study was conducted in Bunda district, which is a Tanzanian community with high annual incidents of human-elephant interactions, to determine a geographical configuration of hidden impacts. These are indirect impacts and largely unreported adverse effects resulting from human and elephant interactions. These are the effects which usually go unnoticed and unreported due to the lack of visible damage. Spatial analyses of patterns of human-elephant interactions have focused on environmental to socio-economic perspectives rather than spatial aspects of hidden patterns. The study analyzed the distribution, proximity to protected areas, kernel density and hotspots analysis of hidden impacts. The study identified 327 hidden impacts, categorized into the abandonment of farms, marriage problems, delayed school attendances and restriction on movement. It ascertained the highest number of incidents (18.35%) from Kihumbu village and the lowest from Nyangere village (0.01%). Abandonment of farms constituted the largest number (77.4%) while marriage problems formed the lowest number (0.6%) of hidden impacts. The most hidden impacts occurred between 0 and 2000 meters from the boundaries of protected areas. There was a higher concentration of hidden impacts in villages bordering Grumeti Game Reserve than Serengeti National Park. The significant statistical level of adverse hidden impacts occurred in Kihumbu village. Imprecisely execution of tourist hunting operations could presumably be the causing factor for the high concentration of hidden effects nearby Grumeti Game Reserve. However, we recommend a comprehensive study for an intensive understanding of the spatial characteristics of other types of hidden impacts adjacent to protected areas.
{"title":"Analysing the Geospatial Patterns of Hidden Impacts from Human-Elephant Interactions in the Bunda District, Tanzania","authors":"A. Mamboleo, C. Doscher, A. Paterson","doi":"10.3808/jeil.201900016","DOIUrl":"https://doi.org/10.3808/jeil.201900016","url":null,"abstract":"study was conducted in Bunda district, which is a Tanzanian community with high annual incidents of human-elephant interactions, to determine a geographical configuration of hidden impacts. These are indirect impacts and largely unreported adverse effects resulting from human and elephant interactions. These are the effects which usually go unnoticed and unreported due to the lack of visible damage. Spatial analyses of patterns of human-elephant interactions have focused on environmental to socio-economic perspectives rather than spatial aspects of hidden patterns. The study analyzed the distribution, proximity to protected areas, kernel density and hotspots analysis of hidden impacts. The study identified 327 hidden impacts, categorized into the abandonment of farms, marriage problems, delayed school attendances and restriction on movement. It ascertained the highest number of incidents (18.35%) from Kihumbu village and the lowest from Nyangere village (0.01%). Abandonment of farms constituted the largest number (77.4%) while marriage problems formed the lowest number (0.6%) of hidden impacts. The most hidden impacts occurred between 0 and 2000 meters from the boundaries of protected areas. There was a higher concentration of hidden impacts in villages bordering Grumeti Game Reserve than Serengeti National Park. The significant statistical level of adverse hidden impacts occurred in Kihumbu village. Imprecisely execution of tourist hunting operations could presumably be the causing factor for the high concentration of hidden effects nearby Grumeti Game Reserve. However, we recommend a comprehensive study for an intensive understanding of the spatial characteristics of other types of hidden impacts adjacent to protected areas.","PeriodicalId":143718,"journal":{"name":"Journal of Environmental Informatics Letters","volume":"133 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116268608","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Temperature is one of the most important parameters in climate modeling, as it has significant impacts on various geophysical processes such as evaporation and precipitation. Applying multiple climate models for prediction generally outperforms the use of individual climate models, and neural networks perform well at capturing nonlinear relationships, which can provide more reliable temperature projections. In this study, three neural network algorithms, including Multi-layer Perceptron (MLP), Time-lagged Feed-forward Neural Networks (TLFN) and Nonlinear Auto-Regressive Networks with exogenous inputs (NARX), were used to develop data-driven models for predicting daily mean near-surface temperature based on North American Coordinated Regional Downscaling Experiment (NA-CORDEX) output. A case study of Big Trout Lake in Ontario, Canada was carried out to demonstrate the applications and to evaluate the performance of the proposed neural network based models. The results showed that MLP, TLFN, and NARX performed well in generating accurate daily near-surface temperature predictions with the coefficient of determination (R2) values above 0.84. The three neural network based models had similar performance with no significant difference in terms of root mean square error and R2. Neural network based climate prediction models outperformed each of the individual regional climate models and generated smoother predictions with less fluctuation. This study provides a technical basis for generating reliable predictions of daily temperature using neural networks based model.
{"title":"Prediction of Long-Term Near-Surface Temperature Based on NA-CORDEX Output","authors":"X. Li, Z. Li, Q. Zhang, P. Zhou, W. Huang","doi":"10.3808/jeil.201900012","DOIUrl":"https://doi.org/10.3808/jeil.201900012","url":null,"abstract":"Temperature is one of the most important parameters in climate modeling, as it has significant impacts on various geophysical processes such as evaporation and precipitation. Applying multiple climate models for prediction generally outperforms the use of individual climate models, and neural networks perform well at capturing nonlinear relationships, which can provide more reliable temperature projections. In this study, three neural network algorithms, including Multi-layer Perceptron (MLP), Time-lagged Feed-forward Neural Networks (TLFN) and Nonlinear Auto-Regressive Networks with exogenous inputs (NARX), were used to develop data-driven models for predicting daily mean near-surface temperature based on North American Coordinated Regional Downscaling Experiment (NA-CORDEX) output. A case study of Big Trout Lake in Ontario, Canada was carried out to demonstrate the applications and to evaluate the performance of the proposed neural network based models. The results showed that MLP, TLFN, and NARX performed well in generating accurate daily near-surface temperature predictions with the coefficient of determination (R2) values above 0.84. The three neural network based models had similar performance with no significant difference in terms of root mean square error and R2. Neural network based climate prediction models outperformed each of the individual regional climate models and generated smoother predictions with less fluctuation. This study provides a technical basis for generating reliable predictions of daily temperature using neural networks based model.","PeriodicalId":143718,"journal":{"name":"Journal of Environmental Informatics Letters","volume":"180 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-08-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124503222","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Eutrophication is one of the main causes of the degradation of lake ecosystems. In this paper, multiple linear regression (MLR) and neural network (NN) methods were developed as empirical models to predict chlorophyll-a (Chl-a) concentrations in Lake Champlain. The models were developed using a large dataset collected from Lake Champlain over a 24-year period from 1992 to 2016. The dataset consisted of monitoring depth (Depth), total phosphorus (TP), total nitrogen (TN), alkalinity (RegAlk), temperature (TempC), chloride (Cl) and secchi depth (Secchi). Statistical analyses showed that TP, Secchi, TN and Depth demonstrated strong relationships with Chl-a concentrations. The simulation results revealed that both the MLR and NN models performed well in predicting Chl-a concentrations, especially for low to moderate concentrations of Chl-a ( 7.5 μg/L). These models can be useful for improving lake management and providing early warnings regarding the problem of eutrophication.
{"title":"Examination of Multiple Linear Regression (MLR) and Neural Network (NN) Models to Predict Eutrophication Levels in Lake Champlain","authors":"L. E. Farra, K. Wang, Z. Chen, Y. Zhu","doi":"10.3808/JEIL.201900007","DOIUrl":"https://doi.org/10.3808/JEIL.201900007","url":null,"abstract":"Eutrophication is one of the main causes of the degradation of lake ecosystems. In this paper, multiple linear regression (MLR) and neural network (NN) methods were developed as empirical models to predict chlorophyll-a (Chl-a) concentrations in Lake Champlain. The models were developed using a large dataset collected from Lake Champlain over a 24-year period from 1992 to 2016. The dataset consisted of monitoring depth (Depth), total phosphorus (TP), total nitrogen (TN), alkalinity (RegAlk), temperature (TempC), chloride (Cl) and secchi depth (Secchi). Statistical analyses showed that TP, Secchi, TN and Depth demonstrated strong relationships with Chl-a concentrations. The simulation results revealed that both the MLR and NN models performed well in predicting Chl-a concentrations, especially for low to moderate concentrations of Chl-a ( 7.5 μg/L). These models can be useful for improving lake management and providing early warnings regarding the problem of eutrophication.","PeriodicalId":143718,"journal":{"name":"Journal of Environmental Informatics Letters","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133014108","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Influent flow rate is a crucial parameter closely related to the plant-wide control of wastewater treatment plants (WWTPs). In this study, a random forest (RF) model and a multi-layer perceptron (MLP) model are developed for hourly influent flow rate prediction at a confidential WWTP in Canada. Both models perform well on predicting influent flow rate one-step ahead. The coefficient of determination (R2) values of MLP and RF for the testing data set are 0.927 and 0.925, respectively. Furthermore, the multi-step ahead prediction accuracy of the proposed models is discussed. To improve the multi-step ahead prediction accuracy of the RF model, time-tag information is transformed to numerical values and then fed into the RF model as input. The R2 values of the RF model for the testing data set with and without time-tag information are 0.334 and 0.811, respectively. The results show that the RF model’s performance for multi- step ahead prediction is heavily affected by the time-tag information. Including time-tag information as input could dramatically improve the multi-step ahead prediction accuracy. In this study, the RF model shows more robust performance than the MLP model on solving short-term wastewater influent prediction problems. Influent flow rate is a crucial parameter closely related to the plant-wide control of wastewater treatment plants (WWTPs). In this study, a random forest (RF) model and a multi-layer perceptron (MLP) model are developed for hourly influent flow rate prediction at a confidential WWTP in Canada. Both models perform well on predicting influent flow rate one-step ahead. The coefficient of determination (R2) values of MLP and RF for the testing data set are 0.927 and 0.925, respectively. Furthermore, the multi-step ahead prediction accuracy of the proposed models is discussed. To improve the multi-step ahead prediction accuracy of the RF model, time-tag information is transformed to numerical values and then fed into the RF model as input. The R2 values of the RF model for the testing data set with and without time-tag information are 0.334 and 0.811, respectively. The results show that the RF model’s performance for multi-step ahead prediction is heavily affected by the time-tag information. Including time-tag information as input could dramatically improve the multi-step ahead prediction accuracy. In this study, the RF model shows more robust performance than the MLP model on solving short-term wastewater influent prediction problems.
{"title":"Short-Term Wastewater Influent Prediction Based on Random Forests and Multi-Layer Perceptron","authors":"P. Zhou, Z. Li, S. Snowling, R. Goel, Q. Zhang","doi":"10.3808/JEIL.201900010","DOIUrl":"https://doi.org/10.3808/JEIL.201900010","url":null,"abstract":"Influent flow rate is a crucial parameter closely related to the plant-wide control of wastewater treatment plants (WWTPs). In this study, a random forest (RF) model and a multi-layer perceptron (MLP) model are developed for hourly influent flow rate prediction at a confidential WWTP in Canada. Both models perform well on predicting influent flow rate one-step ahead. The coefficient of determination (R2) values of MLP and RF for the testing data set are 0.927 and 0.925, respectively. Furthermore, the multi-step ahead prediction accuracy of the proposed models is discussed. To improve the multi-step ahead prediction accuracy of the RF model, time-tag information is transformed to numerical values and then fed into the RF model as input. The R2 values of the RF model for the testing data set with and without time-tag information are 0.334 and 0.811, respectively. The results show that the RF model’s performance for multi- step ahead prediction is heavily affected by the time-tag information. Including time-tag information as input could dramatically improve the multi-step ahead prediction accuracy. In this study, the RF model shows more robust performance than the MLP model on solving short-term wastewater influent prediction problems. Influent flow rate is a crucial parameter closely related to the plant-wide control of wastewater treatment plants (WWTPs). In this study, a random forest (RF) model and a multi-layer perceptron (MLP) model are developed for hourly influent flow rate prediction at a confidential WWTP in Canada. Both models perform well on predicting influent flow rate one-step ahead. The coefficient of determination (R2) values of MLP and RF for the testing data set are 0.927 and 0.925, respectively. Furthermore, the multi-step ahead prediction accuracy of the proposed models is discussed. To improve the multi-step ahead prediction accuracy of the RF model, time-tag information is transformed to numerical values and then fed into the RF model as input. The R2 values of the RF model for the testing data set with and without time-tag information are 0.334 and 0.811, respectively. The results show that the RF model’s performance for multi-step ahead prediction is heavily affected by the time-tag information. Including time-tag information as input could dramatically improve the multi-step ahead prediction accuracy. In this study, the RF model shows more robust performance than the MLP model on solving short-term wastewater influent prediction problems.","PeriodicalId":143718,"journal":{"name":"Journal of Environmental Informatics Letters","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114951111","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Y. Odeyemi, M. Pollind, Ryan Peeler, K. Nozawa, D. Vesely, Anthony M. Page, C. Rakovski, H. El-Askary
This research builds a multinomial regression framework to conduct a meta-analysis of trends in climate research and funding as related to the state of affairs in the last twenty-five years in this area of research. We used a climate research query-based strategy searching the Web of Science, National Science Foundation, Australia Department of Environment and Energy, African Development Bank’s African Climate Change Fund, the Asian Development Bank Climate Change Fund and Australia’s Department of Environment and Energy databases to perform quantitative and qualitative trend analysis. Data were harvested using a web scraper and filtered for the 1993 ~ 2017 window. Comparative analysis was carried out to evaluate the climate research output per continent. Also, we evaluated the role funding plays in the climate research outcomes. Different text processing and mining techniques were used to extract information and data needed for trend analysis and statistical modeling. The text processing revealed trends such as major key- words, key opinion leaders, and individual country’s contribution, monthly and yearly spread of published articles in the climate research domain. From these trends, we engineered some of the variables to build a multinomial regression model to further understand future trends in the climate research space. It is probabilistic in nature with the assumption of no inter correlation between variables, hence outputs are more significant. We found that funding for climate research has been on a steady increase in the last twenty-five years, with the US and European investing hundreds of millions of dollars in alternative and renewable energy. Lastly, the multinomial logistic regression assesses the impact of number of investigators, abstract word count and institution types on the class of grant awarded by NSF. This research builds a multinomial regression framework to conduct a meta-analysis of trends in climate research and funding as related to the state of affairs in the last twenty-five years in this area of research. We used a climate research query-based strategy searching the Web of Science, National Science Foundation, Australia Department of Environment and Energy, African Development Bank’s African Climate Change Fund, the Asian Development Bank Climate Change Fund and Australia’s Department of Environment and Energy databases to perform quantitative and qualitative trend analysis. Data were harvested using a web scraper and filtered for the 1993 ~ 2017 window. Comparative analysis was carried out to evaluate the climate research output per continent. Also, we evaluated the role funding plays in the climate research outcomes. Different text processing and mining techniques were used to extract information and data needed for trend analysis and statistical modeling. The text processing revealed trends such as major keywords, key opinion leaders, and individual country’s contribution, monthly and yearly spread of published
本研究建立了一个多项回归框架，对过去25年气候研究和资金状况的趋势进行了荟萃分析。我们使用基于气候研究查询的策略，检索Web of Science、美国国家科学基金会、澳大利亚环境与能源部、非洲开发银行的非洲气候变化基金、亚洲开发银行气候变化基金和澳大利亚环境与能源部的数据库，进行定量和定性趋势分析。使用web scraper收集数据，并对1993年至2017年的窗口进行过滤。对各大洲的气候研究成果进行了对比分析。此外，我们还评估了资金在气候研究成果中所起的作用。使用不同的文本处理和挖掘技术来提取趋势分析和统计建模所需的信息和数据。文本处理揭示了气候研究领域的主要关键词、主要意见领袖、单个国家的贡献、每月和每年发表的文章的传播等趋势。根据这些趋势，我们设计了一些变量来构建多项式回归模型，以进一步了解气候研究领域的未来趋势。它本质上是概率性的，假设变量之间没有相互关联，因此输出更重要。我们发现，在过去的25年里，气候研究的资金一直在稳步增长，美国和欧洲在替代能源和可再生能源上投入了数亿美元。最后，采用多项逻辑回归方法评估了研究人员数量、摘要字数和机构类型对NSF资助类别的影响。本研究建立了一个多项回归框架，对过去25年气候研究和资金状况的趋势进行了荟萃分析。我们使用基于气候研究查询的策略，检索Web of Science、美国国家科学基金会、澳大利亚环境与能源部、非洲开发银行的非洲气候变化基金、亚洲开发银行气候变化基金和澳大利亚环境与能源部的数据库，进行定量和定性趋势分析。使用web scraper收集数据，并对1993年至2017年的窗口进行过滤。对各大洲的气候研究成果进行了对比分析。此外，我们还评估了资金在气候研究成果中所起的作用。使用不同的文本处理和挖掘技术来提取趋势分析和统计建模所需的信息和数据。文本处理揭示了气候研究领域的主要关键词、主要意见领袖、单个国家的贡献、每月和每年发表的文章的传播等趋势。根据这些趋势，我们设计了一些变量来构建多项式回归模型，以进一步了解气候研究领域的未来趋势。它本质上是概率性的，假设变量之间没有相互关联，因此输出更重要。我们发现，在过去的25年里，气候研究的资金一直在稳步增长，美国和欧洲在替代能源和可再生能源上投入了数亿美元。最后，采用多项逻辑回归方法评估了研究人员数量、摘要字数和机构类型对NSF资助类别的影响。
{"title":"Review of Climate Research and Funding 1993 ~ 2017: A Multinomial Logistic Regression Approach","authors":"Y. Odeyemi, M. Pollind, Ryan Peeler, K. Nozawa, D. Vesely, Anthony M. Page, C. Rakovski, H. El-Askary","doi":"10.3808/JEIL.201900011","DOIUrl":"https://doi.org/10.3808/JEIL.201900011","url":null,"abstract":"This research builds a multinomial regression framework to conduct a meta-analysis of trends in climate research and funding as related to the state of affairs in the last twenty-five years in this area of research. We used a climate research query-based strategy searching the Web of Science, National Science Foundation, Australia Department of Environment and Energy, African Development Bank’s African Climate Change Fund, the Asian Development Bank Climate Change Fund and Australia’s Department of Environment and Energy databases to perform quantitative and qualitative trend analysis. Data were harvested using a web scraper and filtered for the 1993 ~ 2017 window. Comparative analysis was carried out to evaluate the climate research output per continent. Also, we evaluated the role funding plays in the climate research outcomes. Different text processing and mining techniques were used to extract information and data needed for trend analysis and statistical modeling. The text processing revealed trends such as major key- words, key opinion leaders, and individual country’s contribution, monthly and yearly spread of published articles in the climate research domain. From these trends, we engineered some of the variables to build a multinomial regression model to further understand future trends in the climate research space. It is probabilistic in nature with the assumption of no inter correlation between variables, hence outputs are more significant. We found that funding for climate research has been on a steady increase in the last twenty-five years, with the US and European investing hundreds of millions of dollars in alternative and renewable energy. Lastly, the multinomial logistic regression assesses the impact of number of investigators, abstract word count and institution types on the class of grant awarded by NSF. This research builds a multinomial regression framework to conduct a meta-analysis of trends in climate research and funding as related to the state of affairs in the last twenty-five years in this area of research. We used a climate research query-based strategy searching the Web of Science, National Science Foundation, Australia Department of Environment and Energy, African Development Bank’s African Climate Change Fund, the Asian Development Bank Climate Change Fund and Australia’s Department of Environment and Energy databases to perform quantitative and qualitative trend analysis. Data were harvested using a web scraper and filtered for the 1993 ~ 2017 window. Comparative analysis was carried out to evaluate the climate research output per continent. Also, we evaluated the role funding plays in the climate research outcomes. Different text processing and mining techniques were used to extract information and data needed for trend analysis and statistical modeling. The text processing revealed trends such as major keywords, key opinion leaders, and individual country’s contribution, monthly and yearly spread of published ","PeriodicalId":143718,"journal":{"name":"Journal of Environmental Informatics Letters","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131218420","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Frequently occurring ice jams often cause concern in northern regions. Breakup timing is directly related to emergency responses preparation and thus its early accurate forecasting is beneficial to ice-related flooding management. The stepwise cluster analysis (SCA) is a non-parameter regression method, which generates a classification tree in the sense of probability through cutting or merging operations according to certain statistic criteria. To enhance SCA’s predictive performance, a SCA ensemble (SCAE) method is developed and applied to forecasting of annual river ice breakup dates (BDs). In detail, the SCA is employed as a base model at the lower level while the simple average method is selected as combining models at the upper level. The SCA base models are selected according to different performance selection criteria and searched for further combination. A site on a representative river prone to river ice flooding in Alberta, Canada is selected to demonstrate the effectiveness of the proposed SCAE. The results mainly show that: the SCA base models with multiple combinations of inputs and internal parameters are able to predict the BDs with good performances (the highest average of correlation coefficients for training can be 0.958); the optimal SCA base model has three inputs, which indicates that the temperatures before breakup and just after freeze-up as well as the maximum of water flow in March are relatively important indicators of BD. The optimal SCAE, including base models from different performance selection criteria, has the lowest average of root mean squared error, which improves upon the optimal SCA base model by 25.3%. It indicates the different model selection criteria do improve the diversity and thus further help to improve the performance of ensemble models. This first application of the SCAE to river ice forecasting highlights the possibility of using the ensemble learning paradigm to enhance the SCA. The potential applications of the SCAE to other forecasting problems are expected.
{"title":"Ensemble Learning Enhanced Stepwise Cluster Analysis for River Ice Breakup Date Forecasting","authors":"W. Sun, Q. Shi, Y. Huang, Y. Lv","doi":"10.3808/JEIL.201900005","DOIUrl":"https://doi.org/10.3808/JEIL.201900005","url":null,"abstract":"Frequently occurring ice jams often cause concern in northern regions. Breakup timing is directly related to emergency responses preparation and thus its early accurate forecasting is beneficial to ice-related flooding management. The stepwise cluster analysis (SCA) is a non-parameter regression method, which generates a classification tree in the sense of probability through cutting or merging operations according to certain statistic criteria. To enhance SCA’s predictive performance, a SCA ensemble (SCAE) method is developed and applied to forecasting of annual river ice breakup dates (BDs). In detail, the SCA is employed as a base model at the lower level while the simple average method is selected as combining models at the upper level. The SCA base models are selected according to different performance selection criteria and searched for further combination. A site on a representative river prone to river ice flooding in Alberta, Canada is selected to demonstrate the effectiveness of the proposed SCAE. The results mainly show that: the SCA base models with multiple combinations of inputs and internal parameters are able to predict the BDs with good performances (the highest average of correlation coefficients for training can be 0.958); the optimal SCA base model has three inputs, which indicates that the temperatures before breakup and just after freeze-up as well as the maximum of water flow in March are relatively important indicators of BD. The optimal SCAE, including base models from different performance selection criteria, has the lowest average of root mean squared error, which improves upon the optimal SCA base model by 25.3%. It indicates the different model selection criteria do improve the diversity and thus further help to improve the performance of ensemble models. This first application of the SCAE to river ice forecasting highlights the possibility of using the ensemble learning paradigm to enhance the SCA. The potential applications of the SCAE to other forecasting problems are expected.","PeriodicalId":143718,"journal":{"name":"Journal of Environmental Informatics Letters","volume":"59 3-4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132062615","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Thousands of new, emerging chemicals are produced each year, making thorough investigations infeasible regarding their potential detrimental dimensions. As an important step for estimating whether a chemical will result in an exposure pathway and therefore create the potential for a detrimental impact, a coefficient-based strategy consisting of eight key coefficients, is proposed. The strategy is based upon key factors which are used to assess the potential for a chemical to attenuate or change its phase or medium, as part of its fate and transport pathway. The eight key coefficients are described, knowledge of which will assist in determining whether a chemical will result in a fate and exposure pathway change and/or attenuate, as a means of developing a strategy to assess the risks of emerging contaminants. The need for attention to this next water revolution to develop a strategy to assess some of the risks of emerging contaminants is already upon us.
{"title":"Removal of Emerging Contaminants: The Next Water Revolution","authors":"E. McBean","doi":"10.3808/JEIL.201900001","DOIUrl":"https://doi.org/10.3808/JEIL.201900001","url":null,"abstract":"Thousands of new, emerging chemicals are produced each year, making thorough investigations infeasible regarding their potential detrimental dimensions. As an important step for estimating whether a chemical will result in an exposure pathway and therefore create the potential for a detrimental impact, a coefficient-based strategy consisting of eight key coefficients, is proposed. The strategy is based upon key factors which are used to assess the potential for a chemical to attenuate or change its phase or medium, as part of its fate and transport pathway. The eight key coefficients are described, knowledge of which will assist in determining whether a chemical will result in a fate and exposure pathway change and/or attenuate, as a means of developing a strategy to assess the risks of emerging contaminants. The need for attention to this next water revolution to develop a strategy to assess some of the risks of emerging contaminants is already upon us.","PeriodicalId":143718,"journal":{"name":"Journal of Environmental Informatics Letters","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126857529","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Environmental decision-making commonly involves multifaceted problems that demonstrate considerable uncertainty. Monte Carlo simulation approaches have been employed in a variety of environmental planning venues to address these uncertain aspects. Simulation-based outputs are frequently presented in the form of probability distributions. Recently an approach referred to as simulation decomposition (SD) has been introduced that extends the analysis of Monte Carlo results by enhancing the explanatory power of the cause-effect relationships between the multi-variable combinations of inputs and the simulated outputs. SD constructs sub-distributions of the simulation output by pre-classifying some of the uncertain input variables into states, clustering the various combinations of these different states into scenarios, and then collecting simulated outputs attributable to each multi-variable input scenario. Since the contribution of subdivided scenarios to the overall output is easily portrayed visually, SD can highlight and disclose previously unidentified connections between the multi-variable combinations of inputs on the outputs. An SD approach is generalizable to any Monte Carlo model with negligible additional computational overhead and, hence, can be readily used for environmental analyses that employ simulation models. This study illustrates the efficacy of SD in environmental analysis using a carbon capture and storage project from China.
{"title":"Multi-Variable Simulation Decomposition in Environmental Planning: An Application to Carbon Capture and Storage","authors":"M. Kozlova, J. Yeomans","doi":"10.3808/JEIL.201900003","DOIUrl":"https://doi.org/10.3808/JEIL.201900003","url":null,"abstract":"Environmental decision-making commonly involves multifaceted problems that demonstrate considerable uncertainty. Monte Carlo simulation approaches have been employed in a variety of environmental planning venues to address these uncertain aspects. Simulation-based outputs are frequently presented in the form of probability distributions. Recently an approach referred to as simulation decomposition (SD) has been introduced that extends the analysis of Monte Carlo results by enhancing the explanatory power of the cause-effect relationships between the multi-variable combinations of inputs and the simulated outputs. SD constructs sub-distributions of the simulation output by pre-classifying some of the uncertain input variables into states, clustering the various combinations of these different states into scenarios, and then collecting simulated outputs attributable to each multi-variable input scenario. Since the contribution of subdivided scenarios to the overall output is easily portrayed visually, SD can highlight and disclose previously unidentified connections between the multi-variable combinations of inputs on the outputs. An SD approach is generalizable to any Monte Carlo model with negligible additional computational overhead and, hence, can be readily used for environmental analyses that employ simulation models. This study illustrates the efficacy of SD in environmental analysis using a carbon capture and storage project from China.","PeriodicalId":143718,"journal":{"name":"Journal of Environmental Informatics Letters","volume":"123 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132252091","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Dry climate, clayey soil, and flat topography govern water balance in the southern part of the Canadian Praries. The main purpose of this work was to assess flood-drought hazard using Regina as a typical urban centre in the region. Results indicate that extreme weather patterns are frequent and meteorological parameters have changed from 1970 to 2015: precipitation (+50 mm), air temperature (+0.9oC), relative humidity (+6%), wind speed (-1.35 km/hr), and solar radiation (+0.9 MJ/m2). In the dry climate (Dfb), 77% of the total annual precipitation (386 mm/year) occurs from April to September. The runoff coefficient of 0.6 relates to 63% impervious areas (commercial, industrial and residential) and 35% near-impervious areas (open spaces with low hydraulic conductivity). The flat topography (570 m through 600 m asl over 124 km2) along with a low channel slope of up to 0.4% results in water ponding during short-term and high-intensity rainfalls. Water is managed through the Wascana Creek that holds 98% of the total water volume (84 x 106 m3) in the city. From April to September, volume fluctuations depend on antecedent water levels and meteorological conditions. The city has recently received several events of flash floods (2010 and 2014) and long-term droughts (1984 and 2017). The negative average change in storage indicates drought-like conditions during spring-summer.
{"title":"Flood-Drought Hazard Assessment for a Flat Clayey Deposit in the Canadian Prairies","authors":"A. Akhter, S. Azam","doi":"10.3808/JEIL.201900002","DOIUrl":"https://doi.org/10.3808/JEIL.201900002","url":null,"abstract":"Dry climate, clayey soil, and flat topography govern water balance in the southern part of the Canadian Praries. The main purpose of this work was to assess flood-drought hazard using Regina as a typical urban centre in the region. Results indicate that extreme weather patterns are frequent and meteorological parameters have changed from 1970 to 2015: precipitation (+50 mm), air temperature (+0.9oC), relative humidity (+6%), wind speed (-1.35 km/hr), and solar radiation (+0.9 MJ/m2). In the dry climate (Dfb), 77% of the total annual precipitation (386 mm/year) occurs from April to September. The runoff coefficient of 0.6 relates to 63% impervious areas (commercial, industrial and residential) and 35% near-impervious areas (open spaces with low hydraulic conductivity). The flat topography (570 m through 600 m asl over 124 km2) along with a low channel slope of up to 0.4% results in water ponding during short-term and high-intensity rainfalls. Water is managed through the Wascana Creek that holds 98% of the total water volume (84 x 106 m3) in the city. From April to September, volume fluctuations depend on antecedent water levels and meteorological conditions. The city has recently received several events of flash floods (2010 and 2014) and long-term droughts (1984 and 2017). The negative average change in storage indicates drought-like conditions during spring-summer.","PeriodicalId":143718,"journal":{"name":"Journal of Environmental Informatics Letters","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-04-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121500694","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The stability of earthen structures is governed by field uncertainties arising from material properties, environment loading, and slope geometry. This research devised a systematic approach to capture the effects of field uncertainties on the stability of natural and manmade slopes. New design charts were developed by incrementally changing slope geometry and randomly generating shear strength parameters. Subset simulation was used to determine the safe range of soil properties for various slopes. The charts were applied to three published case studies with distinct triggering mechanisms resulting from complex field settings. All of the investigated slopes were found to be stable (factor of safety (FOS) > 1.0) under the reported geometry and shear strength parameters while assuming no water table. The effects of soil properties’ variation and environmental conditions on fluctuating water table were captured through history matching. Results indicated three distinct failure mechanisms: foundation settlement of a glacial moraine till (Vernon, British Columbia) due to an increased pore water pressure during construction of the compacted fill (FOS = 1.45 without berms and 2.24 with berms); instability in the natural cut (FOS = 0.98) comprising layered glacio-lacustrine clays (Labret, Saskatchewan) due to saturation of the entire slope resulting from a long duration rainfall; and collapse of a compacted fill (FOS = 0.98) on a glacial moraine till (Hamelin Creek, Alberta) due to soil saturation arising from thawing of a frozen layer in the slope. This validation illustrates that the new approach fully captures environmental loading (resulting in water table variation in the slope) and partly captures construction practice and site geology via soil properties.
{"title":"Parametric Design and Field Behavior of Earthen Structures","authors":"R. Paranthaman, S. Azam","doi":"10.3808/jeil.202300102","DOIUrl":"https://doi.org/10.3808/jeil.202300102","url":null,"abstract":"The stability of earthen structures is governed by field uncertainties arising from material properties, environment loading, and slope geometry. This research devised a systematic approach to capture the effects of field uncertainties on the stability of natural and manmade slopes. New design charts were developed by incrementally changing slope geometry and randomly generating shear strength parameters. Subset simulation was used to determine the safe range of soil properties for various slopes. The charts were applied to three published case studies with distinct triggering mechanisms resulting from complex field settings. All of the investigated slopes were found to be stable (factor of safety (FOS) > 1.0) under the reported geometry and shear strength parameters while assuming no water table. The effects of soil properties’ variation and environmental conditions on fluctuating water table were captured through history matching. Results indicated three distinct failure mechanisms: foundation settlement of a glacial moraine till (Vernon, British Columbia) due to an increased pore water pressure during construction of the compacted fill (FOS = 1.45 without berms and 2.24 with berms); instability in the natural cut (FOS = 0.98) comprising layered glacio-lacustrine clays (Labret, Saskatchewan) due to saturation of the entire slope resulting from a long duration rainfall; and collapse of a compacted fill (FOS = 0.98) on a glacial moraine till (Hamelin Creek, Alberta) due to soil saturation arising from thawing of a frozen layer in the slope. This validation illustrates that the new approach fully captures environmental loading (resulting in water table variation in the slope) and partly captures construction practice and site geology via soil properties.","PeriodicalId":143718,"journal":{"name":"Journal of Environmental Informatics Letters","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126767425","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：