Dehua Zhao

Remote sensing of aquatic vegetation distribution in Taihu Lake using an improved classification tree with modified thresholds.

Classification trees (CT) have been used successfully in the past to classify aquatic vegetation from spectral indices (SI) obtained from remotely-sensed images. However, applying CT models developed for certain image dates to other time periods within the same year or among different years can reduce the classification accuracy. In this study, we developed CT models with modified thresholds using extreme SI values (CT(m)) to improve the stability of the models when applying them to different time periods. A total of 903 ground-truth samples were obtained in September of 2009 and 2010 and classified as emergent, floating-leaf, or submerged vegetation or other cover types. Classification trees were developed for 2009 (Model-09) and 2010 (Model-10) using field samples and a combination of two images from winter and summer. Overall accuracies of these models were 92.8% and 94.9%, respectively, which confirmed the ability of CT analysis to map aquatic vegetation in Taihu Lake. However, Model-10 had only 58.9-71.6% classification accuracy and 31.1-58.3% agreement (i.e., pixels classified the same in the two maps) for aquatic vegetation when it was applied to image pairs from both a different time period in 2010 and a similar time period in 2009. We developed a method to estimate the effects of extrinsic (EF) and intrinsic (IF) factors on model uncertainty using Modis images. Results indicated that 71.1% of the instability in classification between time periods was due to EF, which might include changes in atmospheric conditions, sun-view angle and water quality. The remainder was due to IF, such as phenological and growth status differences between time periods. The modified version of Model-10 (i.e. CT(m)) performed better than traditional CT with different image dates. When applied to 2009 images, the CT(m) version of Model-10 had very similar thresholds and performance as Model-09, with overall accuracies of 92.8% and 90.5% for Model-09 and the CT(m) version of Model-10, respectively. CT(m) decreased the variability related to EF and IF and thereby improved the applicability of the models to different time periods. In both practice and theory, our results suggested that CT(m) was more stable than traditional CT models and could be used to map aquatic vegetation in time periods other than the one for which the model was developed.

Spatio-Temporal Variability of Aquatic Vegetation in Taihu Lake over the Past 30 Years

It is often difficult to track the spatio-temporal variability of vegetation distribution in lakes because of the technological limitations associated with mapping using traditional field surveys as well as the lack of a unified field survey protocol. Using a series of Landsat remote sensing images (i.e. MSS, TM and ETM+), we mapped the composition and distribution area of emergent, floating-leaf and submerged macrophytes in Taihu Lake, China, at approximate five-year intervals over the past 30 years in order to quantify the spatio-temporal dynamics of the aquatic vegetation. Our results indicated that the total area of aquatic vegetation increased from 187.5 km2 in 1981 to 485.0 km2 in 2005 and then suddenly decreased to 341.3 km2 in 2010. Similarly, submerged vegetation increased from 127.0 km2 in 1981 to 366.5 km2 in 2005, and then decreased to 163.3 km2. Floating-leaf vegetation increased continuously through the study period in both area occupied (12.9 km2 in 1981 to 146.2 km2 in 2010) and percentage of the total vegetation (6.88% in 1981 to 42.8% in 2010). In terms of spatial distribution, the aquatic vegetation in Taihu Lake has spread gradually from the East Bay to the surrounding areas. The proportion of vegetation in the East Bay relative to that in the entire lake has decreased continuously from 62.3% in 1981, to 31.1% in 2005 and then to 21.8% in 2010. Our findings have suggested that drastic changes have taken place over the past 30 years in the spatial pattern of aquatic vegetation as well as both its relative composition and the amount of area it occupies.

Artificial regulation of water level and its effect on aquatic macrophyte distribution in Taihu Lake.

Management of water levels for flood control, water quality, and water safety purposes has become a priority for many lakes worldwide. However, the effects of water level management on the distribution and composition of aquatic vegetation has received little attention. Relevant studies have used either limited short-term or discrete long-term data and thus are either narrowly applicable or easily confounded by the effects of other environmental factors. We developed classification tree models using ground surveys combined with 52 remotely sensed images (15–30 m resolution) to map the distributions of two groups of aquatic vegetation in Taihu Lake, China from 1989–2010. Type 1 vegetation included emergent, floating, and floating-leaf plants, whereas Type 2 consisted of submerged vegetation. We sought to identify both inter- and intra-annual dynamics of water level and corresponding dynamics in the aquatic vegetation. Water levels in the ten-year period from 2000–2010 were 0.06–0.21 m lower from July to September (wet season) and 0.22–0.27 m higher from December to March (dry season) than in the 1989–1999 period. Average intra-annual variation (CVa) decreased from 10.21% in 1989–1999 to 5.41% in 2000–2010. The areas of both Type 1 and Type 2 vegetation increased substantially in 2000–2010 relative to 1989–1999. Neither annual average water level nor CVa influenced aquatic vegetation area, but water level from January to March had significant positive and negative correlations, respectively, with areas of Type 1 and Type 2 vegetation. Our findings revealed problems with the current management of water levels in Taihu Lake. To restore Taihu Lake to its original state of submerged vegetation dominance, water levels in the dry season should be lowered to better approximate natural conditions and reinstate the high variability (i.e., greater extremes) that was present historically.

A Hardy Plant Facilitates Nitrogen Removal via Microbial Communities in Subsurface Flow Constructed Wetlands in Winter.

The plants effect in subsurface flow constructed wetlands (SSF-CWs) is controversial, especially at low temperatures. Consequently, several SSF-CWs planted with Iris pseudacorus (CWI) or Typha orientalis Presl. (CWT) and several unplanted ones (CWC) were set up and fed with secondary effluent of sewage treatment plant during the winter in Eastern China. The 16S rDNA Illumina Miseq sequencing analysis indicated the positive effects of I. pseudacorus on the bacterial community richness and diversity in the substrate. Moreover, the community compositions of the bacteria involved with denitrification presented a significant difference in the three systems. Additionally, higher relative abundances of nitrifying bacteria (0.4140%, 0.2402% and 0.4318% for Nitrosomonas, Nitrosospira and Nitrospira, respectively) were recorded in CWI compared with CWT (0.2074%, 0.0648% and 0.0181%, respectively) and CWC (0.3013%, 0.1107% and 0.1185%, respectively). Meanwhile, the average removal rates of NH4+-N and TN in CWI showed a prominent advantage compared to CWC, but no distinct advantage was found in CWT. The hardy plant I. pseudacorus, which still had active root oxygen release in cold temperatures, positively affected the abundance of nitrifying bacteria in the substrate, and accordingly was supposed to contribute to a comparatively high nitrogen removal efficiency of the system during the winter.

A Method for Application of Classification Tree Models to Map Aquatic Vegetation Using Remotely Sensed Images from Different Sensors and Dates

In previous attempts to identify aquatic vegetation from remotely-sensed images using classification trees (CT), the images used to apply CT models to different times or locations necessarily originated from the same satellite sensor as that from which the original images used in model development came, greatly limiting the application of CT. We have developed an effective normalization method to improve the robustness of CT models when applied to images originating from different sensors and dates. A total of 965 ground-truth samples of aquatic vegetation types were obtained in 2009 and 2010 in Taihu Lake, China. Using relevant spectral indices (SI) as classifiers, we manually developed a stable CT model structure and then applied a standard CT algorithm to obtain quantitative (optimal) thresholds from 2009 ground-truth data and images from Landsat7-ETM+, HJ-1B-CCD, Landsat5-TM and ALOS-AVNIR-2 sensors. Optimal CT thresholds produced average classification accuracies of 78.1%, 84.7% and 74.0% for emergent vegetation, floating-leaf vegetation and submerged vegetation, respectively. However, the optimal CT thresholds for different sensor images differed from each other, with an average relative variation (RV) of 6.40%. We developed and evaluated three new approaches to normalizing the images. The best-performing method (Method of 0.1% index scaling) normalized the SI images using tailored percentages of extreme pixel values. Using the images normalized by Method of 0.1% index scaling, CT models for a particular sensor in which thresholds were replaced by those from the models developed for images originating from other sensors provided average classification accuracies of 76.0%, 82.8% and 68.9% for emergent vegetation, floating-leaf vegetation and submerged vegetation, respectively. Applying the CT models developed for normalized 2009 images to 2010 images resulted in high classification (78.0%–93.3%) and overall (92.0%–93.1%) accuracies. Our results suggest that Method of 0.1% index scaling provides a feasible way to apply CT models directly to images from sensors or time periods that differ from those of the images used to develop the original models.

Estimation of leaf area index and plant area index of a submerged macrophyte canopy using digital photography.

Non-destructive estimation using digital cameras is a common approach for estimating leaf area index (LAI) of terrestrial vegetation. However, no attempt has been made so far to develop non-destructive approaches to LAI estimation for aquatic vegetation. Using the submerged plant species Potamogeton malainus, the objective of this study was to determine whether the gap fraction derived from vertical photographs could be used to estimate LAI of aquatic vegetation. Our results suggested that upward-oriented photographs taken from beneath the water surface were more suitable for distinguishing vegetation from other objects than were downward-oriented photographs taken from above the water surface. Exposure settings had a substantial influence on the identification of vegetation in upward-oriented photographs. Automatic exposure performed nearly as well as the optimal trial exposure, making it a good choice for operational convenience. Similar to terrestrial vegetation, our results suggested that photographs taken for the purpose of distinguishing gap fraction in aquatic vegetation should be taken under diffuse light conditions. Significant logarithmic relationships were observed between the vertical gap fraction derived from upward-oriented photographs and plant area index (PAI) and LAI derived from destructive harvesting. The model we developed to depict the relationship between PAI and gap fraction was similar to the modified theoretical Poisson model, with coefficients of 1.82 and 1.90 for our model and the theoretical model, respectively. This suggests that vertical upward-oriented photographs taken from below the water surface are a feasible alternative to destructive harvesting for estimating PAI and LAI for the submerged aquatic plant Potamogeton malainus.

Decreasing but still significant facilitation effect of cold-season macrophytes on wetlands purification function during cold winter

To identify the facilitation effect of a cool-season aquatic macrophyte (FEam) for use in effluent purification via constructed floating wetlands (CFWs) and to determine the possible pathways used during a winter period with an average temperature of less than 5 °C, pilot-scale CFWs were planted with the cold-season macrophyte Oenanthe clecumbens and were operated as batch systems. Although some leaves withered, the roots retained relatively high levels of activity during the winter, which had average air and water temperatures of 3.63 and 5.04 °C, respectively. The N and P removal efficiencies in CFWs decreased significantly in winter relative to those in late autumn. The presence of cool-season plants resulted in significant improvements in N and P removal, with a FEam of 15.23–25.86% in winter. Microbial N removal accounted for 71.57% of the total N removed in winter, and the decrease in plant uptake was the dominant factor in the wintertime decrease in N removal relative to that in late autumn. These results demonstrate the importance of cold-season plants in CFWs for the treatment of secondary effluent during cold winters.

Performance of a large-scale wetland treatment system in treating tailwater from a sewage treatment plant

Water quality standards pertaining to effluent from sewage treatment plants (STPs) in China have become more stringent, requiring upgrading of STPs and entailing huge capital expenditure. Wetland treatment systems (WTSs) are a low-cost and highly efficient approach for deep purification of tailwater from STPs. The Hongze WTS (HZ-WTS), a large-scale surface-flow constructed wetland, with a total area of 55.58ha and a treatment capacity of 4×104m3day–1, was built for the disposal of tailwater from STPs. The aim of the present study was to evaluate the performance of HZ-WTP with regard to seasonal variations and to compare treatment costs with those of other STPs. The performance of the HZ-WTS was evaluated in 2013 using online monitoring. HZ-WTS exhibited significant removal efficiency of ammonia nitrogen (NH4+-N), chemical oxygen demand and total phosphorus (mean±s.d., percentage removal efficiency 56.33±70.44, 55.64±18.58 and 88.44±22.71% respectively), whereas there was significant seasonal variation in the efficiency of NH4+-N removal. In addition, the average treatment cost was ¥0.17m–3, significantly lower than the corresponding value for other STPs. Therefore, WTSs are recommended for use with STPs in order to improve waste water quality in a cost-effective manner.

Elevated salinity inhibits nitrogen removal by changing the microbial community composition in constructed wetlands during the cold season

In the present study we investigated whether subsurface flow constructed wetlands (SSF-CWs) can remove nitrogen from saline waste water and whether salinity affects nitrogen removal during the cold season (mean water temperature <10°C). Eight Iris pseudacorus-planted SSF-CWs were fed with normal (salinity 1.3–1.5‰; CWP) or saline (salinity 6.3–6.5‰; CWP+) waste water; similarly, eight unplanted SSF-CWs were fed with normal (CWU) or saline waste water (CWU+). The systems were run continuously at a hydraulic loading rate of 187.5mmday–1 and a hydraulic retention time of 4 days. Nitrogen removal efficiency, plant parameters and bacterial abundance and community composition were measured. In CWP, 80% of NH4+-N and 52% of total nitrogen (TN) were removed. In contrast, the removal rates of NH4+-N and TN in CWP+ were reduced by 27 and 37% respectively. In the presence of higher salinity, not only were there decreases in plant biomass (32.1%) and nitrogen uptake (50.1%), but the growth, activity and oxygen release of roots were also reduced (by 37.8, 68.0 and 62.9% respectively). Bacterial community composition also differed under conditions of elevated salinity. Elevated salinity is associated with lower nitrogen removal in SSF-CWs, which we speculate is a result of suppressed wetland macrophyte growth and activity, as well as changes in microbial community composition.

Nitrogen removal during the cold season by constructed floating wetlands planted with Oenanthe javanica

Constructed floating wetlands (CFWs) are used to treat waste waters of various origins either alone or as part of waste water treatment trains. The aim of the present study was to determine the extent of nitrogen removal by CFWs planted with Oenanthe javanica (Blume) DC. at low temperatures (<10°C) and whether CFWs with vesuvianite as a substrate perform better than those without substrate. A batch model was used, with CFWs planted with O. javanica (Tc), CFWs without O. javanica (Ts), CFWs without substrate (Tp) and floating mats only (To) as a control. The average removal rates of NH4+-N, NO3–-N and total nitrogen were 78.3, 44.4 and 49.7% respectively in Tc; 72.0, 40.0 and 39.5% respectively in Ts; and 73.1, 33.7 and 44.0% respectively in Tp. In addition to a gradual increase in chemical oxygen demand during the experimental period, Tc had higher microbial richness and diversity, as well as a higher abundance of bacteria, archaea, anaerobic ammonium oxidation (Anammox) bacteria and key genes (ammonia mono-oxygenase, amoA, nitrous oxide reductase, nosZ, dissimilatory cd1-containing nitrite reductase, nirS, and dissimilatory copper-containing nitrite reductase, nirK) involved in nitrogen metabolism in the substrate than Ts. Further analysis of microbial community composition revealed a difference at multiple taxonomic levels among different systems. These results demonstrate the positive roles of O. javanica and vesuvianite in CFWs in nitrogen removal from waste water during the cold season (mean water temperature <10°C).