Guo Y.,Capital Normal University |
Guo Y.,Laboratory of 3D Information Acquisition and Application |
Guo Y.,Beijing Municipal Key Laboratory of Resources Environment and GIS |
Gong H.,Urban Environmental Processes and Digital Modeling Laboratory |
And 5 more authors.
Applied Microbiology and Biotechnology
Wetland plant rhizosphere microorganisms play a significant role in the purification of ecological restoration of reclaimed water replenishment wetlands. In this study, water quality discriminant analysis indicated the wetland had a distinctive role in the purification of total nitrogen (TN), total phosphorus (TP), and nitrate (NO3 −) from reclaimed water, of which removal rates were 42.15, 47.34, and 28.56 % respectively. All the sequences of 16S ribosomal RNA (16S rRNA) gene clone library were affiliated with Proteobacteria (74.50 %), Bacteroidetes (6.54 %), Gemmatimonadetes (5.88 %), Chloroflexi (4.25 %), Chlorobi (2.94 %), Nitrospira (2.61 %), Acidobacteria (2.29 %), and Actinobacteria (0.98 %). Assessment of water quality purification and rhizosphere bacterial properties revealed that the major biogeochemical reactions were nitrogen, phosphorus, carbon, and sulfur cycles (33.70, 15.40, 14.40, and 4.90 %, respectively). In addition, approximately 5.90 and 4.60 % of the clones are closely related with the minor biogeochemical degradations of antibiotics and halogenated hydrocarbons, which were the typical characteristics of reclaimed water wetland different from freshwater wetlands. The finding of water quality discriminant is consistent with that of bacterial community, but the latter was a more powerful method than the former which reveals possible implications of wetland plant purification on the reclaimed water. © 2014, Springer-Verlag Berlin Heidelberg. Source
Guo C.,Capital Normal University |
Guo C.,Urban Environmental Processes and Digital Modeling Laboratory |
Guo C.,Laboratory of 3D Information Acquisition and Application |
Guo X.,Capital Normal University |
And 2 more authors.
Increasing amounts of recycled water have been used in wetlands; as such, monitoring the growth status of wetland plants with remote sensing technology is of great significance to restore and reconstruct wetland added by recycled water. In this paper, Phragmites australis and Typha angustifolia were selected. Spectral reflectance was determined in the visible domain to avoid the influence of leaf water on the established model; chlorophyll and nitrogen contents were measured as data sources. Statistical analysis, including correlation analysis and stepwise regression analysis, was conducted to estimate chlorophyll and nitrogen contents with different derivative transform spectra, namely, first derivative reflectance spectrum (FDS), band depth (BD), continuum-removed derivative reflectance (CRDR), band depth ratio (BDR), and normalized band depth index (NBDI). After the selected bands were statistically analyzed, we found that the range of 550-750 nm, particularly 700-750 nm (red edge range), was the key range to estimate biochemical parameters. Single removal cross-validation results showed that the optimum models of chlorophyll content inversion in reeds, cattails, and combined samples were BD, CRDR, and NBDI, respectively; the corresponding R2 were 0.87, 0.83, and 0.81. The respective RMSE values were 0.16, 0.15, and 0.33; the optimal models of nitrogen content inversion in reeds, cattails, and combined samples were BD, CRDR, and BD, respectively; the corresponding R2 were 0.69, 0.72, and 0.79. The respective RMSE values were 0.15, 0.11, and 0.17. Kruskal-Wallis non-parametric test and multi-way ANOVA were then performed; results revealed that data types and biological differences greatly influence the inversion model; by contrast, spectral derivative transform and vegetation type difference did not significantly affect the inversion model. Our results could be used as a scientific basis of the non-destructive growth status monitoring of wetland plants. It also provided a reference for restoration and management of wetlands supplied with recycled water. © 2016 Taylor and Francis Group, LLC. Source
Wang Z.,Capital Normal University |
Li X.,Capital Normal University |
Li X.,Urban Environmental Processes and Digital Modeling Laboratory |
Li X.,Laboratory of 3D Information Acquisition and Application |
And 5 more authors.
Proceedings of SPIE - The International Society for Optical Engineering
In order to take full advantage of the multi-source remote sensing data, the concept of image fusion was proposed and nowadays, many fusion algorithms have been produced, so the adaptability of the algorithm has become a hot topic. This article aiming at the panchromatic and multispectral images of Worldview-2, uses the different fusion methods, including PCA, HPF, Gram-Schmidt and wavelet transform, to fuse the image, and evaluates the results from the aspects of the statistical information, spectral information and spatial characteristics. Then it takes the extracting vegetation as an example, analyzes the applicability of different fusion methods. The research shows that Gram-Schmidt and wavelet transform have better fusion quality and Gram-Schmidt is the most suitable method to extract vegetation for Worldview-2. © 2014 SPIE. Source
Guo C.F.,Capital Normal University |
Guo C.F.,Urban Environmental Processes and Digital Modeling Laboratory |
Guo C.F.,Laboratory of 3D Information Acquisition and Application |
Guo C.F.,Beijing Municipal Key Laboratory of Resources Environment and GIS |
And 4 more authors.
Advanced Materials Research
Wetlands is a superior way to purify water environment. However monitoring and discriminating among vegetation covers types are critical to understanding population distribution, biogeochemical functioning and the process of wetland recovery. Recently, remote sensing technology has become an important tool to monitor wetland vegetation. Typical wetland plants Bidens Pilosa, Scirpus planiculumis, Phragmites australis and Typha angustifolia in Wild duck wetland were chosen, and Mann-Whitney U-test were used to analyze the spectral characteristics. There were typical vegetation spectral characteristics of dominant species spectral reflectance curve. There are differences among reflective spctral characteristic of species, especially in the "green peak" and "steep reflection". According to the results of Mann-Whitney U-test, the highest frequency bands appeared in702-715,732-745,747nm, the classification precision of Bidens pilosa, Scirpus planiculumis and Phragmites australis are 100%, 100%, 80%. While It is difficult to distinguish Typha angustifolia from other species effectively using the three spectrum regions. © (2014) Trans Tech Publicutions, Switzerland. Source
Hu Z.,Capital Normal University |
Hu Z.,Key Laboratory of Resources Environment |
Hu Z.,Laboratory of 3D Information Acquisition and Application |
Wang Z.,Tianjin Institute of Urban Construction |
And 4 more authors.
Journal of Natural Disasters
In this study, a road flooding information extraction model based on high resolution satellite remote sensing imagery was researched. A road flooding information extraction system was developed through the realization of inter-programming among different software environments by combining remote sensing and geographic information system technologies. First, road information extraction rules were built by analyzing spectral, geometrical and textural characteristics of fused image, which can be generated from original high resolution multi-spectral satellite remote sensing imagery. The rules were utilized to make an object-oriented road information extraction model. Second, with the knowledge of pure water s very low reflecting rate in near and mid-infrared electromagnetic wave band, and considering the complexity caused by different environmental background, an adaptive water information extraction model was built. On above bases, a component-based communication and data sharing mechanism between the application environment of remote sensing and geographic information system was established. An interactive user interface that supports model running was then developed. It can provide a spatial overlap function, extract malfunctioned road from original roads, and apply to systems with specific requirements of road flooding information extraction for correlative works. Source