Asymmetric circular-linear multivariate regression models with applications to environmental data

We propose asymmetric angular-linear multivariate regression models, which were motivated by the need to predict some environmental characteristics based on some circular and linear predictors. A measure of fit is provided through the residual analysis. Some applications using data from solar energy radiation experiment and wind energy are given. Received: September 2003 / Revised: February 2005     

Optimal patch time allocation for time-limited foragers

The Charnov Marginal Value Theorem (MVT) predicts the optimal foraging duration of animals exploiting patches of resources. The predictions of this model have been verified for various animal species. However, the model is based on several assumptions that are likely too simplistic. One of these assumptions is that animals are living forever (i.e., infinite horizon). Using a simple dynamic programming model, we tested the importance of this assumption by analysing the optimal strategy for time-limited foragers. We found that, for time-limited foragers, optimal patch residence times should be greater than those predicted from the classic, static MVT, and the deviation should increase when foragers are approaching the end of their life. These predictions were verified for females of the parasitoid Anaphes victus (Hymenoptera: Mymaridae) exploiting egg patches of its host, the carrot weevil Listronotus oregonensis (Coleoptera: Curculionidae). As predicted by the model, females indeed remained for a longer time on host patches when they approached the end of their life. Experimental results were finally analysed with a Cox regression model to identify the patch-leaving decision rules females used to behave according to the model’s predictions.     

降水pH值的支持向量回归预测模型构建

将支持向量回归用于降水pH值预测模型的构建,结果表明,该模型具有较好的稳定性和较高的预测精度,降水的pH值主要受大气中碱性离子浓度的影响,起主导作用的是碱性离子的中和作用;其预测结果优于多元线性回归、主成分回归、偏最小二乘回归和投影寻踪回归等模型.     

Modeling Species' Distributions to Improve Conservation in Semiurban Landscapes: Koala Case Study

Abstract:  Models of species' distributions are commonly used to inform landscape and conservation planning. In urban and semiurban landscapes, the distributions of species are determined by a combination of natural habitat and anthropogenic impacts. Understanding the spatial influence of these two processes is crucial for making spatially explicit decisions about conservation actions. We present a logistic regression model for the distribution of koalas (  Phascolarctos cinereus ) in a semiurban landscape in eastern Australia that explicitly separates the effect of natural habitat quality and anthropogenic impacts on koala distributions. We achieved this by comparing the predicted distributions from the model with what the predicted distributions would have been if anthropogenic variables were at their mean values. Similar approaches have relied on making predictions assuming anthropogenic variables are zero, which will be unreliable if the training data set does not include anthropogenic variables close to zero. Our approach is novel because it can be applied to landscapes where anthropogenic variables are never close to zero. Our model showed that, averaged across the study area, natural habitat was the main determinant of koala presence. At a local scale, however, anthropogenic impacts could be more important, with consequent implications for conservation planning. We demonstrated that this modeling approach, combined with the visual presentation of predictions as a map, provides important information for making decisions on how different conservation actions should be spatially allocated. This method is particularly useful for areas where wildlife and human populations exist in close proximity.     

A condition metric for Eucalyptus woodland derived from expert evaluations

The evaluation of ecosystem quality is important for land‐management and land‐use planning. Evaluation is unavoidably subjective, and robust metrics must be based on consensus and the structured use of observations. We devised a transparent and repeatable process for building and testing ecosystem metrics based on expert data. We gathered quantitative evaluation data on the quality of hypothetical grassy woodland sites from experts. We used these data to train a model (an ensemble of 30 bagged regression trees) capable of predicting the perceived quality of similar hypothetical woodlands based on a set of 13 site variables as inputs (e.g., cover of shrubs, richness of native forbs). These variables can be measured at any site and the model implemented in a spreadsheet as a metric of woodland quality. We also investigated the number of experts required to produce an opinion data set sufficient for the construction of a metric. The model produced evaluations similar to those provided by experts, as shown by assessing the model's quality scores of expert‐evaluated test sites not used to train the model. We applied the metric to 13 woodland conservation reserves and asked managers of these sites to independently evaluate their quality. To assess metric performance, we compared the model's evaluation of site quality with the managers’ evaluations through multidimensional scaling. The metric performed relatively well, plotting close to the center of the space defined by the evaluators. Given the method provides data‐driven consensus and repeatability, which no single human evaluator can provide, we suggest it is a valuable tool for evaluating ecosystem quality in real‐world contexts. We believe our approach is applicable to any ecosystem.     

Temporal change estimation of mercury concentrations in northern pike (Esox lucius L.) in Swedish lakes

Adequate temporal trend analysis of mercury (Hg) in freshwater ecosystems is critical to evaluate if actions from the human society have affected Hg concentrations ([Hg]) in fresh water biota. This study examined temporal change in [Hg] in Northern pike (Esox lucius L.) in Swedish freshwater lakes between 1994 and 2006. To achieve this were lake-specific, multiple-linear-regression models used to estimate pike [Hg], including indicator variables representing time and fish weight and their interactions. This approach permitted estimation of the direction and magnitude of temporal changes in 25 lakes selected from the Swedish national database on Hg in freshwater biota. A significant increase was found in 36% of the studied lakes with an average increase in pike [Hg] of 3.7 ± 6.7% per year that was found to be positively correlated with total organic carbon. For lakes with a significant temporal change the dataset was based on a mean of 30 fish, while for lakes with no temporal change it was based on a mean of 13 fish.     

Impacts of Meteorological Factors on PM10: Artificial Neural Networks (ANN) and Multiple Linear Regression (MLR) Approaches

In this study, prediction capacities of multi-linear regression (MLR) and artificial neural networks (ANN) onto coarse particulate matter (PM₁₀) concentrations were investigated. Different meteorological factors on particulate pollution were chosen for operating variables in the model analyses. Two different regions (urban and industrial) were identified in the region of Kocaeli, Turkey. All data sets were obtained from air quality monitoring network of the Ministry of Environment and Urban Planning, and 120 data sets were used in the MLR and ANN models. Regression equations explained the effects of the meteorological factors in MLR analyses. In the ANN model, backpropagation network with two hidden layers has achieved the best prediction efficiency. Determination coefficients and error values were examined for each model. ANN models displayed more accurate results compared to MLR.     

利用后向轨迹模式研究上海市PM2.5来源分布及传输特征

利用后向轨迹模型,结合上海 PM2．5的浓度数据计算了2012年6月27日-2013年6月26日以上海为起始点的后向轨迹,并通过轨迹相关的分析方法,研究不同来源区域对上海 PM2．5浓度的贡献影响。结果表明：长三角地区的排放对上海的贡献最为显著;苏北、山东等地区的排放对上海也有较明显的贡献;来自海面的贡献总体低于大陆。所采用的轨迹多元回归分析法为 PM2．5的来源分布及传输特征研究提供了新思路。     

中国北方农牧交错带2000—2010年生态环境状况分析

北方农牧交错带是中国生态环境的一条过渡带,也是中国重要的生态安全屏障带,生态环境脆弱,具有明显的不稳定性和敏感性。以中国北方农牧交错带为研究对象,构建了反映区域生态生产力、功能支撑力、制约因子和胁迫因子的农牧交错带生态环境状况评估体系,利用多源遥感数据、土地生态数据、环境数据、气象数据等多源信息综合评估2000—2010年间该地区生态质量状况及变化趋势。根据Linear Regression和K-means cluster分析表明,该区域生态环境质量以"一般"为主,10年间生态环境质量呈现整体"稳中变好"的特征,"良"和"一般"呈增加趋势,16个县生态功能改善,18个县生态生产力呈恢复状态,16个县土地退化程度降低。水分条件和环境负荷是影响农牧交错带的主要限制和胁迫因子,局部地区存在显著生态风险。