Factor Structure Underlying Components of Allostatic Load

Allostatic load is a commonly used metric of health risk based on the hypothesis that recurrent exposure to environmental demands (e.g., stress) engenders a progressive dysregulation of multiple physiological systems. Prominent indicators of response to environmental challenges, such as stress-related hormones, sympatho-vagal balance, or inflammatory cytokines, comprise primary allostatic mediators. Secondary mediators reflect ensuing biological alterations that accumulate over time and confer risk for clinical disease but overlap substantially with a second metric of health risk, the metabolic syndrome. Whether allostatic load mediators covary and thus warrant treatment as a unitary construct remains to be established and, in particular, the relation of allostatic load parameters to the metabolic syndrome requires elucidation. Here, we employ confirmatory factor analysis to test: 1) whether a single common factor underlies variation in physiological systems associated with allostatic load; and 2) whether allostatic load parameters continue to load on a single common factor if a second factor representing the metabolic syndrome is also modeled. Participants were 645 adults from Allegheny County, PA (30–54 years old, 82% non-Hispanic white, 52% female) who were free of confounding medications. Model fitting supported a single, second-order factor underlying variance in the allostatic load components available in this study (metabolic, inflammatory and vagal measures). Further, this common factor reflecting covariation among allostatic load components persisted when a latent factor representing metabolic syndrome facets was conjointly modeled. Overall, this study provides novel evidence that the modeled allostatic load components do share common variance as hypothesized. Moreover, the common variance suggests the existence of statistical coherence above and beyond that attributable to the metabolic syndrome.     

Coral disease following massive bleaching in 2005 causes 60% decline in coral cover on reefs in the US Virgin Islands

In the northeast Caribbean, doldrum-like conditions combined with elevated water temperatures in the summer/fall 2005 created the most severe coral bleaching event ever documented within this region. Video monitoring of 100 randomly chosen, permanent transects at five study sites in the US Virgin Islands revealed over 90% of the scleractinian coral cover showed signs of thermal stress by paling or becoming completely white. Lower water temperatures in October allowed some re-coloring of corals; however, a subsequent unprecedented regional outbreak of coral disease affected all sites. Five known diseases or syndromes were recorded; however, most lesions showed signs similar to white plague. Nineteen scleractinian species were affected by disease, with >90% of the disease-induced lesions occurring on the genus Montastraea. The disease outbreak peaked several months after the onset of bleaching at all sites but did not occur at the same time. The mean number of disease-induced lesions increased 51-fold and the mean area of disease-associated mortality increased 13-fold when compared with pre-bleaching disease levels. In the 12 months following the onset of bleaching, coral cover declined at all sites (average loss: 51.5%, range: 42.4–61.8%) reducing the five-site average from 21.4% before bleaching to 10.3% with most mortality caused by white plague disease, not bleaching. Continued losses through October 2007 reduced the average coral cover of the five sites to 8.3% (average 2-year loss: 61.1%, range: 53.0–79.3%). Mean cover by M. annularis (complex) decreased 51%, Colpophyllia natans 78% and Agaricia agaricites 87%. Isolated disease outbreaks have been documented before in the Virgin Islands, but never as widespread or devastating as the one that occurred after the 2005 Caribbean coral-bleaching event. This study provides insight into the effects of continued seawater warming and subsequent coral bleaching events in the Caribbean and highlights the need to understand links between coral bleaching and disease.     

Predictor species: Improving assessments of rare species occurrence by modeling environmental co‐responses

Designing an effective conservation strategy requires understanding where rare species are located. Because rare species can be difficult to find, ecologists often identify other species called conservation surrogates that can help inform the distribution of rare species. Species distribution models typically rely on environmental data when predicting the occurrence of species, neglecting the effect of species' co‐occurrences and biotic interactions. Here, we present a new approach that uses Bayesian networks to improve predictions by modeling environmental co‐responses among species. For species from a European peat bog community, our approach consistently performs better than single‐species models and better than conventional multi‐species approaches that include the presence of nontarget species as additional independent variables in regression models. Our approach performs particularly well with rare species and when calibration data are limited. Furthermore, we identify a group of “predictor species” that are relatively common, insensitive to the presence of other species, and can be used to improve occurrence predictions of rare species. Predictor species are distinct from other categories of conservation surrogates such as umbrella or indicator species, which motivates focused data collection of predictor species to enhance conservation practices.     

Predictive Modeling of Coral Disease Distribution within a Reef System

Diseases often display complex and distinct associations with their environment due to differences in etiology, modes of transmission between hosts, and the shifting balance between pathogen virulence and host resistance. Statistical modeling has been underutilized in coral disease research to explore the spatial patterns that result from this triad of interactions. We tested the hypotheses that: 1) coral diseases show distinct associations with multiple environmental factors, 2) incorporating interactions (synergistic collinearities) among environmental variables is important when predicting coral disease spatial patterns, and 3) modeling overall coral disease prevalence (the prevalence of multiple diseases as a single proportion value) will increase predictive error relative to modeling the same diseases independently. Four coral diseases: Porites growth anomalies (PorGA), Porites tissue loss (PorTL), Porites trematodiasis (PorTrem), and Montipora white syndrome (MWS), and their interactions with 17 predictor variables were modeled using boosted regression trees (BRT) within a reef system in Hawaii. Each disease showed distinct associations with the predictors. Environmental predictors showing the strongest overall associations with the coral diseases were both biotic and abiotic. PorGA was optimally predicted by a negative association with turbidity, PorTL and MWS by declines in butterflyfish and juvenile parrotfish abundance respectively, and PorTrem by a modal relationship with Porites host cover. Incorporating interactions among predictor variables contributed to the predictive power of our models, particularly for PorTrem. Combining diseases (using overall disease prevalence as the model response), led to an average six-fold increase in cross-validation predictive deviance over modeling the diseases individually. We therefore recommend coral diseases to be modeled separately, unless known to have etiologies that respond in a similar manner to particular environmental conditions. Predictive statistical modeling can help to increase our understanding of coral disease ecology worldwide.     

Caribbean coral diseases: primary transmission or secondary infection?

Over the last 40 years, disease outbreaks have significantly reduced coral populations throughout the Caribbean. Most coral‐disease models assume that coral diseases are contagious and that pathogens are transmitted from infected to susceptible hosts. However, this assumption has not been rigorously tested. We used spatial epidemiology to examine disease clustering, at scales ranging from meters to tens of kilometers, to determine whether three of the most common Caribbean coral diseases, (i) yellow‐band disease, (ii) dark‐spot syndrome, and (iii) white‐plague disease, were spatially clustered. For all three diseases, we found no consistent evidence of disease clustering and, therefore, these diseases did not follow a contagious‐disease model. We suggest that the expression of some coral diseases is instead a two‐step process. First, environmental thresholds are exceeded. Second, these environmental conditions either weaken the corals, which are then more susceptible to infection, or the conditions increase the virulence or abundance of pathogens. Exceeding such environmental thresholds will most likely become increasingly common in rapidly warming oceans, leading to more frequent coral‐disease outbreaks.     

Bird species richness patterns of northern Taiwan: primary productivity, human population density, and habitat heterogeneity

Energy, climate, habitat heterogeneity, and human activity are important correlates of spatial variation in species richness. We examined the correlation between species richness and these variables using the birds that breed in northern Taiwan. We conducted general linear models (GLMs) and spatial correlation models to examine the relationship between bird species richness (BSR) and environmental variables. We found that normalized difference vegetation index (NDVI) was the most important predictor of BSR. We suggest productivity is the primary process of BSR. Additionally, we hypothesized that scale dependency might exist in the relationship between BSR and NDVI in Taiwan. Human population density, the second most important factor, was inversely correlated with BSR. The factor and BSR did not have similar response to NDVI, which contradicted observations in most of the previous studies on human population vs. species richness. We proposed that the human population density had an effect on NDVI, which in turn had an effect on BSR. Moreover, we hypothesized that the contradiction between our study and the previous studies might arise from a higher level of human disturbance in Taiwan than in other areas. The necessity of conserving native species in intensively developed lowlands of Taiwan cannot be overemphasized. Number of land cover type was another significant predictor of BSR. Habitat heterogeneity may have an effect on BSR in Taiwan.     

A high-latitude coral community with an uncertain future: Stetson Bank, northwestern Gulf of Mexico

Limited data exist that detail trends in benthic community composition of high-latitude coral communities. As anthropogenic stressors are projected to increase in number and intensity, long-term monitoring datasets are essential to understanding community stability and ecosystem resilience. In 1993, a long-term monitoring program was initiated at Stetson Bank, in the Gulf of Mexico. Over the course of this monitoring, a major shift in community structure occurred, in which the coral-sponge community was replaced by an algal-dominated community. During the initial years of this study, the coral community at Stetson Bank was relatively stable. Beginning in the late 1990s, sponge cover began a steady decline from over 30 % to less than 25 %. Then, in 2005, the benthic community underwent a further significant change when living coral cover declined from 30 % to less than 8 % and sponges declined to less than 20 % benthic cover. This abrupt shift corresponded with a Caribbean-wide bleaching event in 2005 that caused major mortality of Stetson Bank corals. Previous bleaching events at Stetson Bank did not result in wide-scale coral mortality. Several environmental parameters may have contributed to the rapid decline in this benthic community. We suggest that the combined effects of coastal runoff and elevated temperatures contributed to the observed shift. We present an analysis of 15 years of monitoring data spanning from 1993 to 2008; this dataset provides both a biological baseline and a multiyear trend analysis of the community structure for a high-latitude coral-sponge community in the face of changing climatic conditions.     

Emergence of white pine needle damage in the northeastern United States is associated with changes in pathogen pressure in response to climate change

The defoliation of the eastern white pine (Pinus strobus) across the northeastern United States is an escalating concern threatening the ecological health of northern forests and economic vitality of the region's lumber industry. First documented in the spring of 2010 affecting 24 328 hectares in the state of Maine, white pine needle damage (WPND) has continued to spread and is now well established in all New England states. While causal agents of WPND are known, current research is lacking in both sampling distribution and the specific environmental factor(s) that affect the development and spread of this disease complex. This study aims to construct a more detailed distribution map of the four primary causal agents within the region, as well as utilize long‐term WPND monitoring plots and data collected from land‐based weather stations to develop a climatic model to predict the severity of defoliation events in the proceeding year. Sampling results showed a greater distribution of WPND than previously reported. WPND was generally found in forest stands that compromised >50% eastern white pine by basal area. No single species, nor a specific combination of species had a dominating presence in particular states or regions, thus supporting the disease complex theory that WPND is neither caused by an individual species nor by a specific combination of species. In addition, regional weather data confirmed the trend of increasing temperature and precipitation observed in this region with the previous year's May, June, and July rainfall being the best predictor of defoliation events in the following year. Climatic models were developed to aid land managers in predicting disease severity and accordingly adjust their management decisions. Our results clearly demonstrate the role changing climate patterns have on the health of eastern white pine in the northeastern United States.     

Models as tools for understanding past,recent and future changes in large lakes

Large lakes currently exhibit ecosystem responses to environmental changes such as climate and land use changes, nutrient loading, toxic contaminants, hydrological modifications and invasive species. These sources have impacted lake ecosystems over a number of years in various combinations and often in a spatially heterogeneous pattern. At the same time, many different kinds of mathematical models have been developed to help to understand ecosystem processes and improve cost-effective management. Here, the advantages and limitations of models and sources of uncertainty will be discussed. From these considerations and in view of the multiple environmental pressures, the following emerging issues still have to be met in order to improve the understanding of ecosystem function and management of large lakes: (1) the inclusion of thresholds and points-of-no-return; (2) construction of general models to simulate biogeochemical processes for a large number of lakes rather than for individual systems; (3) improvement of the understanding of spatio-temporal variability to quantify biogeochemical fluxes accurately; and (4) inclusion of biogeochemical linkages between terrestrial and aquatic ecosystems in model approaches to assess the effects of external environmental pressures such as land-use changes. The inclusion of the above-mentioned issues would substantially improve models as tools for the scientific understanding and cost-effective management of large lakes that are subject to multiple environmental pressures in a changing future.     

Strong relationship between coral bleaching and growth anomalies in massive Porites

Reports of coral diseases are increasing and may result from human land use and climate change conditions such as increased water temperature, coral bleaching, runoff from land, and changes in the ecology of heavily fished reefs. We examined a stable coral syndrome or a growth anomaly [ Porite growth anomaly (PGA)] (skeletal tissue anomaly, hyperplasia, or 'tumor') that was present in 0–15% of massive Porites colonies in 12 Kenyan reef lagoons. At the level of the calice morphology, this growth anomaly showed larger calices with less distance between calices and some calices with higher than normal numbers of septa, which indicate the influence of microboring organisms. Scanning electron micrographs of affected corals revealed a high abundance of fungal hyphae, a potential microboring pathogenic agent. To test the hypothesis that the PGA covaries with environmental variables, we evaluated its prevalence in relationship to 16 parameters of water quality, temperature, intensity of bleaching, benthic composition, and management at the end of the 2005 warm season. Stepwise regression models found eight environmental variables significantly associated with the frequency of the PGA, and the site's bleaching intensity was the most strongly associated variable. When bleaching intensity was removed from the dataset, the concentration of phosphorus was the one significant and positively associated variable, which suggest that the other significant environmental variables were associated with bleaching and not the growth anomalies. Our hypothetical model of causation is that the patchy loss of symbionts, often associated with bleaching, reduces calcification, increases susceptibility to pathogens, and allows endolithic fungi to perforate the skeleton creating a porous and anomalous growth of the skeleton. Consequently, we suggest that the frequency of skeletal growth anomalies is expected to increase with the frequency of coral bleaching.     

Chronic nutrient enrichment increases prevalence and severity of coral disease and bleaching

Nutrient loading is one of the strongest drivers of marine habitat degradation. Yet, the link between nutrients and disease epizootics in marine organisms is often tenuous and supported only by correlative data. Here, we present experimental evidence that chronic nutrient exposure leads to increases in both disease prevalence and severity and coral bleaching in scleractinian corals, the major habitat‐forming organisms in tropical reefs. Over 3 years, from June 2009 to June 2012, we continuously exposed areas of a coral reef to elevated levels of nitrogen and phosphorus. At the termination of the enrichment, we surveyed over 1200 scleractinian corals for signs of disease or bleaching. Siderastrea siderea corals within enrichment plots had a twofold increase in both the prevalence and severity of disease compared with corals in unenriched control plots. In addition, elevated nutrient loading increased coral bleaching; Agaricia spp. of corals exposed to nutrients suffered a 3.5‐fold increase in bleaching frequency relative to control corals, providing empirical support for a hypothesized link between nutrient loading and bleaching‐induced coral declines. However, 1 year later, after nutrient enrichment had been terminated for 10 months, there were no differences in coral disease or coral bleaching prevalence between the previously enriched and control treatments. Given that our experimental enrichments were well within the ranges of ambient nutrient concentrations found on many degraded reefs worldwide, these data provide strong empirical support to the idea that coastal nutrient loading is one of the major factors contributing to the increasing levels of both coral disease and coral bleaching. Yet, these data also suggest that simple improvements to water quality may be an effective way to mitigate some coral disease epizootics and the corresponding loss of coral cover in the future.     

Augmented coral reef monitoring using a stationary reef monitoring system

Coral reef monitoring is a reliable tool to assess the effect of climate change as corals are sensitive to increases in water temperatures between 30 °C and 35 °C resulting in bleaching - a whitening process when the corals lose their color and the reefs begin to die. Existing satellite-based monitoring products facilitate coral bleaching monitoring over large spatial scales, but their use in predicting local scale stress that influences the bleaching severity across reefs is limited. In this paper, we describe a Stationary Reef Monitoring System (SRMS) that monitors the time evolution of coral reefs through the photography of nearby coral clusters. Simultaneously, the SRMS measures and records environmental parameters such as temperature, solar irradiance (PAR), and salinity in the waters surrounding the coral colonies. When deployed in the sea, the SRMS detected a 0.1–0.4 °C variability in temperature between the in situ and satellite datasets. The SRMS uses color photography along with quantitative data on environmental parameters to monitor the health of corals and eliminates the need for physical/visual verification of coral health by a diver. By this approach, one can determine the stress thresholds of corals and identify the vulnerable and resilient reefs so as to prioritize conservation efforts.     

Characterization of Households and its Implications for the Vegetation of Urban Ecosystems

Our understanding of the dynamics of urban ecosystems can be enhanced by examining the multidimensional social characteristics of households. To this end, we investigated the relative significance of three social theories of household structure—population, lifestyle behavior, and social stratification—to the distribution of vegetation cover in Baltimore, Maryland, USA. Our ability to assess the relative significance of these theories depended on fine-scale social and biophysical data. We distinguished among vegetation in three areas hypothesized to be differentially linked to these social theories: riparian areas, private lands, and public rights-of-way (PROWs). Using a multimodel inferential approach, we found that variation of vegetation cover in riparian areas was not explained by any of the three theories and that lifestyle behavior was the best predictor of vegetation cover on private lands. Surprisingly, lifestyle behavior was also the best predictor of vegetation cover in PROWs. The inclusion of a quadratic term for housing age significantly improved the models. Based on these research results, we question the exclusive use of income and education as the standard variables to explain variations in vegetation cover in urban ecological systems. We further suggest that the management of urban vegetation can be improved by developing environmental marketing strategies that address the underlying household motivations for and participation in local land management.     

Extended geographic distribution of several Indo-Pacific coral reef diseases

Other than coral bleaching, few coral diseases or diseases of other reef organisms have been reported from Japan. This is the first report of lesions similar to Porites ulcerative white spots (PUWS), brown band disease (BrB), pigmentation response (PR), and crustose coralline white syndrome (CCWS) for this region. To assess the health status and disease prevalence, qualitative and quantitative surveys (3 belt transects of 100 m2 each on each reef) were performed in March and September 2010 on 2 reefs of the Ginowan-Ooyama reef complex off Okinawa, and 2 protected reefs off Zamani Island, in the Kerama Islands 40 km west of Okinawa. Overall, mean (±SD) disease prevalence was higher in Ginowan-Ooyama (9.7 ± 7.9%) compared to Zamami (3.6 ± 4.6%). Porites lutea was most affected by PUWS at Ooyama (23.1 ± 10.4 vs. 4.5 ± 5.2%). White syndrome (WS) mostly affected Acropora cytherea (12. 5 ± 18.0%) in Zamami and Oxipora lacera (10.2 ± 10%) in Ooyama. Growth anomalies (GA) and BrB were only observed on A. cytherea (8.3 ± 6.2%) and A. nobilis (0.8%) at Zamami. Black band disease affected Pachyseris speciosa (6.0 ± 4.6%) in Ooyama only. Pigmentation responses (PR) were common in massive Porites in both localities (2.6 ± 1.9 and 5.6 ± 2.3% respectively). Crustose coralline white syndrome (CCWS) was observed in both localities. These results significantly expand the geographic distribution of PUWS, BrB, PR and CCWS in the Indo-Pacific, indicating that the northernmost coral reefs in the western Pacific are susceptible to a larger number of coral diseases than previously thought.     

What is winter? Modeling spatial variation in bat host traits and hibernation and their implications for overwintering energetics

White‐nose syndrome (WNS) has decimated hibernating bat populations across eastern and central North America for over a decade. Disease severity is driven by the interaction between bat characteristics, the cold‐loving fungal agent, and the hibernation environment. While we further improve hibernation energetics models, we have yet to examine how spatial heterogeneity in host traits is linked to survival in this disease system. Here, we develop predictive spatial models of body mass for the little brown myotis (Myotis lucifugus) and reassess previous definitions of the duration of hibernation of this species. Using data from published literature, public databases, local experts, and our own fieldwork, we fit a series of generalized linear models with hypothesized abiotic drivers to create distribution‐wide predictions of prehibernation body fat and hibernation duration. Our results provide improved estimations of hibernation duration and identify a scaling relationship between body mass and body fat; this relationship allows for the first continuous estimates of prehibernation body mass and fat across the species'' distribution. We used these results to inform a hibernation energetic model to create spatially varying fat use estimates for M. lucifugus. These results predict WNS mortality of M. lucifugus populations in western North America may be comparable to the substantial die‐off observed in eastern and central populations.     

Sea surface temperatures and coral reef bleaching off La Parguera, Puerto Rico (northeastern Caribbean Sea)

 Much recent attention has been given to coral reef bleaching because of its widespread occurrence, damage to reefs, and possible connection to global change. There is still debate about the relationship between temperature and widespread bleaching. We compared coral reef bleaching at La Parguera, Puerto Rico to a 30-y (1966–1995) record of sea surface temperature (SST) at the same location. The last eight years of the La Parguera SST record have all had greater than average maximum temperatures; over the past 30 y maximum summer temperature has increased 0.7 °C. Coral reef bleaching has been particularly frequent since the middle 1980s. The years 1969, 1987, 1990, and 1995 were especially noteworthy for the severity of bleaching in Puerto Rico. Seven different annual temperature indices were devised to determine the extent to which they could predict severe coral bleaching episodes. Three of these, maximum daily SST, days >29.5 °C, and days >30 °C predict correctly the four years with severe bleaching. A log-log linear relationship was found between SST and the number of days in a given year above that SST at which severe coral beaching was observed. However, the intra-annual relationship between temperature and the incidence of bleaching suggests that no one simple predictor of the onset of coral bleaching within a year may be applicable. Accepted: 17 March 1998     

Learning to predict large-scale coral bleaching from past events: A Bayesian approach using remotely sensed data,in-situ data,and environmental proxies

Ocean warming and coral bleaching are patchy phenomena over a wide range of scales. This paper is part of a larger study that aims to understand the relationship between heat stress and ecological impact caused by the 2002-bleaching event in the Great Barrier Reef (GBR). We used a Bayesian belief network (BBN) as a framework to refine our prior beliefs and investigate dependencies among a series of proxies that attempt to characterize potential drivers and responses: the remotely sensed environmental stress (sea surface temperature — SST); the geographic setting; and topographic and ecological attributes of reef sites for which we had field data on bleaching impact. Sensitivity analyses helped us to refine and update our beliefs in a manner that improved our capacity to hindcast areas of high and low bleaching impact. Our best predictive capacity came by combining proxies for a sites heat stress in 2002 (remotely sensed), acclimatization temperatures (remote sensed), the ease with which it could be cooled by tidal mixing (modeled), and type of coral community present at a sample of survey sites (field data). The potential for the outlined methodology to deliver a transparent decision support tool to aid in the process of identifying a series of locations whose inclusion in a network of protected areas would help to spread the risk of bleaching is discussed.     

Modeling Soil Organic Carbon at Regional Scale by Combining Multi-Spectral Images with Laboratory Spectra

There is a great challenge in combining soil proximal spectra and remote sensing spectra to improve the accuracy of soil organic carbon (SOC) models. This is primarily because mixing of spectral data from different sources and technologies to improve soil models is still in its infancy. The first objective of this study was to integrate information of SOC derived from visible near-infrared reflectance (Vis-NIR) spectra in the laboratory with remote sensing (RS) images to improve predictions of topsoil SOC in the Skjern river catchment, Denmark. The second objective was to improve SOC prediction results by separately modeling uplands and wetlands. A total of 328 topsoil samples were collected and analyzed for SOC. Satellite Pour l’Observation de la Terre (SPOT5), Landsat Data Continuity Mission (Landsat 8) images, laboratory Vis-NIR and other ancillary environmental data including terrain parameters and soil maps were compiled to predict topsoil SOC using Cubist regression and Bayesian kriging. The results showed that the model developed from RS data, ancillary environmental data and laboratory spectral data yielded a lower root mean square error (RMSE) (2.8%) and higher R² (0.59) than the model developed from only RS data and ancillary environmental data (RMSE: 3.6%, R²: 0.46). Plant-available water (PAW) was the most important predictor for all the models because of its close relationship with soil organic matter content. Moreover, vegetation indices, such as the Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI), were very important predictors in SOC spatial models. Furthermore, the ‘upland model’ was able to more accurately predict SOC compared with the ‘upland & wetland model’. However, the separately calibrated ‘upland and wetland model’ did not improve the prediction accuracy for wetland sites, since it was not possible to adequately discriminate the vegetation in the RS summer images. We conclude that laboratory Vis-NIR spectroscopy adds critical information that significantly improves the prediction accuracy of SOC compared to using RS data alone. We recommend the incorporation of laboratory spectra with RS data and other environmental data to improve soil spatial modeling and digital soil mapping (DSM).     

环境微生物研究中机器学习算法及应用

微生物在环境中无处不在，它们不仅是生物地球化学循环和环境演化的关键参与者，也在环境监测、生态治理和保护中发挥着重要作用。随着高通量技术的发展，大量微生物数据产生，运用机器学习对环境微生物大数据进行建模和分析，在微生物标志物识别、污染物预测和环境质量预测等领域的科学研究和社会应用方面均具有重要意义。机器学习可分为监督学习和无监督学习2大类。在微生物组学研究当中，无监督学习通过聚类、降维等方法高效地学习输入数据的特征，进而对微生物数据进行整合和归类。监督学习运用有特征和标记的微生物数据集训练模型，在面对只有特征没有标记的数据时可以判断出标记，从而实现对新数据的分类、识别和预测。然而，复杂的机器学习算法通常以牺牲可解释性为代价来重点关注模型预测的准确性。机器学习模型通常可以看作预测特定结果的“黑匣子”，即对模型如何得出预测所知甚少。为了将机器学习更多地运用于微生物组学研究、提高我们提取有价值的微生物信息的能力，深入了解机器学习算法、提高模型的可解释性尤为重要。本文主要介绍在环境微生物领域常用的机器学习算法和基于微生物组数据的机器学习模型的构建步骤，包括特征选择、算法选择、模型构建和评估等，并对各种机器学习模型在环境微生物领域的应用进行综述，深入探究微生物组与周围环境之间的关联，探讨提高模型可解释性的方法，并为未来环境监测、环境健康预测提供科学参考。     

The effects of sea surface temperature anomalies on oceanic coral reef systems in the southwestern tropical Atlantic

In 2010, high sea surface temperatures that were recorded in several parts of the world and caused coral bleaching and coral mortality were also recorded in the southwest Atlantic Ocean, between latitudes 0°S and 8°S. This paper reports on coral bleaching and diseases in Rocas Atoll and Fernando de Noronha archipelago and examines their relationship with sea surface temperature (SST) anomalies recorded by PIRATA buoys located at 8°S30°W, 0°S35°W, and 0°S23°W. Adjusted satellite data were used to derive SST climatological means at buoy sites and to derive anomalies at reef sites. The whole region was affected by the elevated temperature anomaly that persisted through 2010, reaching 1.67 °C above average at reef sites and 1.83 °C above average at buoys sites. A significant positive relationship was found between the percentage of coral bleaching that was observed on reef formations and the corresponding HotSpot SST anomaly recorded by both satellite and buoys. These results indicate that the warming observed in the ocean waters was followed by a warming at the reefs. The percentage of bleached corals persisting after the subsidence of the thermal stress, and disease prevalence increased through 2010, after two periods of thermal stress. The in situ temperature anomaly observed during the 2009–2010 El Niño event was equivalent to the anomaly observed during the 1997–1998 El Niño event, explaining similar bleaching intensity. Continued monitoring efforts are necessary to further assess the relationship between bleaching severity and PIRATA SST anomalies and improve the use of this new dataset in future regional bleaching predictions.