Influence of UV radiation on chlorophyll, and antioxidant enzymes of wetland plants in different types of constructed wetland

A surface- and vertical subsurface-flow-constructed wetland were designed to study the response of chlorophyll and antioxidant enzymes to elevated UV radiation in three types of wetland plants (Canna indica, Phragmites austrail, and Typha augustifolia). Results showed that (1) chlorophyll content of C. indica, P. austrail, and T. augustifolia in the constructed wetland was significantly lower where UV radiation was increased by 10 and 20 % above ambient solar level than in treatment with ambient solar UV radiation (p < 0.05). (2) The malondialdehyde (MDA) content, guaiacol peroxidase (POD), superoxide dismutase (SOD), and catalase (CAT) activities of wetland plants increased with elevated UV radiation intensity. (3) The increased rate of MDA, SOD, POD, and CAT activities of C. indica, P. australis, and T. angustifolia by elevated UV radiation of 10 % was higher in vertical subsurface-flow-constructed wetland than in surface-flow-constructed wetland. The sensitivity of MDA, SOD, POD, and CAT activities of C. indica, P. austrail, and T. augustifolia to the elevated UV radiation was lower in surface-flow-constructed wetland than in the vertical subsurface-flow-constructed wetland, which was related to a reduction in UV radiation intensity through the dissolved organic carbon and suspended matter in the water. C. indica had the highest SOD and POD activities, which implied it is more sensitive to enhanced UV radiation. Therefore, different wetland plants had different antioxidant enzymes by elevated UV radiation, which were more sensitive in vertical subsurface-flow-constructed wetland than in surface-flow-constructed wetland.

Fast identification algorithms for Gaussian process model

A class identification algorithms is introduced for Gaussian process(GP)models.The fundamental approach is to propose a new kernel function which leads to a covariance matrix with low rank,a property that is consequently exploited for computational efficiency for both model parameter estimation and model predictions.The objective of either maximizing the marginal likelihood or the Kullback–Leibler (K–L) divergence between the estimated output probability density function(pdf)and the true pdf has been used as respective cost functions.For each cost function,an efficient coordinate descent algorithm is proposed to estimate the kernel parameters using a one dimensional derivative free search, and noise variance using a fast gradient descent algorithm. Numerical examples are included to demonstrate the effectiveness of the new identification approaches.

Estimation of Gaussian process regression model using probability distance measures

A new class of parameter estimation algorithms is introduced for Gaussian process regression (GPR) models. It is shown that the integration of the GPR model with probability distance measures of (i) the integrated square error and (ii) Kullback–Leibler (K–L) divergence are analytically tractable. An efficient coordinate descent algorithm is proposed to iteratively estimate the kernel width using golden section search which includes a fast gradient descent algorithm as an inner loop to estimate the noise variance. Numerical examples are included to demonstrate the effectiveness of the new identification approaches.

Gaussian processes autoencoder for dimensionality reduction

Learning low dimensional manifold from highly nonlinear data of high dimensionality has become increasingly important for discovering intrinsic representation that can be utilized for data visualization and preprocessing. The autoencoder is a powerful dimensionality reduction technique based on minimizing reconstruction error, and it has regained popularity because it has been efficiently used for greedy pretraining of deep neural networks. Compared to Neural Network (NN), the superiority of Gaussian Process (GP) has been shown in model inference, optimization and performance. GP has been successfully applied in nonlinear Dimensionality Reduction (DR) algorithms, such as Gaussian Process Latent Variable Model (GPLVM). In this paper we propose the Gaussian Processes Autoencoder Model (GPAM) for dimensionality reduction by extending the classic NN based autoencoder to GP based autoencoder. More interestingly, the novel model can also be viewed as back constrained GPLVM (BC-GPLVM) where the back constraint smooth function is represented by a GP. Experiments verify the performance of the newly proposed model.

Vertical structure and physical processes of the Madden-Julian oscillation: exploring key model physics in climate simulations

Aimed at reducing deficiencies in representing the Madden-Julian oscillation (MJO) in general circulation models (GCMs), a global model evaluation project on vertical structure and physical processes of the MJO was coordinated. In this paper, results from the climate simulation component of this project are reported. It is shown that the MJO remains a great challenge in these latest generation GCMs. The systematic eastward propagation of the MJO is only well simulated in about one-fourth of the total participating models. The observed vertical westward tilt with altitude of the MJO is well simulated in good MJO models, but not in the poor ones. Damped Kelvin wave responses to the east of convection in the lower troposphere could be responsible for the missing MJO preconditioning process in these poor MJO models. Several process-oriented diagnostics were conducted to discriminate key processes for realistic MJO simulations. While large-scale rainfall partition and low-level mean zonal winds over the Indo-Pacific in a model are not found to be closely associated with its MJO skill, two metrics, including the low-level relative humidity difference between high and low rain events and seasonal mean gross moist stability, exhibit statistically significant correlations with the MJO performance. It is further indicated that increased cloud-radiative feedback tends to be associated with reduced amplitude of intraseasonal variability, which is incompatible with the radiative instability theory previously proposed for the MJO. Results in this study confirm that inclusion of air-sea interaction can lead to significant improvement in simulating the MJO.

The forms of bioavailability of phosphorus in integrated vertical flow constructed wetland with earthworms and different substrates

A sequential extraction method was utilized to analyze seven forms of P in an integrated vertical-flow constructed wetland (IVFCW) containing earthworms and different substrates. The aluminum-bound P (Al-P) content was found to be lower, and the occluded P (Oc-P) content was higher in the IVFCW. The addition of earthworms into the influent chamber of IVFCW increased the exchange P (Ex-P), iron-bound P (Fe-P), calcium bound P (Ca-P), Oc-P, detritus-bound (De-P) and organic P (Org-P) content in the influent chamber, and also enhanced P content uptake by wetland plants. A significantly positive correlation between P content of above-ground wetland plants and the Ex-P, Fe-P, Oc-P and Org-P content in the rhizosphere was found (P < 0.05), which indicated that the Ex-P, Fe-P, Oc-P and Org-P could be bio-available P. The Ex-P, Fe-P, De-P, Oc-P and Ca-P content of the influent chamber was higher where the substrate contained a mixture of Qing sand and river sand rather than only river sand. Also the IVFCW with earthworms and both Qing sand and river sand had a higher removal efficiency of P, which was related to higher P content uptake by wetland plants and P retained in IVFCW. These findings suggest that addition of earthworms in IVFCW increases the bioavailable P content, resulting in enhanced P content uptake by wetland plants.