Predicting tree species presence and basal area in Utah: A comparison of stochastic gradient boosting,generalized additive models,and tree-based methods

Many efforts are underway to produce broad-scale forest attribute maps by modelling forest class and structure variables collected in forest inventories as functions of satellite-based and biophysical information. Typically, variants of classification and regression trees implemented in Rulequest's^© See5 and Cubist (for binary and continuous responses, respectively) are the tools of choice in many of these applications. These tools are widely used in large remote sensing applications, but are not easily interpretable, do not have ties with survey estimation methods, and use proprietary unpublished algorithms. Consequently, three alternative modelling techniques were compared for mapping presence and basal area of 13 species located in the mountain ranges of Utah, USA. The modelling techniques compared included the widely used See5/Cubist, generalized additive models (GAMs), and stochastic gradient boosting (SGB). Model performance was evaluated using independent test data sets. Evaluation criteria for mapping species presence included specificity, sensitivity, Kappa, and area under the curve (AUC). Evaluation criteria for the continuous basal area variables included correlation and relative mean squared error. For predicting species presence (setting thresholds to maximize Kappa), SGB had higher values for the majority of the species for specificity and Kappa, while GAMs had higher values for the majority of the species for sensitivity. In evaluating resultant AUC values, GAM and/or SGB models had significantly better results than the See5 models where significant differences could be detected between models. For nine out of 13 species, basal area prediction results for all modelling techniques were poor (correlations less than 0.5 and relative mean squared errors greater than 0.8), but SGB provided the most stable predictions in these instances. SGB and Cubist performed equally well for modelling basal area for three species with moderate prediction success, while all three modelling tools produced comparably good predictions (correlation of 0.68 and relative mean squared error of 0.56) for one species.     

Modelling the relationship of driver license and offense history with fatal and serious injury (FSI) crash involvement

Introduction: Previous research has indicated that increases in traffic offenses are linked to increased crash involvement rates, making reductions in offending an appropriate measure for evaluating road safety interventions in the short-term. However, the extent to which traffic offending predicts fatal and serious injury (FSI) crash involvement risk is not well established, prompting this new Victorian (Australia) study. Method: A preliminary cluster analysis was performed to describe the offense data and assess FSI crash involvement risk for each cluster. While controlling demographic and licensing variables, the key traffic offenses that predict future FSI crash involvement were then identified. The large sample size allowed the use of machine learning methods such as random forests, gradient boosting, and Least Absolute Shrinkage and Selection Operator (LASSO) regression. This was done for the ‘all driver’ sample and five sometimes overlapping groups of drivers; the young, the elderly, and those with a motorcycle license, a heavy vehicle license endorsement and/or a history of license bans. Results: With the exception of the group of drivers who had a history of bans, offense history significantly improved the accuracy of models predicting future FSI crash involvement using demographic and licensing data, suggesting that traffic offenses may be an important factor to consider when analyzing FSI crash involvement risk and the effects of road safety countermeasures. Conclusions: The results are helpful for identifying driver groups to target with further road safety countermeasures, and for showing that machine learning methods have an important role to play in research of this nature. Practical Application: This research indicates with whom road safety interventions should particularly be applied. Changes to driver demerit policies to better target offenses related to FSI crash involvement and repeat traffic offenders, who are at greater risk of FSI crash involvement, are recommended.     

Mechanism on minimization of excess sludge in oxic-settlinganaerobic (OSA) process

The oxic-settling-anaerobic (OSA) process is a promising wastewater treatment technique for efficiently reducing sludge production and improving the stability of process operation. In this paper, the possible factors of sludge reduction such as sludge decay, uncoupled metabolism, and anaerobic oxidation with low sludge production were discussed in the OSA process. It has been confirmed that sludge decay is the decisive cause in the OSA process, accounting for 66.7% of sludge production reduction. Sludge decay includes hydrolysis and acidogenesis of dead microorganisms and particle organic carbon adsorbed in sludge floc and endogenous metabolism. By batch experiments, it has been proven that there is energetic uncoupling in the OSA system since microorganisms were exposed to alternative anaerobic and aerobic environment. It accounts for about 7.5% of sludge production reduction. Soluble chemical oxygen demand (SCOD) released from the anaerobic sludge tank in the OSA process was used as the substrate for cryptic growth. The substrate was used for anoxic denitrifying, anaerobic phosphorus release, sulfate reduction, and methane production. These anaerobic reactions in the sludge anaerobic tank have lower sludge production than in the aerobic oxidation when equivalent SCOD is consumed, which may lead to approximately 23% of sludge reduction in the OSA process. It has been concluded that multiple causes resulted in the minimization of excess sludge in the OSA system. The microbial community structure and diversity of sludge samples from the CAS (conventional activated sludge) and OSA systems were investigated by 16 SrDNA PCR-DG-DGGE (polymerase chain reaction-double gradient-denaturing gradient gel electrophoresis). DGGE profile and cluster analysis showed more abundant species in the OSA system contrasting to microbial communities in the CAS system.     

根肿(Plasmodiophora brassicae)休眠孢子的纯化和超声波破碎方法研究

根肿(Plasmodiophora brassicae Woronin)是严重危害十字花科植物的世界性病害.采用蔗糖密度梯度离心和超声波处理方法从油菜染病根部组织分离和提纯根肿休眠孢子.众多因素可以影响提纯休眠孢子的纯度,为了得到更高纯度和均匀度的休眠孢子的优化方法,我们对一系列影响因素如超声波破碎功率、pH、孢子破碎浓度、处理时间和不同的水溶物质等进行了研究.在pH值5.5左右、添加0.5%的二甲基亚砜、孢子浓度约为7×108、345W功率下超声波处理10min条件下获得了最优数量和均匀度的孢子,杂菌数量减少了51%以上,杂菌种类减少超过75%.超声波处理50min后,孢子破碎率稳定地达到90%以上,除菌率达99%以上.用此方法能在较短时间内获得高纯度和高均匀度的休眠孢子,为休眠孢子的深入研究提供了资料.图4表3参10     

用梯度氧化铝膜净化空气

研究了孔径沿径向由大到小自然过渡的梯度氧化铝陶瓷膜管对空气的过滤净化。梯度氧化铝膜管控制层平均孔径为 0 2 μm ,对于空气中粒径大于或等于 0 2 2 μm的颗粒 ,膜管的截留率达到 10 0 % ;细菌的总滤除率为 99 99%。提高操作压力和错流线速度都有利于提高膜的过滤通量。空气净化过程中截留在空气中的微粒沉积于梯度氧化铝膜的管内壁 ,防止膜管的深层堵塞 ,使用过程通量下降缓慢 ,一次性使用寿命为 4年 ,易于清洗 ,清洗后过滤通量和过滤效率仍达 10 0 %     

Environmentally stratified sampling design for the development of Great Lakes environmental indicators

Understanding the relationship between human disturbance and ecological response is essential to the process of indicator development. For large-scale observational studies, sites should be selected across gradients of anthropogenic stress, but such gradients are often unknown for a population of sites prior to site selection. Stress data available from public sources can be used in a geographic information system (GIS) to partially characterize environmental conditions for large geographic areas without visiting the sites. We divided the U.S. Great Lakes coastal region into 762 units consisting of a shoreline reach and drainage-shed and then summarized over 200 environmental variables in seven categories for the units using a GIS. Redundancy within the categories of environmental variables was reduced using principal components analysis. Environmental strata were generated from cluster analysis using principal component scores as input. To protect against site selection bias, sites were selected in random order from clusters. The site selection process allowed us to exclude sites that were inaccessible and was shown to successfully distribute sites across the range of environmental variation in our GIS data. This design has broad applicability when the goal is to develop ecological indicators using observational data from large-scale surveys.     

黄河口盐沼湿地盐地碱蓬和互花米草凋落物的分解特征

凋落物分解在湿地生态系统中扮演重要的角色。为探索潮汐梯度下凋落物分解过程中质量和营养元素变化情况,采用分解袋的方法,选择盐地碱蓬（Suaeda salsa）和互花米草（Spartina alterniflora）凋落物作为研究对象,沿着潮汐梯度开展凋落物分解的野外实验。结果表明：凋落物分解速率与凋落物类型、土壤盐度、含水率以及潮汐干扰强弱有关;潮汐干扰强的区域互花米草分解快于盐地碱蓬,潮汐干扰弱的区域盐地碱蓬分解快于互花米草,两种凋落物分解速率在0.00134~0.00234 d^-1之间。分解末期,凋落物都呈现C和N的净释放,盐地碱蓬凋落物平均释放了36.9%的C和55.8%的N,互花米草凋落物平均释放了53.1%的C和47.1%的N。本研究强调关注潮汐梯度下的凋落物分解,其在调节生物地球化学循环以及碳累积上具有重要意义。     

基于区域梯度式学习情景的长江经济带环境治理研究

长江经济带作为国家重大战略,面临开发与保护的巨大挑战,构建科学、高效的生态环境保护机制为其保驾护航成为重中之重。作为一个流域式生态共同体,区域统筹协调下的生态环境保护和治理是根本关键。本研究导入区域协同治理理念,提出构建梯度式渐进学习环境协同治理框架,并采用数据进行长江经济带环境污染治理的多情景测度与评价实证检验了梯度间以及梯度内部的环境保护与污染治理协调机制的潜在影响。最终,从中央、区域和地方三个层面提出梯度式学习视野下长江经济带环境治理策略。     

大气污染控制中高梯度磁分离技术的研究与应用

本文概述了高梯度磁分离技术在燃煤脱硫除灰、烟气除尘方面的应用与研究进展以及超导磁分离技术在燃煤脱硫方面的研究，并展望了超导磁分离技术在环境保护中的应用前景。     

Early development of vegetation in restored dune plant microhabitats on a nourished beach at Ocean City, New Jersey

Topography and vegetation of restored dunes on a developed barrier island were examined after a large-scale beach nourishment project. Restoration began in 1993 using sand-trapping fences andAmmophila breviligulata Fern. plantings. Subsequent growth of dunes was favored by installing new fences and suspending beach raking to accommodate nesting birds. Plant species richness, percent cover of vegetation, and height ofA. breviligulata were sampled in 1999 on seven shore perpendicular transects in six dune microhabitats (backdune, primary crest, mid-foredune, swale, seaward-most fenced ridge, incipient dune on the backbeach). A total of 26 plant taxa were found at all seven sites. Richness and percent cover were greatest in the backdune and crest, especially in locations that predated the 1992 nourishment. Richness was greater where fences enhanced stabilization. Fences initially compensate for time and space and allow vegetation to develop rapidly, but maintenance nourishment is required to protect against wave erosion and ensure long-term viability of habitat. An expanded environmental gradient is an option, where beach nourishment provides space for a species-rich crest and backdune to develop, while the incipient dune remains dynamic. Options where space is restricted include a dynamic, full-sized seaward section of a naturally functioning dune (truncated gradient) or a spatially restricted sampler of a wider natural dune (compressed gradient) maintained using fences. Expanded and truncated gradients may become self-sustaining and provide examples of natural cycles of change. Compressed gradients provide greater species richness and flood protection for the available space, but habitats are vulnerable to erosion, and resident views may be impaired.