Using artificial neural network for reservoir eutrophication prediction

Reservoirs provide approximately 70% of water supply for domestic and industrial use in Taiwan. The water quality of reservoirs is now one of the key factors in the operation and water quality management of reservoirs. Transient weather patterns result in highly variable magnitudes of precipitation and thereby sharp fluctuations in the surface elevation of the reservoirs. In addition, excessive watershed development in the past two decades has contributed to continuing increase in nutrient loads to the reservoirs. The difficulty in quantifying watershed nutrient loads and uncentainties in kinetic mechanism in the water column present a technical challenge to the mass balance based modeling of reservoir eutrophication. This study offers an alternative approach to quantifying the cause-and-effect relationship in reservoir eutrophication with a data-driven method, i.e., capturing non-linear relationships among the water quality variables in the reservoir. A commonly used back-propagation neural network was used to relate the key factors that influence a number of water quality indicators such as dissolved oxygen (DO), total phosphorus (TP), chlorophyll-a (Chl-a), and secchi disk depth (SD) in a reservoir in central Taiwan. Study results show that the neural network is able to predict these indicators with reasonable accuracy, suggesting that the neural network is a valuable tool for reservoir management in Taiwan.     

Prediction of longitudinal dispersion coefficients in natural rivers using artificial neural network

An artificial neural network (ANN) model is developed for predicting the longitudinal dispersion coefficient in natural rivers. The model uses few rivers’ hydraulic and geometric characteristics, that are readily available, as the model input, and the target output is the longitudinal dispersion coefficient (K). For performance evaluation of the model, using published field data, predictions by the developed ANN model are compared with those of other reported important models. Based on various performance indices, it is concluded that the new model predicts the longitudinal dispersion coefficient more accurately. Sensitive analysis performed on input parameters indicates stream width, flow depth, stream sinuosity, flow velocity, and shear velocity to be the most influencing parameters for accurate prediction of the longitudinal dispersion coefficient.     

Applying artificial neural network methodology to ocean color remote sensing

Artificial neural networks (ANN) are widely used as continuous models to fit non-linear transfer functions. In this study we used ANN to retrieve chlorophyll pigments in the near-surface of oceans from Ocean Color measurements. This bio-optical inversion is established by analyzing concomitant sun-light spectral reflectances over the ocean surface and pigment concentration. The relationships are complex, non-linear, and their biological nature implies a significant variability. Moreover, the sun-light reflectances are usually measured by satellite radiometers flying at 800 km over the ocean surface, which affect the data by adding radiometric noise and atmospheric correction errors. By comparison with the polynomial fit usually employed to treat this problem, we show the advantages of neural function approximation like the association of non-linear complexity and noise filtering.     

神经网络模型森林生物量遥感估测方法的研究

森林生物量的估测是全球变化研究的基础,而遥感宏观、综合、动态、快速的特点决定了基于遥感的生物量模型为森林生物量估测的发展方向,目前的遥感生物量估测方法大多基于回归分析,需要预先假设、事后检验,仅为经验性的统计模型。神经网络的分布并行处理、非线性映射、自适应学习和容错等特性,使其具有独特的信息处理和计算能力,在机制尚不清楚的高维非线性系统体现出强大优势,可以用于遥感生物量估测。文章在野外调查的基础上,尝试应用BP网络和RBF网络技术,建立广州TM遥感影像数据与森林样方生物量实测数据之间的神经网络模型,通过训练和仿真,与生物量实测数据进行比较。结果表明,在独立样地估测中,人工神经网络估测的相对误差均小于15.18%,获得了满意的效果。而RBF网络与BP网络相比,在识别精度上、稳定性、速度上,均优于BP网络,其最大相对误差不超过10.12%,平均相对误差为4.76%。可见应用神经网络方法的“黑箱”操作虽然难以归纳出指导性规律,但可以获得很高的精度。尤其RBF网络,在训练完成后,可以应用该模型进行大区域生物量估算,对于森林的规划及管理具有深远意义。     

Estimation of stream temperature in firtina creek (Rize-Turkiye) using artificial neural network model

Water temperature is one of the most important environmental variables in aquatic ecosystem. Temperature changes may have positive or negative effects on organisms. High water temperatures have caused mortalities in salmonid fishes. Therefore, monitoring and prediction of potential adverse changes in water temperature is very important. Here, we have developed and tested an artificial neural network (ANN) model to predict stream temperature of Firtina Creekin Black Sea region; using local water temperature, dissolved oxygen, pH and other available meteorological data (air temperature, rainfall). Thus, enabling define suitable habitat for native Sea Trout (Salmo trutta labrax, Pallas 1811) under past drought or other adverse envIronmental conditions.     

Review and comparison of methods to study the contribution of variables in artificial neural network models

Convinced by the predictive quality of artificial neural network (ANN) models in ecology, we have turned our interests to their explanatory capacities. Seven methods which can give the relative contribution and/or the contribution profile of the input factors were compared: (i) the ‘PaD’ (for Partial Derivatives) method consists in a calculation of the partial derivatives of the output according to the input variables; (ii) the ‘Weights’ method is a computation using the connection weights; (iii) the ‘Perturb’ method corresponds to a perturbation of the input variables; (iv) the ‘Profile’ method is a successive variation of one input variable while the others are kept constant at a fixed value; (v) the ‘classical stepwise’ method is an observation of the change in the error value when an adding (forward) or an elimination (backward) step of the input variables is operated; (vi) ‘Improved stepwise a’ uses the same principle as the classical stepwise, but the elimination of the input occurs when the network is trained, the connection weights corresponding to the input variable studied is also eliminated; (vii) ‘Improved stepwise b’ involves the network being trained and fixed step by step, one input variable at its mean value to note the consequences on the error. The data tested in this study concerns the prediction of the density of brown trout spawning redds using habitat characteristics. The PaD method was found to be the most useful as it gave the most complete results, followed by the Profile method that gave the contribution profile of the input variables. The Perturb method allowed a good classification of the input parameters as well as the Weights method that has been simplified but these two methods lack stability. Next came the two improved stepwise methods (a and b) that both gave exactly the same result but the contributions were not sufficiently expressed. Finally, the classical stepwise methods gave the poorest results.     

应用自组织人工神经网络预报黄土高原生态破坏程度的初探

运用三维ＴＫｏｈｏｎｅｎ自组织人工神经网络，分析预测黄土高原生态经济破坏程度，预测成功率１００％。结果表明，神经网络方法性能良好，可望成为生态经济破坏程度预测的有效的辅助手段。     

基于人工神经网络方法的河北省近海沉积物重金属污染综合评价

利用误差反相传播神经（BP）网络对河北省近海沉积物中的铅、镉、锌、汞、砷5种重金属元素的污染水平进行分析,利用自组织特征映射（SOFM）网络对上述重金属元素分布特征进行分类,通过分类与污染水平量化值的结合,进行综合评价。SOFM把52个沉积物样品分别划分为3、4、6类和9类。对比各种分类,分为3类的物理意义较明确。每个类别分别对应高中低不同的污染物浓度水平,差异显著、分类方式比较合理。通过此种分类可以判断河北省近海的沉积物重金属污染在不同海域存在一定的差别,整体上是离海岸越远,沉积物的重金属污染水平越高,距海岸较近的海域内,沉积物的重金属污染水平较低,但渤海湾内的重金属污染水平高于其他海域。     

基于人工神经网络方法的河北省近海沉积物重金属污染综合评价

利用误差反相传播神经(BP)网络对河北省近海沉积物中的铅、镉、锌、汞、砷5种重金属元素的污染水平进行分析,利用自组织特征映射(SOFM)网络对上述重金属元素分布特征进行分类,通过分类与污染水平量化值的结合,进行综合评价。SOFM把52个沉积物样品分别划分为3、4、6类和9类。对比各种分类,分为3类的物理意义较明确,每个类别分别对应高中低不同的污染物浓度水平,差异显著、分类方式比较合理。通过此种分类可以判断河北省近海的沉积物重金属污染在不同海域存在一定的差别,整体上是离海岸越远,沉积物的重金属污染水平越高,距海岸较近的海域内,沉积物的重金属污染水平较低,但渤海湾内的重金属污染水平高于其他海域。     

Modeling and inference of animal movement using artificial neural networks

Movement of animals in relation to objects in their environment is important in many areas of ecology and wildlife conservation. Tools for analysis of movement data, however, still remain rather limited. In previous work, we developed nonlinear regression models for movement in relation to a single landscape feature. Here we greatly expand these previous models by using artificial neural networks. The new models add substantial flexibility and capabilities, including the ability to incorporate multiple factors and covariates. We devise a likelihood-based model fitting procedure that utilizes genetic algorithms and demonstrate the approach with movement data for red diamond rattlesnakes. The proposed methodology can be useful for global positioning system tracking data that are becoming more common in studies of animal movement behavior.     

Optimization of an artificial neural network for identifying fishing set positions from VMS data: An example from the Peruvian anchovy purse seine fishery

The spatial behavior of numerous fishing fleets is nowadays well documented thanks to satellite Vessel Monitoring Systems (VMS). Vessel positions are recorded on a frequent and regular basis which opens promising perspectives for improving fishing effort estimation and management. However, no specific information is provided on whether the vessel is fishing or not. To answer that question, existing works on VMS data usually apply simple criteria (e.g. threshold on speed). Those simple criteria generally focus in detecting true positives (a true fishing set detected as a fishing set); conversely, estimation errors are given no attention. For our case study, the Peruvian anchovy fishery, those criteria overestimate the total number of fishing sets by 182%. To overcome this problem an artificial neural network (ANN) approach is presented here. In order to set both the optimal parameterization and use “rules” for this ANN, we perform an extensive sensitivity analysis on the optimization of (1) the internal structure and training algorithm of the ANN and (2) the “rules” used for choosing both the relative size and the composition of the databases (DBs) used for training and inferring with the ANN. The “optimized” ANN greatly improves the estimates of the number and location of fishing events. For our case study, ANN reduces the total estimation error on the number of fishing sets to 1% (in average) and obtains 76% of true positives. This spatially explicit information on effort, provided with error estimation, should greatly reduce misleading interpretations of catch per unit effort and thus significantly improve the adaptive management of fisheries. While fitted on Peruvian anchovy fishery data, this type of neural network approach has wider potential and could be implemented in any fishery relying on both VMS and at-sea observer data. In order to increase the accuracy of the ANN results, we also suggest some criteria for improving sampling design by at-sea observers and VMS data.     

Pesticide prediction in ground water in North Carolina domestic wells using artificial neural networks

This paper illustrates the application of artificial neural networks (ANN) for prediction of pesticide occurrence in rural domestic wells from the available limited information. Among the three ANN models (a feed-forward back propagation [BP], a radial basis function [RBF] and an adaptive neural network-based fuzzy inference system [ANFIS]) employed for this investigation, the BP neural network was found to be superior to RBF and ANFIS type networks for the detection of pesticide occurrences in wells. For improved model prediction efficiency, optimization of network structure (e.g., number of hidden layers and number of nodes in each hidden layer) and spread (the width of the radial basis function) are important for BP and RBF type of network, respectively. A four layer BP network with a 3:2 neurons ratio of the first hidden layer to the second hidden layer produced better prediction performance efficiencies in terms of the pesticide detection efficiency (E_f), the root mean square error (RMSE), and the correlation coefficient (R) and the overall E_f of the BP neural network was found greater than 85%. Sensitivity analysis was performed to measure the relative importance of one input parameter over the other in pesticide occurrence in wells. It was shown in terms of the prediction efficiencies (E_f, RMSE, and R) of a four-layer BP neural network that the time of sample collection (TSC; month of the year), the depths of wells, and pesticide travel times (PTT) were more important parameters in the prediction of the pesticide occurrences in rural domestic wells. This means that the wells having shallow ground water table are more vulnerable to pesticide occurrence.     

Prediction of sulphur removal with Acidithiobacillus sp. using artificial neural networks

Artificial neural network (ANN) model was used to predict the extent of sulphur removal from three types of coal using native cultures of Acidithiobacillus ferrooxidans. The type of coal, initial pH, pulp density, particle size, residence time, media composition and initial sulphur content of coal were fed as input to the network. The output of the model was sulphur removal. The resulting ANN showed satisfactory prediction of sulphur removal percentages with mean absolute deviations varying from 0.003 to 0.5. A three layer feed forward neural network model consisting of an input layer, one hidden layer and an output layer was found to give satisfactory results. Although the number of data sets were limited, the parity plot shows that the model estimations for the test set was good.     

Testing the generalization of artificial neural networks with cross-validation and independent-validation in modelling rice tillering dynamics

Neural networks (NN) rely on the inner structure of available data sets rather than on comprehension of the modeled processes between inputs and outputs. Therefore, neural networks have been regarded as highly empirical models with limited extrapolation capability to situations outside the range of the training and validation data sets. In this study, the generalization ability of neural networks in predicting rice tillering dynamics was tested and several techniques inducing the generalization ability of neural networks were compared. We compared the performance of cross-validated neural networks with independent-validated neural networks and found that neural networks were able to extrapolate and predict tillering dynamics if the data were within the range of inputs of the training set. An inadequate training set resulted in overfitting of available data and neural networks that were not generalizable. The training set size required to enable a neural network to generalize and predict rice tillering dynamics was found to be at least 9 times as many training patterns for each weight. When a large number of variables are included in the input vector of a neural network with inadequate amounts of training data, we strongly recommend that the dimension of the input vector is reduced using principle component analysis (PCA), correspondence analysis (CA) or similar techniques to decrease the number of weights before the training procedure to improve the generalization ability of the NN. If the amount of training data still is not sufficient after the dimension of the input vector is reduced, regularization techniques, such as early stopping, jittering, and especially the embedment of estimated results by a theoretical model into the training set, should be used to improve the generalization ability of the neural network. The generalization of neural networks presents a wide spectrum of problems, and the proposed approaches are not confined strictly to modelling rice tillering dynamics but can be applied to other agricultural and ecological systems.     

应用于水文预报的优化BP神经网络研究

利用广东省滨江流域的水文观测资料，建立了以前期降水量为预报因子、以水位为输出的BP人工神经网络水文预报模型。首先采用了合理的方法进行样本组织，进而利用最优子集回归技术进行输入因子的确定，然后进行了不同隐层节点数、不同转移函数、不同训练算法的组合试验，确定了应用于水文预报中的优化BP神经网络：网络结构为8-9-1；转移函数的组合方式为tansig-线性函数；训练算法为采用evenberg-Marquardt(Lm)算法。为便于精度分析，还采用了最优子集回归模型作了研究。结果表明，优化BP网络模型无论在拟合精度还是在预测精度上都高于最优子集模型。总的来说BP网络是一种精度较高的水文预测模型。     

Applications of artificial neural networks for patterning and predicting aquatic insect species richness in running waters

Two artificial neural networks (ANNs), unsupervised and supervised learning algorithms, were applied to suggest practical approaches for the analysis of ecological data. Four major aquatic insect orders (Ephemeroptera, Plecoptera, Trichoptera, and Coleoptera, i.e. EPTC), and four environmental variables (elevation, stream order, distance from the source, and water temperature) were used to implement the models. The data were collected and measured at 155 sampling sites on streams of the Adour–Garonne drainage basin (South-western France). The modelling procedure was carried out following two steps. First, a self-organizing map (SOM), an unsupervised ANN, was applied to classify sampling sites using EPTC richness. Second, a backpropagation algorithm (BP), a supervised ANN, was applied to predict EPTC richness using a set of four environmental variables. The trained SOM classified sampling sites according to a gradient of EPTC richness, and the groups obtained corresponded to geographic regions of the drainage basin and characteristics of their environmental variables. The SOM showed its convenience to analyze relationships among sampling sites, biological attributes, and environmental variables. After accounting for the relationships in data sets, the BP used to predict the EPTC richness with a set of four environmental variables showed a high accuracy (r=0.91 and r=0.61 for training and test data sets respectively). The prediction of EPTC richness is thus a valuable tool to assess disturbances in given areas: by knowing what the EPTC richness should be, we can determine the degree to which disturbances have altered it. The results suggested that methodologies successively using two different neural networks are helpful to understand ecological data through ordination first, and then to predict target variables.     

Estimation of total sediment load concentration obtained by experimental study using artificial neural networks

Estimation of sediment concentration in rivers is very important for water resources projects planning and managements. The sediment concentration is generally determined from the direct measurement of sediment concentration of river or from sediment transport equations. Direct measurement is very expensive and cannot be conducted for all river gauge stations. However, sediment transport equations do not agree with each other and require many detailed data on the flow and sediment characteristics. The main purpose of the study is to establish an effective model which includes nonlinear relations between dependent (total sediment load concentration) and independent (bed slope, flow discharge, and sediment particle size) variables. In the present study, by performing 60 experiments for various independent data, dependent variables were obtained, because of the complexity of the phenomena, as a soft computing method artificial neural networks (ANNs) which is the powerful tool for input–output mapping is used. However, ANN model was compared with total sediment transport equations. The results show that ANN model is found to be significantly superior to total sediment transport equations.     

Artificial neural network modeling of caste odor discrimination based on cuticular hydrocarbons in termites

Summary. Individuals in an insect colony need to identify one another according to caste. Nothing is known about the sensory process allowing nestmates to discriminate minute variations in the cuticular hydrocarbon mixture. The purpose of this study was to attempt to model caste odors discrimination in four species of Reticulitermes termites for the first time by a non-linear mathematical approach using an "artificial neural network" (ANN). Several rounds of testing were carried out using 1 – the whole hydrocarbon mixtures 2 – mixtures containing the hydrocarbons selected by principal component analysis (PCA) as the most implicated in caste discrimination. Discrimination between worker and soldier castes was tested in all four species. For two species we tested discrimination of four castes (workers, soldiers, nymphs, neotenics). To test cuticular pattern similarity in two sibling species (R. santonensis and R. flavipes), we performed two experiments using one species for training and the other for query. Using whole hydrocarbons mixtures, worker/soldier discrimination was always successful in all species. Network performance decreased with the number of hydrocarbons used as inputs. Four-caste discrimination was less successful. In the experiment with the sibling species, the ANN was able to distinguish soldiers but not workers. The results of this study suggest that non-linear mathematical analysis is a good tool for classification of castes based on cuticular hydrocarbon mixture. In addition this study confirms that hydrocarbon mixtures observed are real chemical entities and constitute a true chemical signature or odor. Whole mixtures are not always necessary for discrimination. Received 23 July 1998; accepted 9 October 1998.     

Water and carbon fluxes above European coniferous forests modelled with artificial neural networks

Artificial neural networks are used to select a minimal set of input variables to model water vapour and carbon exchange of coniferous forest ecosystems, independently of tree species and without detailed physiological information. Neural networks are used because of their power to fit highly non-linear relations between input and output-variables. Radiation, temperature, vapour pressure deficit and time of the day showed to be the dynamic input variables that determine ecosystem water fluxes. The same variables, together with projected leaf area index are needed for modelling CO₂-fluxes. The results for the individual sites show that the neural networks found mean water and carbon flux responses to the driving variables valid for all sites. The sensitivity analysis of the derived neural networks shows that the LAI-effect of the CO₂-flux model is overfitted because of the low variability of LAI. However, the predictions of CO₂-fluxes of sites not included in the calibration set indicate that the LAI-response of the network is reliable and that results can be used as a first estimate of the net ecosystem carbon exchange of the forest sites. Independent predictions of forest ecosystem vapour fluxes were equally satisfying as empirical models specifically calibrated for the individual sites. The results indicate that both short term water and carbon fluxes of European coniferous forests can be modelled without using detailed physiological and site specific information.