模型,确定了地表水环境质量评价的吸收系数阈值,模型可以广泛应用于不同类型(深水、浅水、大型、中型、小型)和不同营养状态(贫、中、富营养)湖库水体有机物浓度的定量表征,具有一定普适性.通过对比有色可溶性有机物吸收系数和传统的高锰酸钾、重铬酸钾法优势和不足,明确了有色可溶性有机物吸收系数替代化学需氧量用于湖库水体开展有...     

洱海沉积物有色可溶性有机物(CDOM)三维荧光空间分布特性及指示意义

采用荧光光谱区域体积积分法定量研究洱海沉积物有色可溶性有机物（CDOM）三维荧光空间分布特性,结果表明,洱海沉积物CDOM中类腐殖酸组分占比最高（44.5%~74.2%）,其次为溶解性微生物代谢产物（10.7%~20.4%）和类富里酸物质（8.19%~19.7%）,而类蛋白组分占比最低,类腐殖酸占比与H/C和N/C均呈显著负相关;南部湖心平台区域沉积物类富里酸和类蛋白组分占比较高,其CDOM自生源占比较高;北部水生植物分布区溶解性微生物代谢产物占比较高,CDOM陆源性较强;中部深水区及南北湖湾区域类腐殖酸占比较高,随水深增加沉积物CDOM自生源特征增强.随沉积物深度增加类腐殖酸和溶解性微生物代谢产物占比呈下降趋势,类富里酸占比呈上升趋势,CDOM自生源特征增强,脂肪化程度增高.随湖泊富营养化程度增加沉积物CDOM受微生物代谢产物影响增强,陆源性增加,而自生源性降低,湖泊沉积物CDOM具有富营养化指示意义.     

Total waterborne carbon export and DOC composition from ten nested subarctic peatland catchments—importance of peatland cover,groundwater influence,and inter‐annual variability of precipitation patterns

Waterborne carbon (C) export from terrestrial ecosystems is a potentially important flux for the net catchment C balance and links the biogeochemical C cycling of terrestrial ecosystems to their downstream aquatic ecosystems. We have monitored hydrology and stream chemistry over 3 years in ten nested catchments (0.6–15.1 km²) with variable peatland cover (0%–22%) and groundwater influence in subarctic Sweden. Total waterborne C export, including dissolved and particulate organic carbon (DOC and POC) and dissolved inorganic carbon (DIC), ranged between 2.8 and 7.3 g m^–2 year^–1, representing ~10%–30% of catchment net ecosystem exchange of CO₂. Several characteristics of catchment waterborne C export were affected by interacting effects of peatland cover and groundwater influence, including magnitude and timing, partitioning into DOC, POC, and DIC and chemical composition of the exported DOC. Waterborne C export was greater during the wetter years, equivalent to an average change in export of ~2 g m^–2 year^–1 per 100 mm of precipitation. Wetter years led to a greater relative increase in DIC export than DOC export due to an inferred relative shift in dominance from shallow organic flow pathways to groundwater sources. Indices of DOC composition (SUVA₂₅₄ and a₂₅₀/a₃₆₅) indicated that DOC aromaticity and average molecular weight increased with catchment peatland cover and decreased with increased groundwater influence. Our results provide examples on how waterborne C export and DOC composition might be affected by climate change. Copyright © 2012 John Wiley & Sons, Ltd.     

基于Hyperion数据的太湖水体叶绿素a浓度遥感估算

通过对2004年8月19日太湖Hyperion高光谱遥感数据的处理和分析,文章首先采用比值和一阶微分处理技术进行了叶绿素a浓度的估算．为了弥补此两种方法在模型的适用性和通用性方面的不足,本文尝试了利用混合光谱分析模型进行太湖水体叶绿素a浓度的提取和成图．实验结果说明高光谱遥感数据Hyperion可以进行水体叶绿素a浓度的监测,并且作为高光谱处理技术之一的混合光谱分析技术是水体叶绿素a浓度估算的另一条佳径．