Handling covariates subject to limits of detection in regression

In the environmental health sciences, measurements of toxic exposures are often constrained by a lower limit called the limit of detection (LOD), with observations below this limit called non-detects. Although valid inference may be obtained by excluding non-detects in the estimation of exposure effects, this practice can lead to substantial reduction in power to detect a significant effect, depending on the proportion of censoring and the closeness of the effect size to the null value. Therefore, a variety of methods have been commonly used in the environmental science literature to substitute values for the non-detects for the purpose of estimating exposure effects, including ad hoc values such as ${LOD/2, LOD/\sqrt{2}}$ and LOD. Another method substitutes the expected value of the non-detects, i.e., E[X|X ≤?LOD] but this requires that the inference be robust to mild miss-specifications in the distribution of the exposure variable. In this paper, we demonstrate that the estimate of the exposure effect is extremely sensitive to ad-hoc substitutions and moderate distribution miss-specifications under the conditions of large sample sizes and moderate effect size, potentially leading to biased estimates. We propose instead the use of the generalized gamma distribution to estimate imputed values for the non-detects, and show that this method avoids the risk of distribution miss-specification among the class of distributions represented by the generalized gamma distribution. A multiple imputation-based procedure is employed to estimate the regression parameters. Compared to the method of excluding non-detects, the proposed method can substantially increase the power to detect a significant effect when the effect size is close to the null value in small samples with moderate levels of censoring (?≤ 50%), without compromising the coverage and relative bias of the estimates.     

Influence of Tibetan Plateau on vegetation distributions in East Asia: a modeling perspective

Understanding the current distribution of vegetation and its interaction with climate regularity and surface irregularity is important for predicting its future change. In comparison with other regions of the world, the vegetation distribution in East Asia is unique, such as the location of desert belt, subtropical evergreen forests and deciduous broadleaved forest; but the underlying causes are still not clear. While some researchers have proposed on the effect of monsoons as causal mechanisms, others have suggested that the direct cause might be the rising of the Tibetan Plateau (TP). However, the relative importance of these two mechanisms remains unknown, and cannot be directly tested at large spatial and temporal scales. Here we construct the possible vegetation distribution in this area (15°N–60°N, 60°E–150°E) with assumption that there were no TP (e.g., its elevation were 1000 m and roughly equal to its surroundings), using a General Circulation Model and the Holdridge Life Zones System. Our simulations demonstrated that TP affected the vegetation distribution and patterns in East Asia significantly. Without TP the area of warm temperate forest and subtropical forest would increase, but desert area would decrease. The effects of TP should be considered when studying vegetation dynamics in East Asia under climate change. The results are also useful for explaining the plant biogeography in this region.     

基于SPOT-VGT的黄河流域植被覆盖时空演变

植被是土地覆盖中的最主要部分,是连接土壤、大气和生物等要素的自然"纽带"。植被覆盖动态变化对全球能量循环和物质循环具有重要影响,是全球变化研究的重要内容之一。黄河流域作为我国重要的粮食生产基地,其环境变化直接影响到流域经济的可持续发展。为了快速准确地提取地表植被状况,了解黄河流域生态环境,利用1998—2011年的SPOT-VGT遥感数据,结合地理信息技术,采用均值法和趋势分析法对黄河流域植被NDVI时空分布特征和变化趋势进行了动态监测。结果表明,(1)黄河流域14年NDVI均值的空间分布整体特征是东南部平原、盆地和西部山地植被状况要好于北部地区。其次,黄河流域属于干旱半干旱地区,植被发育主要依赖于水文条件,所以沿黄河干流和支流区域也具有较高的植被NDVI值。(2)黄河流域植被NDVI年均值近14年间整体呈缓慢增长趋势,1998—2000年呈现急剧减少态势,2001—2003年出现了较为快速的增长,2004—2011年又出现了较长时间的连续增长过程。(3)黄河流域植被NDVI基本不变的区域约占研究区总面积的71.13%;植被NDVI轻微改善的区域约占流域总面积的27.30%,且主要分布于流域东南部的盆地、平原和西部的山地、丘陵地区,植被NDVI退化的区域面积约占流域总面积的0.98%。黄河流域自1998年以来,植被NDVI整体在不断提高,生态环境在不断改善。     

基于SPOT—VGT的黄河流域植被覆盖时空演变

植被是土地覆盖中的最主要部分,是连接土壤、大气和生物等要素的自然“纽带”。植被覆盖动态变化对全球能量循环和物质循环具有重要影响,是全球变化研究的重要内容之一。黄河流域作为我国重要的粮食生产基地,其环境变化直接影响到流域经济的可持续发展。为了快速准确地提取地表植被状况,了解黄河流域生态环境,利用1998—2011年的SPOT-VGT遥感数据,结合地理信息技术,采用均值法和趋势分析法对黄河流域植被NDVI时空分布特征和变化趋势进行了动态监测。结果表明,（1）黄河流域14年NDVI均值的空间分布整体特征是东南部平原、盆地和西部山地植被状况要好于北部地区。其次,黄河流域属于干旱半干旱地区,植被发育主要依赖于水文条件,所以沿黄河干流和支流区域也具有较高的植被NDVI值。（2）黄河流域植被NDVl年均值近14年间整体呈缓慢增长趋势,1998--2000年呈现急剧减少态势,2001--2003年出现了较为快速的增长,2004--2011年又出现了较长时间的连续增长过程。（3）黄河流域植被NDVI基本不变的区域约占研究区总面积的71．13％;植被NDVI轻微改善的区域约占流域总面积的27．30％,且主要分布于流域东南部的盆地、平原和西部的山地、丘陵地区,植被NDVI退化的区域面积约占流域总面积的0．98％。黄河流域自1998年以来,植被NDVI整体在不断提高,生态环境在不断改善。     

Spatio-temporal modelling of ground vegetation development in mountain spruce forests

More complex models of forest ecosystems are required to understand how land-cover changes can impact vegetation dynamics and spatial pattern. In order to document spatio-temporal modelling abilities, the observations conducted in the declined climax mountain Norway spruce forest during the recovery period (1995-2006) are used for simulation and spatial analysis in the GIS environment. The developed spatio-temporal model is used for simulation of forest vegetation dynamics in a mountain spruce forest in the framework of regeneration processes after stress from air pollution. In order to explore the spatial and temporal phenomena of regeneration processes, the spatio-temporal model is based on a large set of ordinary differential equations that solve dynamic processes in sets of microsites arranged in grids for each ground vegetation species and each age group of Norway spruce seedlings. The spatial extent of the explored site is composed of a set of 50 × 50 microsites. Each microsite is represented by a square with dimensions of 1 m × 1 m. The presented simulation studies are mainly focused on seedlings from the seed year 1992, in order to explore the longest monitored time series of survival. It is based on exponential growth models that are related to the environmental conditions for each microsite. The canopy gaps based on estimates of the local crown projected area, the soil type layer, and the dominant grass density are used to provide case simulation studies. The first case study simulates the influence of microsite positions in relation to the local tree crown projections on the survival of spruce seedlings. It is assumed that the density of the trees is the main factor that determines the light and heat supply to the ground level of the Norway spruce seedlings. The second case study extends the previous study to include terms that determine the growth ratio in dependence on the crown projection area. The third case study provides further extensions in order to simulate growth ratio relations to the local soil type. The fourth case study demonstrates the local influence of the dominant grasses, such as Avenella flexuosa and Calamagrostis villosa, on the natural regeneration of Norway spruce. Starting from the conditions at the sites before the recovery period, the case simulation studies are able to project the short-term succession for a regeneration decade and the approximate long-term development. In addition to the standard simulation procedures based on solution of ordinary differential equations, spatio-temporal modelling in the GIS environment is able to provide spatial data management, analysis and visualization of the data.     

Spatio-temporal analysis of alpine ecotones: A spatial explicit model targeting altitudinal vegetation shifts

There is general agreement in literature that Alpine vegetation belt ecotones have shown a trend of upward migration in the last few decades. Despite the potential of such shifts as indicators of global change effects in mountain ecosystems, there are relatively few works focused on their assessment in a systematic and spatially explicit way. In this work our aim is to quantify the altitudinal shifts and analyse the spatial pattern dynamics of mountain ecotones. We developed a novel procedure to delineate the current and former state of three characteristic mountain ecotones, which we formalised as forest, tree and tundra lines. Our approach is based on the recognition of altitudinal extreme outposts identified with ecotone locations at a slope scale. The integration of multi-temporal datasets allows the identification and quantification of altitudinal advances and retreats in the outpost locations for a given period. We tested the method in a section of the Italian Alps for the period 1957-2003. Results show a general trend of an increase in altitude for the three ecotones, despite the occurrence of occasional decreases. We estimate decadal altitude increments of 25 m for forest line, 13 m for treeline and 11 m for tundra line. We also identified changes in ecotone spatial morphology between the two dates, with significant implications in connectivity and colonisation dynamics.     

江苏省植被NPP时空特征及气候因素的影响

利用2000—2006年的EOS/MODIS卫星遥感资料,对江苏省植被净初级生产力（NPP）时空特征及气候变化对其影响进行分析。在ArcGIS软件中,建立线性回归方程获得NPP的变化斜率,分析7 a间各像元NPP的空间变化趋势。计算各像元的NPP数值与气候要素的线性相关系数,为定量阐述气候变化对植被生长的影响提供依据。结果表明,江苏省植被NPP 7 a平均值为506.6 g.m-2.a-1（以C计）,比全国同期NPP数值高出约40%。NPP表现出明显的年际变化,2004年植被年均NPP最大为530.6 g.m-2.a-1（以C计）,2000年最小,为481.1 g.m-2.a-1（以C计）。空间分布上NPP表现为东南高于西北,沿海高于内陆。2000—2006年江苏省有76%的区域植被NPP表现为显著增加,仅江苏南部少数区域表现出减少的趋势。除苏南少数区域外,气候因素控制着NPP的时空变化规律。其中气温的升高和太阳辐射的增加促进NPP提高,而降水量的增加引起NPP的降低。     

广州市1990-2005年植被覆盖度的时空变化特征

植被是生态系统的最重要组成部分,以植被覆盖度为指标研究区域植被的时空动态特征,是生态系统健康评价的前提和必要基础。以1990、1995、2000和2005年4个时相的TM遥感影像为数据源,对其进行大气辐射校正以及进行空间图像运算,生成广州市不同时序的植被覆盖度图,以此分析广州市植被覆盖度的时空变化特征。结果表明,广州市过去15a植被覆盖度变化很大,从58.75%下降到46.06%,特别是中心城区和番禺区内的植被覆盖度变化更显著。植被覆盖度下降最多的时间段是1990—1995年,恢复阶段是2000—2005年间。但是,从空间分布上看,植被恢复的区域主要是从化市,市区仍然处于继续下降状态,值得各方面关注,因为市区是人口密度和经济高度集中的区域,也是生态元素最缺乏的区域。     

Using single- and multi-target regression trees and ensembles to model a compound index of vegetation condition

An important consideration in conservation and biodiversity planning is an appreciation of the condition or integrity of ecosystems. In this study, we have applied various machine learning methods to the problem of predicting the condition or quality of the remnant indigenous vegetation across an extensive area of south-eastern Australia—the state of Victoria. The field data were obtained using the ‘habitat hectares’ approach. This rapid assessment technique produces multiple scores that describe the condition of various attributes of the vegetation at a given site. Multiple sites were assessed and subsequently circumscribed with GIS and remote-sensed data.     

Projected vegetation changes for the American Southwest: combined dynamic modeling and bioclimatic-envelope approach

This study focuses on potential impacts of 21st century climate change on vegetation in the Southwest United States, based on debiased and interpolated climate projections from 17 global climate models used in the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Among these models a warming trend is universal, but projected changes in precipitation vary in sign and magnitude. Two independent methods are applied: a dynamic global vegetation model to assess changes in plant functional types and bioclimatic envelope modeling to assess changes in individual tree and shrub species and biodiversity. The former approach investigates broad responses of plant functional types to climate change, while considering competition, disturbances, and carbon fertilization, while the latter approach focuses on the response of individual plant species, and net biodiversity, to climate change. The dynamic model simulates a region-wide reduction in vegetation cover during the 21st century, with a partial replacement of evergreen trees with grasses in the mountains of Colorado and Utah, except at the highest elevations, where tree cover increases. Across southern Arizona, central New Mexico, and eastern Colorado, grass cover declines, in some cases abruptly. Due to the prevalent warming trend among all 17 climate models, vegetation cover declines in the 21st century, with the greatest vegetation losses associated with models that project a drying trend. The inclusion of the carbon fertilization effect largely ameliorates the projected vegetation loss. Based on bioclimatic envelope modeling for the 21st century, the number of tree and shrub species that are expected to experience robust declines in range likely outweighs the number of species that are expected to expand in range. Dramatic shifts in plant species richness are projected, with declines in the high-elevation evergreen forests, increases in the eastern New Mexico prairies, and a northward shift of the Sonoran Desert biodiversity maximum.     

Predictive vegetation modeling for conservation: impact of error propagation from digital elevation data.

The effect of digital elevation model (DEM) error on environmental variables, and subsequently on predictive habitat models, has not been explored. Based on an error analysis of a DEM, multiple error realizations of the DEM were created and used to develop both direct and indirect environmental variables for input to predictive habitat models. The study explores the effects of DEM error and the resultant uncertainty of results on typical steps in the modeling procedure for prediction of vegetation species presence/absence. Results indicate that all of these steps and results, including the statistical significance of environmental variables, shapes of species response curves in generalized additive models (GAMs), stepwise model selection, coefficients and standard errors for generalized linear models (GLMs), prediction accuracy (Cohen's kappa and AUC), and spatial extent of predictions, were greatly affected by this type of error. Error in the DEM can affect the reliability of interpretations of model results and level of accuracy in predictions, as well as the spatial extent of the predictions. We suggest that the sensitivity of DEM-derived environmental variables to error in the DEM should be considered before including them in the modeling processes.     

Spatio-temporal modeling of striped-bass egg, larval movement, and fate in the San Francisco Bay-Delta

Most models developed for the movement and fate of eggs and larvae of aquatic species are based on a particle tracking approach. Although this method has many advantages due to its high flexibility, particle tracking may become computationally intensive for complex geometries and when large numbers of particles are needed to simulate the population properly. In continuous models based on advection and dispersion mechanisms, the computational burden is independent of the size of the population. We developed a continuous fate and transport model for striped bass eggs and larvae in the San Francisco Bay-Delta. The model predicts the concentration of eggs and larvae at any location over time. The method of moments was used to account for the effect of temperature and age on the transition of eggs to larvae and larvae to juveniles. Egg and larval mortality were represented as functions of temperature, and eggs also experienced settling mortality. The fate and transport model used the same one-dimensional spatial grid as the existing Delta Simulation Model II (DSM2) hydrodynamics model. DSM2 output of flow rates, water depths, and cross-sectional areas were inputted into the fate and transport model to determine transport. The model was applied to striped bass eggs and larvae data collected during years 1990-1994; agreement between the modeled and the measured data was acceptable in most cases. Exploratory simulations were performed to demonstrate how the model could be used to evaluate the effects on egg and larval survival and total juvenile production of water diversions for supply and agricultural use and changes in the long-term mean water temperature. The model can be further used to examine the impact of various operation strategies in the San Francisco Bay-Delta, where diversion losses of early life stages of fishes remain a major management issue.     

Nutrient deficiency in dune slack pioneer vegetation: a review

A review of results of fertilization experiments in wet dune slacks is presented. In most cases the above-ground biomass appeared to be limited by nitrogen availability. Primary phosphorus limitation was assessed only once in a dune slack where sod cutting had been applied very recently. In most other case studies phosphorus limits biomass production after nitrogen deficiency was lifted. Potassium availability is of minor importance for biomass production in this type of ecosystem. Singular nitrogen additions led to increased dominance ofCarex andJuncus species as well as perennial grasses, such asAgrostis stolonifera andCalamagrostis epigejos. A combined addition of nitrogen and phosphorus led to total dominance of grasses, while the characteristic basiphilous pioneer species (including mosses) decreased or even disappeared. Certain mechanisms are considered which may maintain nutrient availability in slacks with basiphilous pioneer vegetation at a low level, despite of the accumulation of nutrients in the developing organic soil layer. Some implications for management and further research are discussed.     

The effects of disturbance based management on the dynamics of Mediterranean vegetation: A hierarchical and spatially explicit modeling approach

The eastern Mediterranean region has been subjected to intensive human disturbance in the past 10,000 years, mainly in the forms of agro-pastoral activities such as grazing, shrub clearing, and prescribed burning. This disturbance history resulted in the formation of highly heterogeneous landscapes, characterized by high biodiversity. Recent changes in human activities have resulted in a decrease of landscape heterogeneity, leading to decreasing biodiversity and increasing fire risk. To conserve heterogeneity, land managers apply disturbance based management practices, using the same activities that created and maintained landscape heterogeneity in the past. However, the long-term and large-scale outcomes of these disturbances are often unknown, due to the complex response of Mediterranean vegetation to disturbance. Here we report on a spatially explicit, hybrid, and spatially hierarchical ecological model developed by us. The model attempts to predict the outcome of various disturbance based management activities on the long-term spatio-temporal dynamics of five common Mediterranean vegetation types. The model uses a spatially explicit state and transition formulation, with continuous transition functions. Model simulations were conducted on a Mediterranean landscape in Northern Israel, incorporating various disturbance practices that are common in the region. Simulation results highlight the potential of disturbance based management as a tool for conserving landscape heterogeneity, as well as the complex interactions between disturbances and the spatial structure of the landscape in Mediterranean regions.     

Modelling count data based on weakly dependent spatial covariates using a copula approach: application to rat sightings

Not all environmental processes are observed in a way that allows a straight forward easy modelling. Nevertheless, insights can also be gained by exploring weakly dependent covariates paying attention to details of the distribution. Using the concept of copulas, it is possible to explore the dependence of a multivariate distribution without the distortion of the marginal distribution functions acting on typical correlation measures. Furthermore, copulas turn the attention to the dependence across the entire range of the multivariate distribution and do not only summarise it in a single correlation measure. In our application, we study counts of rat sightings in the city of Madrid. The brown rat lives with mankind and adversely affects public health by transmission of diseases, bites and allergies. Better understanding behavioural and spatial correlation aspects of this species can contribute to its effective management and control. We explore weakly to moderately correlated covariates based on distances to broken sewers, feeding grounds and markets as well as population density. The use of copulas is motivated by the different dependence structures of the four covariates and the asymmetries therein. In order to deal with the discrete zero-inflated counts, we present a new approach that assigns conditional random ranks to discrete data. This way, we mimic an underlying continuous variable easing the vine copula estimation, but do not destroy the dependence as in a uniform randomisation. We show that a 5-dimensional vine copula model is able to capture the dependence in our application.     

A varying-coefficient regression approach to modeling the effects of wind speed on the dispersion of pollutants

Environmental and Ecological Statistics - The real-world monitoring system of air pollution ordinarily collects data about pollutant concentration levels at pollution sources and monitors stations...     

Variation in groundwater composition and decalcification depth in a dune slack: effects on basiphilous vegetation

Basiphilous, open, species-rich vegetation types of young dune slacks have declined throughout Europe in recent years, and have largely been replaced by often acidophilous, tall marsh and scrub vegetation. This succession appears to be accelerated by a decrease in the discharge of calcareous groundwater from sandy ridges or small dune hummocks. The present study deals with spatial and temporal variation in the chemical composition of the groundwater in the upper metres of the soil of a degraded dune slack complex on the Dutch barrier island of Schiermonnikoog, with emphasis on (1) groundwater composition, (2) water level and (3) decalcification patterns. The main aim was to assess perspectives for restoring basiphilous vegetation types which had been abundant in this slack from 1954 to 1977. The depth of decalcification was related to former hydrological conditions along a transect of 200 m. Acidifying effects of rainfall were reflected in the chemical composition of the groundwater below small dune hummocks within the slack. Distinct precipitation water lenses, poor in dissolved ions, were formed under the dune hummocks during a wet period. This microtopography did not contribute to the discharge of calcareous groundwater to lowlying parts of the slack. Here, groundwater showed decreasing concentrations of the dissolved ions after a rain shower. Except for the peripheral sections of the slack—where upward seepage of groundwater (exfiltration)still occurs—infiltration conditions are now dominant in the slack. The consequences of the present hydrological conditions for restoration are briefly discussed.     

Mapping woodland vegetation in SW Ethiopia based on a hybrid model

SUMMARY

Despite the fact that indigenous inhabitants have sustainably managed the forests and woodlands for centuries, Ethiopia's woodland resources are now under severe pressure. At the request of the Ethiopian Government, in 1996/97 the Savanna Woodland Management Study Project was initiated by the German Technical Cooperation (GTZ). A detailed analysis of the woodland resources in SW Ethiopia and a vegetation map were produced. The map was generated using multispectral remote-sensing data, and provides details of the variety within Ethiopian woodlands. A hybrid data classification procedure was applied to incorporate relief and climatic conditions. After analysis of the remote sensing data by image processing, a spatial overlay was performed to incorporate the additional site condition factors. The vegetation map shows the spatial distribution of all existing woodland types in the area and enhances our knowledge of the rich biodiversity of SW Ethiopia. Such mapping gives us a better understanding of existing ecosystems and indigenous land-use systems, and hence, can lead to more sustainable management of the natural resources in the future.