基于无人机影像与面向对象-深度学习的滨海湿地植物物种分类

明确滨海湿地植物物种类型及其分布状况是实现滨海湿地精细化生物多样性监测的基础,对于滨海湿地的保护管理与生态可持续发展均具有重要意义。本研究以无人机可见光遥感影像为基础数据源,在定量分析最优分割尺度与最优分类特征组合的基础上,应用面向对象-U-net深度学习方法对闽江河口湿地植物物种类型进行分类,并与K最近邻、决策树、随机森林和贝叶斯分类方法进行精度对比分析,以期为滨海湿地植物物种遥感精细分类与生物多样性保护管理提供方法借鉴与科学参考。研究结果表明,利用面向对象-U-net深度学习方法提取不同滨海湿地植物物种类型的分类精度可达95.67%,总体精度较其他分类方法提高6.67%–13.67%, Kappa系数提高0.12–0.31,且分类整体性好。此外,实现植物物种光谱特征、形状特征、纹理特征与高度特征的最优特征选择对于有效提高湿地植物物种信息分类精度具有重要作用,应用最优分割尺度实现影像分割可提高整体分类效率。     

基于面向对象的QuickBird遥感影像林隙分割与分类

传统的实地调查和人工解译方法已经不能满足区域尺度的林隙获取,高空间分辨率遥感影像的出现为区域尺度的林隙获取提供了可能.本研究采用QuickBird高空间分辨率光学遥感影像,结合面向对象分类技术对福建省三明市将乐县将乐国有林场进行林隙分割与分类.在面向对象分类过程中,采用10种尺度(10~100,步长为10)对QuickBird遥感影像进行分割,应用参考对象相交面积(RA_or)和分割对象相交面积(RA_os)进行分割结果评价.对每个尺度分割结果应用16个光谱特征,采用向量机分类器(SVM)进行林隙、非林隙和其他类型分类.结果表明:通过RA_or和RA_os等值法获得最优分割尺度参数为40.不同尺度参数之间的分类总精度最高相差22%.在最优尺度下,应用SVM分类器对林隙、非林隙和其他类型分类的总精度高达88%(Kappa=0.82).采用高空间分辨率遥感数据并结合面向对象的方法,可以代替传统的实地调查和人工解译对区域尺度的林隙进行识别分类.     

面向对象的优势树种类型信息提取技术

森林植被优势树种类型信息的提取是遥感影像分类中的难点.面向对象分类方法是用高空间分辨率遥感数据实现精确类型信息提取的新方法.本文以2013年Quickbird影像作为基础数据,选择福建省三明市将乐林场为研究区,采用面向对象多尺度分割方法提取耕地、灌草地、未成林造林地、马尾松、杉木和阔叶树等类型信息.分类特征融合植被的光谱、纹理和多种植被指数3类特征信息,建立类层次结构,对不同层次分别用隶属度函数和决策树分类规则,最终完成分类,并与只用纹理与光谱特征相结合的方法进行对比.结果表明:融合纹理、光谱、多种植被指数的面向对象的分类方法提取研究区优势树种类型信息的精度为91.3％,比只用纹理和光谱的方法精度提高了5.7％.     

基于不同决策树的面向对象林区遥感影像分类比较

面向地理对象影像分析技术(GEOBIA)是影像分辨率越来越高的背景下的产物.如何提高高分辨率影像分类精度和分类效率是影像处理的重要议题之一.本研究对QuickBird影像多尺度分割后的对象进行分类,分析了C5.0、C4.5、CART决策树算法在林区面向对象分类中的效率,并与kNN算法的分类精度进行比较.利用eCognition软件对遥感影像进行多尺度分割,分析得到最佳尺度为90和40.在90尺度下分离出植被和非植被后,在40尺度下提取不同类别植被的光谱、纹理、形状等共21个特征,并利用C5.0、C4.5、CART决策树算法分别对其进行知识挖掘,自动建立分类规则.最后利用建立的分类规则分别对植被区域进行分类,并比较分析其精度.结果表明: 基于决策树的分类精度均高于传统的kNN法.其中,C5.0方法的精度最高,其总体分类精度为90.0%,Kappa系数0.87.决策树算法能有效提高林区树种分类精度,且C5.0决策树的Boosting算法对该分类效果具有最明显的提升.     

基于不同空间尺度遥感影像估算森林叶面积指数的差异

地面测量点对遥感像元的代表性如何,怎样获取像元的相对真值,多大的空间分辨率可以真实地反映森林区域的叶面积指数(LAI),这些都是定量遥感中的重要问题.本研究计算LAI-2200和TRAC两种冠层分析仪测量的空间范围,并结合GF-2(4.1 m)、Sentinel-2(10 m)、Landsat-8(30 m)3种不同空间分辨率遥感影像,找到各尺度下像元的相对真值,在保持真值观测面积和遥感获取面积一致的条件下,基于一元指数和多元回归模型,对比分析不同空间分辨率影像对估算森林LAI的影响,并对3种影像模型进行30和100 m尺度下的检验以及各自数据集的空间代表性评价,比较得出最适合表达研究区域森林LAI的尺度.结果表明:对于森林来说,高分辨率并不一定能充分反映森林LAI.基于3种分辨率影像的统计模型都能很好地估测森林LAI,其中,基于Sentinel-2的反演精度最高,基于GF-2的反演精度最低.30和100 m尺度下的检验结果表明,基于GF-2反演模型高估了森林LAI,基于Landsat-8的反演模型低估了森林LAI,基于Sentinel-2分辨率的统计模型可以很好地估测研究区域森林LAI.     

基于中分辨率TM数据的湿地水生植被提取

利用湿地水生植被生长旺盛、光谱反射较强、光谱信息比较丰富的8月份中分辨率Landsat TM和ETM+多光谱遥感影像,采用面向对象的分类方法,进行野鸭湖湿地水生植被的提取。研究表明:在提取过程中,通过对原始影像进行主成分变换和穗帽变换,将主要信息与噪声分离,不仅减小了数据冗余和波段间的相关性,而且增大了影像上湿地水生植被与其他地物类型光谱和空间信息的差异性,并结合野外水生植被光谱特征分析,选择归一化植被指数NDVI与归一化水体指数NDWI辅助分类,构建特征波段或波段组合,然后,确定适当的隶属度函数和阈值范围,构建分类决策树,完成湿地水生植被的自动分类,提高了影像分割与面向对象分类的精度,取得了较为理想的湿地水生植被提取结果。2002年和2008年两景影像的总体分类精度分别达到86.5%和85.44%,表明中分辨率TM影像可以满足湿地水生植被提取的需要,又因为其具有较高的波谱分辨率、极为丰富的信息量、相对较低的价格、长时间序列,可以作为近20a湿地水生植被提取和动态变化监测的主要数据源。     

基于多时相无人机遥感影像优化河口湿地景观分类

河口湿地具有丰富的生物多样性和高度异质化的景观格局。针对河口湿地景观的复杂性,采用传统的基于单幅遥感影像的分类方法并不能得到较好的分类结果。本研究采用多时相无人机遥感影像参与分类,以优化河口湿地景观自动分类结果。选择天目湖上游平桥河河口湿地为研究区,选取4个季节的无人机影像为基础数据源,采用面向对象与决策树相结合的分类方法,针对不同季节组合的影像进行分类。结果表明:采用多时相无人机影像能显著提升分类效果,且参与分类的时相越多,效果越好;单季影像中,春季是最适合进行景观分类的季节,分类总体精度为62.7%,Kappa系数为0.59;当4个季节获取的影像同时参与分类时,分类总体精度为91.7%,Kappa系数为0.90;参与分类的时相光谱特征差异越大,分类效果提升越明显。本研究可为河口湿地景观分类提供技术支持,并提出了一种利用可见光无人机遥感影像进行湿地景观分类的新思路。     

环境灾害遥感小卫星在辽河三角洲湿地景观制图中的应用

及时、准确地获得湿地的空间分布,对湿地的动态监测、保护与可持续利用具有重要的意义.环境灾害遥感小卫星星座A、B星(HJ-1A/1B星)是我国自主发射的陆地资源监测卫星,可为湿地类型的提取提供新的遥感影像数据源.本文通过对比我国环境灾害遥感小卫星CCD相机影像（HJ CCD）数据与Landsat TM5影像数据获取的湿地景观类型图的分类精度和各景观类型面积,验证和探究了HJ CCD数据在湿地景观动态变化监测中的适用性和应用潜力.结果表明：HJ CCD数据在地物识别分类方面可完全替代Landsat TM5数据;在实时动态监测方面,HJ CCD数据获取周期仅为2 d,优于Landsat TM5数据（16 d）.     

基于多光谱影像的森林树种识别及其空间尺度响应

当前,不同空间分辨率卫星影像对森林类型识别结果中普遍存在的尺度效应,而且纹理参量对不同尺度下树种识别精度的影响仍缺乏广泛认知.本研究以中国东北旺业甸林场为研究区,采用观测时相同步、地理坐标匹配的GF-1 PMS、GF-2 PMS、GF-1 WFV,以及Landsat-8 OLI卫星传感器数据组成空间尺度观测序列(1、2、4、8、16、30 m),并结合支持向量机(SVM)模型,探讨了区域内5种优势树种遥感识别结果的尺度变化规律及其纹理特征参数的影响,同时检验了基于尺度上推转换影像的树种识别结果差异.结果表明: 影像空间分辨率对区域树种识别结果具有显著影响,其中,研究区森林树种识别的最佳影像分辨率为4 m,当分辨率降低至30 m时,树种识别结果最差.在1～8 m影像分辨率范围内,增加纹理信息能够显著提高不同优势树种的识别精度,使总分类精度提升了2.0%～3.6%,但纹理信息对16～30 m影像的识别结果没有显著影响.与真实尺度卫星影像相比,基于升尺度转换影像的树种识别结果及其尺度响应特征存在显著差异,表明在面向多个空间尺度的遥感观测和应用研究中,需要采用真实分辨率影像以确保结果的准确性.     

亚热带常绿林型遥感识别及尺度效应

光学遥感是获取宏观地表植被覆盖信息的重要手段,但常绿树种之间物候差异小,关于亚热带地区常绿林型的遥感识别研究相对较少。遥感林型识别存在尺度效应,从实际应用视角出发,常绿林型遥感识别的最优空间分辨率仍然不清楚。本研究以湖南省会同县为例,利用Pléiades(2 m)、RapidEye (5 m)、Landsat-8 (15、30 m) 4种光学遥感影像,结合光谱、纹理、植被覆盖度等特征变量与随机森林模型,探讨了3种典型亚热带常绿林型:杉木林(Chinese fir forest,CFF)、马尾松林(Masson pine forest,MPF)、常绿阔叶林(evergreen broadleaved forest,EBF)的最优遥感识别分辨率以及尺度效应问题。结果表明:研究区地表覆盖分类精度随影像空间分辨率的降低呈现先降低后上升的变化趋势,在2 m时具有最佳分类精度(Kappa=0.70,总精度=0.77)。3种林型的识别精度随空间分辨率的上升均表现出先降低后上升的变化规律,识别率(rate of identification,RI)范围分别为:RI_(CFF)=68%～87%、RI_(MPF)=55%～84%、RI_(EBF)=29%～74%。杉木林与马尾松林的漏分误差(omission error,OE)与错分误差(commission error,CE)低于常绿阔叶林(OE_(CFF)=0.26～0.46,CE_(CFF)=0.32～0.53; OE_(MPF)=0.31～0.50,CE_(MPF)=0.31～0.46; OE_(EBF)=0.47～0.71,CE_(EBF)=0.39～0.66)。本研究证实了亚热带常绿林型的遥感识别存在明显的尺度效应,30 m分辨率的Landsat-8影像相比高分辨率遥感影像因具备更丰富的光谱信息而具有更高的识别精度。本研究表明,常绿林型的遥感识别不宜盲目追求高空间分辨率,需要综合考虑遥感传感器光谱配置与空间分辨率之间的内在权衡。     

Hyperspectral Analysis of Columbia Spotted Frog Habitat

Abstract: Wildlife managers increasingly are using remotely sensed imagery to improve habitat delineations and sampling strategies. Advances in remote sensing technology, such as hyperspectral imagery, provide more information than previously was available with multispectral sensors. We evaluated accuracy of high-resolution hyperspectral image classifications to identify wetlands and wetland habitat features important for Columbia spotted frogs (Rana luteiventris) and compared the results to multispectral image classification and United States Geological Survey topographic maps. The study area spanned 3 lake basins in the Salmon River Mountains, Idaho, USA. Hyperspectral data were collected with an airborne sensor on 30 June 2002 and on 8 July 2006. A 12-year comprehensive ground survey of the study area for Columbia spotted frog reproduction served as validation for image classifications. Hyperspectral image classification accuracy of wetlands was high, with a producer's accuracy of 96% (44 wetlands) correctly classified with the 2002 data and 89% (41 wetlands) correctly classified with the 2006 data. We applied habitat-based rules to delineate breeding habitat from other wetlands, and successfully predicted 74% (14 wetlands) of known breeding wetlands for the Columbia spotted frog. Emergent sedge microhabitat classification showed promise for directly predicting Columbia spotted frog egg mass locations within a wetland by correctly identifying 72% (23 of 32) of known locations. Our study indicates hyperspectral imagery can be an effective tool for mapping spotted frog breeding habitat in the selected mountain basins. We conclude that this technique has potential for improving site selection for inventory and monitoring programs conducted across similar wetland habitat and can be a useful tool for delineating wildlife habitats.     

Satellite remote sensing of wetlands

To conserve and manage wetland resources, it is important to inventoryand monitor wetlands and their adjacent uplands. Satellite remote sensing hasseveral advantages for monitoring wetland resources, especially for largegeographic areas. This review summarizes the literature on satellite remotesensing of wetlands, including what classification techniques were mostsuccessful in identifying wetlands and separating them from other land covertypes. All types of wetlands have been studied with satellite remote sensing.Landsat MSS, Landsat TM, and SPOT are the major satellite systems that have beenused to study wetlands; other systems are NOAA AVHRR, IRS-1B LISS-II and radarsystems, including JERS-1, ERS-1 and RADARSAT. Early work with satellite imageryused visual interpretation for classification. The most commonly used computerclassification method to map wetlands is unsupervised classification orclustering. Maximum likelihood is the most common supervised classificationmethod. Wetland classification is difficult because of spectral confusion withother landcover classes and among different types of wetlands. However,multi-temporal data usually improves the classification of wetlands, as doesancillary data such as soil data, elevation or topography data. Classifiedsatellite imagery and maps derived from aerial photography have been comparedwith the conclusion that they offer different but complimentary information.Change detection studies have taken advantage of the repeat coverage andarchival data available with satellite remote sensing. Detailed wetland maps canbe updated using satellite imagery. Given the spatial resolution of satelliteremote sensing systems, fuzzy classification, subpixel classification, spectralmixture analysis, and mixtures estimation may provide more detailed informationon wetlands. A layered, hybrid or rule-based approach may give better resultsthan more traditional methods. The combination of radar and optical data providethe most promise for improving wetland classification.     

Applying remote sensing techniques to monitor shifting wetland vegetation: A case study of Danshui River estuary mangrove communities,Taiwan

The purpose of this study is to apply different remote sensing techniques to monitor shifting mangrove vegetation in the Danshui River estuary in Taipei, Taiwan, in order to evaluate a long-term wetland conservation strategy compromising between comprehensive wetland ecosystem management and urban development. In the Danshui estuary, mangrove dominated by Kandelia candel is the major vegetation, and a large area of marsh with freshwater grasses has been protected in three reserves along the river shore. This study applied satellite imagery from different remote sensors of various resolutions for spectral analysis in order to compare shifting wetland vegetation communities at different times. A two-stage analytical process was used for extracting vegetation area and types. In the first-stage, a normalized difference vegetation index (NDVI) was adopted to analyze SPOT, Landsat, and QuickBird imagery to obtain the spatial distribution of vegetation covers. In the second stage, a maximum likelihood classification (MLC) program was used to classify mangrove and non-mangrove areas. The results indicated that the spatial distribution of mangroves expanded 15.18 and 40 ha in two monitoring sites in 10 years, demonstrating the success of establishing reserves for protecting mangrove habitats. The analytical results also indicated that satellite imagery can easily discern the difference in characteristics between imagery of mangrove and other vegetation types, and that the logistical disadvantages of monitoring long-term vegetation community changes as well as evaluating an inaccessible area may be overcome by applying remote sensing techniques.     

基于遥感的湿地景观格局季相分析

以中国东北地区三江平原北部为研究区域,利用2012年多季相遥感影像作为数据源,结合野外调查数据,应用面向对象的分类方法,根据影像的物候、时相等特征,提取不同月份的湿地信息,进行景观格局季相分析。结果表明:(1)研究区湿地面积、类型格局在同一年不同季节不同月份会有不同幅度的变化,总体呈现缓增骤减的态势。湿地主要分布在低洼地区,主要湿地类型为草本沼泽,其次为河流,其他湿地占总面积比例较小。(2)研究区各阶段湿地都有转化,主要发生在湿地和非湿地之间,多数表现在草本沼泽和草地之间的转化。(3)湿地分布和湿地转化面积主要集中在低海拔区域和低坡度区域,其中海拔100 m和坡度5°以下范围内的湿地分布面积和湿地转化面积占湿地总面积及湿地转化面积的绝大部分。(4)年内季节性湿地转化与降水、温度和湿地植被物候关系密切。     

Remote sensing imagery in vegetation mapping: a review

Aims Mapping vegetation through remotely sensed images involves various considerations, processes and techniques. Increasing availability of remotely sensed images due to the rapid advancement of remote sensing technology expands the horizon of our choices of imagery sources. Various sources of imagery are known for their differences in spectral, spatial, radioactive and temporal characteristics and thus are suitable for different purposes of vegetation mapping. Generally, it needs to develop a vegetation classification at first for classifying and mapping vegetation cover from remote sensed images either at a community level or species level. Then, correlations of the vegetation types (communities or species) within this classification system with discernible spectral characteristics of remote sensed imagery have to be identified. These spectral classes of the imagery are finally translated into the vegetation types in the image interpretation process, which is also called image processing. This paper presents an overview of how to use remote sensing imagery to classify and map vegetation cover.Methods Specifically, this paper focuses on the comparisons of popular remote sensing sensors, commonly adopted image processing methods and prevailing classification accuracy assessments.Important findings The basic concepts, available imagery sources and classification techniques of remote sensing imagery related to vegetation mapping were introduced, analyzed and compared. The advantages and limitations of using remote sensing imagery for vegetation cover mapping were provided to iterate the importance of thorough understanding of the related concepts and careful design of the technical procedures, which can be utilized to study vegetation cover from remote sensed images.     

Evaluation of a wetland classification system devised for management in a region with a high cover of peatlands: an example from the Cook Inlet Basin,Alaska

Several wetland classification schemes are now commonly used to describe wetlands in the contiguous United States to meet local, regional, and national regulatory requirements. However, these established systems have proven to be insufficient to meet the needs of land managers in Alaska. The wetlands of this northern region are predominantly peatlands, which are not adequately treated by the nationally-used systems, which have few, if any, peatland classes. A new system was therefore devised to classify wetlands in the rapidly urbanizing Cook Inlet Basin of southcentral Alaska, USA. The Cook Inlet Classification (CIC) is based on seven geomorphic and six hydrologic components that incorporate the environmental gradients responsible for the primary sources of variation in peatland ecosystems. The geomorphic and hydrologic components have the added advantage of being detectable on remote sensing imagery, which facilitates regional mapping across large tracts of inaccessible terrain. Three different quantitative measures were used to evaluate the robustness and performance of the CIC classes relative to that of other commonly used systems in the contiguous United States. The high within-group similarity of the classes identified by the CIC was clearly superior to that of the other systems, demonstrating the need for improved wetland classification systems specifically devised for regions with a high cover of peatlands.     

Mapping and monitoring of coastal wetlands of Çukurova Delta in the Eastern Mediterranean region

The aim of this research was to link vegetation characteristics, such as spatial and temporal distribution, and environmental variables, with land cover information derived from remotely sensed satellite images of the Eastern Mediterranean coastal wetlands of Turkey. The research method was based on (i) recording land cover characteristics by means of a vegetation indicator, and (ii) classifying and mapping coastal wetlands utilizing a Landsat Thematic Mapper (TM) image of Çukurova Deltas in Turkey. Vegetation characteristics of various habitats, such as sand dunes, salt marshes, salty plains and afforestation areas, were identified by field surveys. A Landsat TM image of 4 July 1993 was pre-processed and then classified using the Maximum Likelihood (ML) algorithm and Artificial Neural Networks (ANN). As a result of this supervised classification, the land cover types were classified with a largest accuracy of 90.2% by ANN. The classified satellite sensor imagery was linked to vegetation and bird census data, which were available through literature in a Geographical Information System (GIS) environment to determine the spatial distribution of plant and bird biodiversity in this coastal wetland. The resulting data provide an important baseline for further investigations such as monitoring, change detections and designing conservation policies in this coastal ecosystem.     

Use of aerial imagery to assess habitat suitability and predict site occupancy for a declining wetland-dependent bird

Remote sensing is a valuable tool for wetland habitat quantification, monitoring and assessment. Here we show that habitat assessment via aerial image inspection is useful in predicting wetland site occupancy by black terns (Chlidonias niger), an imperiled and declining species throughout much of North America. We used Google Earth^® images and National Wetlands Inventory maps to rank 390 candidate wetlands throughout Wisconsin (USA) according to their apparent suitability as nesting habitat for black terns and quantified habitat features associated with the suitability rankings. We then conducted ground-based suitability assessments and point counts of terns at most wetlands from May to July 2010. Pre-survey assessment resulted in 123 wetlands classified as suitable, 81 as marginal, and 186 as unsuitable. Wetlands ranked as suitable were more likely to be in the hemi-marsh stage, part of a wetland complex and relatively undisturbed. Black terns were present at 47 % of the wetlands considered suitable but only 11 % of the sites considered marginal or unsuitable. Of the 42 sites where nesting was confirmed, 79 % were at wetlands classified as suitable; no nesting was recorded in any wetlands deemed unsuitable. We found strong concordance in wetland suitability rankings between the two assessment methods (remote sensing, site surveys). We propose that remote sensing is an efficient and inexpensive way to predict site occupancy by wetland birds, such as black terns, that prefer a specific kind of habitat discernible from aerial imagery. This method may be particularly useful in areas, such as the Prairie Pothole region of North America, where ground surveys of all wetlands are not feasible.     

The missing wetlands: using local ecological knowledge to find cryptic ecosystems

Small, temporally dynamic, biologically diverse isolated wetlands are among the most imperiled ecosystems, yet their conservation is hindered by lack of protective legislation and mapping. As part of an effort to better understand isolated wetland ecology in an area undergoing dramatic land use change, we mapped isolated wetlands in South Carolina’s Piedmont and Blue Ridge regions using remote sensing and local ecological knowledge (LEK). Remote detection of isolated wetlands was limited by digital resource resolution, topography, and wetland size. LEK was the most useful tool for locating small isolated wetlands. We sampled 10% of the study area using LEK and discovered 44 wetlands with “isolated” characteristics, none of which had been identified by remote sensing. Only 8 of 44 wetlands found through LEK could be identified using remote sensing after their discovery. LEK fills a gap in cryptic ecosystem detection when adequate remotely sensed data are not available. Though effective, using LEK is neither as rapid nor as repeatable as remote sensing. We suggest a two-pronged approach for finding cryptic ecosystems: remote sensing coupled with LEK where data resolution is inadequate. For remote detection of isolated wetlands, we suggest a minimum resolution of 0.33 m for Color Infrared, leaf-off, high-water photography. Despite great advances in remote sensing, data are not uniformly available worldwide and LEK may serve as an effective tool for locating cryptic resources for biodiversity conservation. Mapping cryptic resources will allow for more accurate resource and biodiversity conservation planning under current and future climate scenarios.     

Vegetation Monitoring of Created Dune Swale Wetlands, Vandenberg Air Force Base, California

A monitoring program was established on San Antonio Terrace at Vandenberg Air Force Base to compare vegetation development at two created wetland sites and six nearby natural wetlands. The reference wetlands were chosen to represent a range of habitats in dune swale wetlands on the Terrace. Vegetation in the reference wetland plant communities varies from low-growing herbaceous marsh species with open canopies to closed canopies dominated by shrub or tree species. Transects and plots for long-term vegetation monitoring were established in all the wetlands, stratified by plant communities in the reference wetlands and by geomorphic location in the newly created wetlands. Quantitative vegetation and environmental data were collected at all the sites; measures included species distributions, species cover, and topographical elevations. Over the first three years of monitoring, variations in groundwater depth at different geomorphic locations in the created wetlands resulted in a variety of physical conditions for plant growth. In the first year, more than 100 plant species were observed, the majority being natives. During the next two years, species richness at the created wetland sites remained relatively stable and was higher than at the reference sites. Statistical comparisons of vegetation parameters by analysis of variance and hierarchical clustering exhibited patterns of increasing similarity between the created and reference wetlands. Long-term monitoring will be continued to track the progress of vegetation at the created sites, and to assess their development relative to the reference wetlands.