Boosted Regression Tree Analysis Research Articles

Climate and geochemistry interactions at different altitudes influence soil organic carbon turnover times in alpine grasslands

Soil organic carbon turnover time (τ) is a key component of the carbon cycle. However, the impact of geochemical and climatic conditions on τ remains poorly understood in high alpine regions. We investigated climatic variables [e.g. mean annual precipitation (MAP) and mean annual temperature (MAT)], as well as geochemical variables [e.g. soil total element content (Mn, Ti, Fe, Si, Al, Mg, Ca, Na, K, and P), soil clay content, and soil pH] and estimated τ in the 0–30 cm soil layer at 169 alpine grassland sites along a 3000 km-long transect on the Tibetan Plateau. We found that τ ranged from 4 to 289 years on the Tibetan Plateau and showed a decreasing trend from the northwest to southeast and an increasing trend with altitude. The estimated τ was 717863 years (mean with 95% confidence interval) in alpine meadows, which did not significantly differ from alpine steppes (768765 yr). Overall, using boosted regression tree analysis, geochemistry was the most important controlling factor for τ (54% of the relative effect on τ), followed by climate (36%), and altitude (10%). When examining the relative contribution of individual variables, we found that MAP was the primary predictor of τ, followed by soil Si content, and altitude. Notably, variation in τ was explained by precipitation rather than temperature. Altitude indirectly affected τ by regulating climatic and soil geochemical conditions. The direct negative effect of climate on τ was opposite the positive indirect effect of climate on τ via soil geochemistry. These results highlight the importance of considering the interactions of climatic and geochemical factors as well as hydrological conditions when predicting how carbon turnover in the soils of semi-arid alpine grasslands will respond to future climate change.

Integrated control of the cattle tick, Rhipicephalus australis (Acari: Ixodidae), in New Caledonia through the Pasture and Cattle Management method.

Development of the Pasture and Cattle Management (PCM) method is a priority to control the cattle tick, Rhipicephalus australis, in New Caledonia. The PCM method provides the foundation for sustainable integrated tick control because approximately 95% of cattle ticks in infested pastures are off the host in the non-parasitic life stages, and the practice of treating cattle intensely with chemical acaricides is a risk for the emergence of resistance to these active ingredients in commercial acaricidal products available for veterinary use. Here, we report the findings of an assessment survey to document the utility of the PCM method. Analyses of questionnaire data provided by 21 beef cattle producers describing their management of 37 herds informed how to (1) assess the ability of PCM to reduce acaricide use and (2) prioritize best practices and define recommendations to breeders promoting efficient tick control with minimum acaricide use. Boosted regression tree analysis showed a significant (p = 0.002) reduction of ≈33% in the number of acaricide treatments from 7.9 to 5.3 per year by using PCM. Of the 24 factors identified as potentially affecting acaricide use, six factors accounted for ≈86% of the variability in number of acaricide treatments applied annually. The six most influential factors involved farm characteristics as well as pasture and herd management recommendations. These results demonstrated the usefulness of PCM for integrated control of R. australis infestations while reducing acaricide use to improve cattle production in New Caledonia.

Ploidy state of aquatic macrophytes: Global distribution and drivers

Despite the increasing evidence that ploidy is a factor of importance for environmental adaptation, little work has hitherto been done about cytotype patterns in aquatic plants. We tested a set of hypotheses concerning the drivers of ploidy in macrophyte communities at worldwide scale, and also how ploidy state may be related to other attributes of aquatic plants such as community diversity, species range size and invasiveness. We built a global ploidy database with 1574 species (representing 45.0 % of the known total global pool) and allocated ploidy state to species occurring in 238 10 × 10° gridcells worldwide. We used Boosted Regression Trees Analysis to analyse the relationship of 16 global-scale spatial, landscape, environmental and biotic variables with ploidy state. Our results demonstrate that (1) Polyploid and mixed ploidy species occupied larger geographical ranges than haploid/diploid species. (2) There were contrasting latitudinal patterns between haploid/diploid and polyploid species with polyploidy increasing in importance with distance from the equator, while species richness measures also influenced ploidy state. (3) Climatic factors, especially temperature and evapotranspiration, are congruent with macrophyte species ploidy state, across the planet. Our data support the assertion that increased chromosome number is an adaptation associated with plant survival of harsh environmental conditions, especially in environmentally-stressed high latitudes. Our findings increase the generality of knowledge of the large-scale factors that drive cytotype patterns in vascular plant species at global scale. The ploidy database provides a resource to support further studies about the effects of varying environmental conditions on ploidy state in aquatic plants, at differing scales.

Movements of Non-Migrant European Eels in an Urbanised Channel Linking a Mediterranean Lagoon to the Sea

Transitional ecosystems and, particularly, Mediterranean lagoons represent important habitats for the European eel (Anguilla anguilla) population. In these habitats many anthropogenic pressures can disturb eel movements and, in turn, negatively affect the population. Despite the importance of movements during the non-migrant growing stage in eels, this topic is understudied in Mediterranean lagoons. We thus aim to describe the diel and seasonal phenology and the effect of environmental drivers on non-migrant eel movements. Videos obtained from an Adaptive Resolution Imaging Sonar (ARIS) acoustic camera that continuously recorded from October 2018 to April 2020 were processed to evaluate the daily number of eels swimming toward the lagoon. More than 60% of the 7207 eels observed were females with a size >45 cm. Movements were year-round and predominantly during the night. A Boosted Regression Tree analysis demonstrated that, among the 10 environmental drivers studied, flow velocity, water temperature, discharge of the main tributary, wind velocity and atmospheric pressure, had the strongest influence on eel movement activity. Non-migrant eel movements should be better incorporated into lagoon management plans through actions such as limiting dredging activities from 18:00 to midnight, especially when the water flows toward the lagoon and when the water temperature is higher than 12 °C.

Spatial-temporal trend of grassland net primary production and their driving factors in the Loess Plateau, China

Net primary productivity (NPP) of grassland is a key link and important part of the ecosystem's carbon cycle. We estimated the changes of NPP in grasslands of the Loess Plateau with unchanged land use types during 2000-2015 and analyzed its responses to the variation of main climate factors (annual precipitation, annual heavy rainfall, annual effective rainfall days, annual average temperature, annual maximum temperature, annual minimum temperature) using piecewise linear regression and Pearson correlation analysis. The driving factors of grassland NPP were further analyzed by pixel-by-pixel with boosted regression tree analysis. The results showed that annual mean grassland NPP in the Loess Plateau showed an increasing trend during the study period, with 51.3% of the total grassland area showing a significant increasing trend. The average increase rate of annual mean NPP declined from 15.23 g C·m-2·a-1 in 2000-2004 to 3.58 g C·m-2·a-1 in 2004-2015. There was a significant positive correlation between grassland NPP and precipitation, but negative correlation with temperature factors. Annual precipitation was the dominant climatic factor affecting NPP of the whole study area with the highest relative importance. Annual maximum temperature was the dominant driving force of grassland NPP of central Loess Plateau, while annual minimum temperature mainly affected the growth of grassland in high-altitude area of the western Loess Plateau.

Biophysical and anthropogenic influences on the status of Tonga's coral reefs and reef fish fishery.

Despite increasing threats to Tonga’s coral reefs from stressors that are both local (e.g. overfishing and pollution) and global (e.g. climate change), there is yet to be a systematic assessment of the status of the country’s coral reef ecosystem and reef fish fishery stocks. Here, we provide a national ecological assessment of Tonga’s coral reefs and reef fish fishery using ecological survey data from 375 sites throughout Tonga’s three main island groups (Ha’apai, Tongatapu and Vava’u), represented by seven key metrics of reef health and fish resource status. Boosted regression tree analysis was used to assess and describe the relative importance of 11 socio-environmental variables associated with these key metrics of reef condition. Mean live coral cover across Tonga was 18%, and showed a strong increase from north to south correlated with declining sea surface temperature, as well as with increasing distance from each provincial capital. Tongatapu, the southernmost island group, had 2.5 times greater coral cover than the northernmost group, Vava’u (24.9% and 10.4% respectively). Reef fish species richness and density were comparable throughout Tongatapu and the middle island group, Ha’apai (~35 species/transect and ~2500 fish/km2), but were significantly lower in Vava’u (~24 species/transect and ~1700 fish/km2). Spatial patterns in the reef fish assemblage were primarily influenced by habitat-associated variables (slope, structural complexity, and hard coral cover). The biomass of target reef fish was greatest in Ha’apai (~820 kg/ha) and lowest in Vava’u (~340 kg/ha), and was negatively associated with higher human influence and fishing activity. Overall mean reef fish biomass values suggest that Tonga’s reef fish fishery can be classified as moderately to heavily exploited, with 64% of sites having less than 500 kg/ha. This study provides critical baseline ecological information for Tonga’s coral reefs that will: (1) facilitate ongoing management and research; and (2) enable accurate reporting on conservation targets locally and internationally.

Species-specific preferences drive the differential effects of lake factors on fish production

As the global human population grows, it remains a top priority for communities, managers, policymakers, and stakeholders to maintain healthy, sustainable, and productive fisheries under continued global change. Here we used a dataset consisting of fish and lake characteristics for 536 lakes across Ontario, Canada, to test whether multiple climate, human, and biological factors differentially affect fish production (i.e., population biomass per hectare per year). We tested the hypothesis that temperature is the key driver of fisheries production by testing for the effects of multiple factors on the production of three top predatory fish species: cold-water lake trout (Salvelinus namaycush), cool-water walleye (Sander vitreus), and warm-water smallmouth bass (Micropterus dolomieu). Using boosted regression tree analyses, we found that lake trout production was most influenced by the volume of hypolimnetic habitat, walleye production was related to other climatic variables, and smallmouth bass production was most influenced by sampling day of the year followed by Secchi depth. Our results suggest that current fish production models — that only include temperature and body size — may oversimplify important ecological complexities and thus misinform management decisions because species respond differently to environmental drivers.

Soil physico-chemical properties are critical for predicting carbon storage and nutrient availability across Australia

Soil carbon and nutrient availability play crucial roles in ecosystem sustainability, and they are controlled by the interaction of climatic, biotic, and soil physico-chemical variables. Although soil physico-chemical properties have been recognized as vital variables for predicting soil organic carbon (SOC) and nutrients, their relative influence across broad geographical scales has yet to be evaluated when simultaneously considering many other drivers. Using boosted regression tree and structural equation modelling analyses of observations from topsoil (0–10 cm) and subsoil (20–30 cm) at 628 sites across Australia, we investigated the effects and relative influence of climate (mean annual temperature and aridity index), plant productivity, soil biodiversity (bacterial and fungal richness), and soil physical (clay and silt) and chemical (pH and iron) properties on SOC content and nutrient availability (i.e. nitrogen, phosphorus, and potassium). Among these variables, we found that soil physico-chemical properties primarily predicted the continent-scale SOC storage and nutrient availability. In contrast, climate, plant productivity, and soil biodiversity played relatively small roles. The importance of physico-chemical properties was evident across soil depths and ecosystem types (i.e. tropical, temperate, arid, and cropland). Our findings point to the need to better understand the role of soil physico-chemical properties in soil carbon and nutrient cycling and including these variables in predictions of SOC and nutrient dynamics at the ecosystem to continental scale.

Species traits predict the aryl hydrocarbon receptor 1 (AHR1) subtypes responsible for dioxin sensitivity in birds

Differences in avian sensitivity to dioxin-like compounds (DLCs) are directly attributable to the identities of amino acids at two sites within the ligand binding domain (LBD) of the aryl hydrocarbon receptor 1 (AHR1). Recent work suggests that by influencing avian exposure to naturally occurring dioxins, differences in diet, habitat, and migration may have influenced the evolution of three AHR1 LBD genotypes in birds: type 1 (high sensitivity), type 2 (moderate sensitivity), and type 3 (low sensitivity). Using a boosted regression tree (BRT) analysis, we built on previous work by examining the relationship between a comprehensive set of 17 species traits, phylogeny, and the AHR1 LBD across 89 avian species. The 17 traits explained a combined 74% of the model deviance, while phylogenetic relatedness explained only 26%. The strongest predictors of AHR1 LBD were incubation period and habitat type. We found that type 3 birds tended to occupy aquatic habitats, and, uniquely, we also found that type 3 birds tended to have slower developmental rates. We speculate that this reflects higher evolutionary exposure to naturally occurring dioxins in waterbirds and species with K-selected life histories. This study highlights the value of trait-based approaches in helping to understand differing avian species sensitivities to environmental contaminants.

Combine performance, draft position and playing position are poor predictors of player career outcomes in the Australian Football League.

Physical testing-based draft combines are undertaken across various sporting codes to inform talent selection. To determine the explanatory power of the Australian football league (AFL) draft combine, participants drafted between 1999–2016 (n = 1488) were assessed. Testing performance, draft selection order and playing position, AFL matches played, AFL player ranking points and AFL player rating points were collected as career outcomes. Boosted regression tree analysis revealed that position and draft selection order were the most explanatory variables of career outcomes. Linear modelling based on testing results is able to explain 4% of matches played and 3% of in-game performance measures. Each individual combine test explained <2% of the matches played outcome. Draft selection order demonstrated mixed results for career outcomes relative to playing position. For instance, key forwards and draft selection order were observed as a slight negative relationship using the AFL Player Ranking points career outcome measure. These findings indicate that the AFL draft combine is a poor measure for informing talent selection, thus providing minimal utility for the practices investigated in this study.

Change of impervious surface area and its impacts on urban landscape: an example of Shenyang between 2010 and 2017

ABSTRACT Introduction One of the most striking features of urbanization is the replacement of the original natural land cover type by artificial impervious surface area (ISA). However, the extent of the contribution of various environmental factors, especially the growth of 3D space to ISA expansion, and the scope and mechanism of their influences in dramatically expanding cities, are yet to be determined. The boosted regression tree (BRT) model was adopted to analyze the main influencing factors and driving mechanisms of ISA change in Shenyang, China between 2010 and 2017. Outcomes The nearly complete-coverage ISA (≥0.7) increased from 42% in 2010 to 47% in 2017. The percentage of landscape with a high ISA fraction increased, while the landscape evenness and diversity of ISA decreased. The BRT analysis revealed that elevation, regional population density, and landscape class had the largest influences on the change of urban ISA, contributing 22.55%, 18.16%, and 11.18% to the model, respectively. Conclusion Overall, topographic and socioeconomic factors had the greatest influence on urban ISA change in Shenyang, followed by land use type and building pattern indices. The trend of high aggregation was strong in large commercial and residential areas. The 3D expansion of the city had an influence on its areal expansion.

Effects of Landscape Pattern Change on Water Yield and Nonpoint Source Pollution in the Hun-Taizi River Watershed, China.

Understanding the influence of landscape pattern changes on water yield (WYLD) and nutrient yield is a key topic for water resource management and nonpoint source (NPS) pollution reduction. The annual WYLD and NPS pollution were estimated in 2004 and 2015 with the calibrated and validated Soil and Water Assessment Tool (SWAT) in the Hun-Taizi River watershed. The impact of land use and landscape pattern changes on the annual WYLD and NPS loading changes were analyzed with a boosted regression tree (BRT) and redundancy analysis (RDA). The results showed that WYLD had a positive correlation with dry farmland and built-up area; however, a negative correlation with paddy field and water area, with the relative contribution of 42.03%, 23.79%, 17.06%, and 13.55%, respectively. The change in nutrient yield was positively correlated with changes in dry farmland, built-up area, and water area but negatively with forestland, according to the BRT model. Landscape patterns had an important influence on WYLD and NPS pollution. A large unfragmented forestland may improve water quality, while a large concentrated dry farmland results in water quality deterioration due to NPS pollution. Water quality is more likely degraded when land uses are complex and scattered with many small patches in a forestland dominated watershed.

Biophysical Habitat Features Explain Colonization and Size Distribution of Introduced Trochus (Gastropoda)

The trochid gastropod, Rochia nilotica is harvested in small-scale fisheries throughout Southeast Asia and the Pacific Islands, many of which were created from inter-country translocations. This species is found on structurally complex shallow reefs, but specific habitat requirements have not yet been quantified in order the maximize the success of future translocations and help future-proof the fishery in changing ocean conditions. At 28 sites around Samoa, where the species was introduced in the early 2000s, we measured a suite of habitat variables along transects in which R. nilotica were counted and the shell sizes measured in a parallel study. Boosted regression tree analyses revealed that R. nilotica were most abundant at reef locations that were shallow, with fairly consistent depth, high cover of branching coral, low cover of erect macroalgae, low wave exposure and high surface complexity. Smaller individuals were associated with wide reef flats and high cover of branching coral, whereas larger animals occurred in deeper water with high surface complexity. Multivariate analyses showed this species to be a habitat generalist, sharing much of its niche with an endemic herbivorous gastropod, Tectus pyramis. Future stocking programs should focus on sites with habitats optimal for both adults and juveniles. R. nilotica populations are likely to be especially affected by broadscale stressors that result in declines in live coral cover and substratum complexity and increasing coverage of macroalgae on coral reefs.

Current climate overrides historical effects on species richness and range size of freshwater plants in Europe and North America

Abstract The latitudinal diversity gradient hypothesis suggests that species richness should be highest at low latitudes, whereas Rapoport's rule states that largest ranges ought to be found for species at high latitudes. However, there is no consensus over these patterns and their underlying drivers in the freshwater realm. We investigated species richness and mean range size of freshwater plants in 50 × 50 km grid cells across Europe (40°N–71°N) and North America (25°N–78°N), supplemented with data based on 1° latitudinal bands for mean range size. We were especially interested to find out whether there are similarities and differences in these ecogeographical patterns and their underlying drivers between the continents due to their contrasting historical characteristics, spatial extent and topography. First, we used partial regression to reveal whether species richness and mean range size of freshwater plants have a linear or quadratic relationship with latitude. Second, we employed variation partitioning based on partial regression to model relationships between plant species richness and mean range size and four explanatory variable groups (i.e. environmental features, current climate, historical climate and geographical location). Third, we utilized boosted regression tree analysis to further investigate species richness and mean range size of freshwater plants in relation to a set of explanatory variables. Our results revealed that species richness showed relatively similar patterns in relation to latitude between the continents. Similarly, mean range size trends were alike in North America whether we used 50 × 50 km grid cell data or 1° latitudinal bands. Instead, different patterns in mean range size emerged between the used datasets in Europe. For both species richness and mean range size, current climate (with different individual predictor variables) was the main driver in both the continents, but historical effects had a small influence on the response variables. Synthesis. Our findings indicated that major ecogeographical rules can strongly vary for the same taxonomic group across broad scales between continents. It is also premature to rely solely on well‐known terrestrial taxonomic groups when drawing generalizations about ecogeographical rules.

Soil physico-chemical properties are more important than microbial diversity and enzyme activity in controlling carbon and nitrogen stocks near Sydney, Australia

Soil organic carbon (SOC) and available nitrogen (N) stocks are controlled by the complex interplay of soil physical, chemical, and biological conditions. However, the interrelations of SOC or available N with these drivers as well as their relative importance are rarely evaluated quantitatively. Using investigations of SOC density (SOCD) and available N density (ND) with other detailed soil properties of topsoil (0–10 cm) and subsoil (20–30 cm) from 33 sites under different ecosystems in the vicinity of an eddy flux tower near Sydney, Australia, we investigated the controls of soil physical, chemical, and biological properties (a total of 19 variables) on SOCD and available ND. Structural equation models showed that only physical and chemical properties significantly and directly affected SOCD and available ND. Among these variables, physical and chemical properties were the most influential factors, while the relative influences of microbial biodiversity and enzyme activity were small based on boosted regression tree analysis. In addition, the effects of variables on SOCD and available ND differed between the topsoil and subsoil. In the topsoil, soil physical properties had the highest relative influence followed by chemical properties, enzyme activities, and microbial biodiversity; in the subsoil, however, soil chemical properties had the highest relative importance followed by physical properties, enzyme activities, and microbial biodiversity. This comprehensive soil characterization provides the biogeochemical context for ecosystem carbon cycling being monitored at a nearby eddy flux tower, and demonstrates the importance of including accurate measurements of soil physical and chemical properties to reduce uncertainty in soil C and N predictions in process-based models. However, this is a local-scale study, and large-scale studies are warranted to gain further understanding on this issue.

Long‐term Dynamics and Drivers of Coral and Macroalgal Cover on Inshore Reefs of the Great Barrier Reef Marine Park

Quantifying the role of biophysical and anthropogenic drivers of coral reef ecosystem processes can inform management strategies that aim to maintain or restore ecosystem structure and productivity. However, few studies have examined the combined effects of multiple drivers, partitioned their impacts, or established threshold values that may trigger shifts in benthic cover. Inshore fringing reefs of the Great Barrier Reef Marine Park (GBRMP) occur in high-sediment, high-nutrient environments and are under increasing pressure from multiple acute and chronic stressors. Despite world-leading management, including networks of no-take marine reserves, relative declines in hard coral cover of 40–50% have occurred in recent years, with localized but persistent shifts from coral to macroalgal dominance on some reefs. Here we use boosted regression tree analyses to test the relative importance of multiple biophysical drivers on coral and macroalgal cover using a long-term (12–18 yr) data set collected from reefs at four island groups. Coral and macroalgal cover were negatively correlated at all island groups, and particularly when macroalgal cover was above 20%. Although reefs at each island group had different disturbance-and-recovery histories, degree heating weeks (DHW) and routine wave exposure consistently emerged as common drivers of coral and macroalgal cover. In addition, different combinations of sea-surface temperature, nutrient and turbidity parameters, exposure to high turbidity (primary) floodwater, depth, grazing fish density, farming damselfish density, and management zoning variously contributed to changes in coral and macroalgal cover at each island group. Clear threshold values were apparent for multiple drivers including wave exposure, depth, and degree heating weeks for coral cover, and depth, degree heating weeks, chlorophyll a, and cyclone exposure for macroalgal cover, however, all threshold values were variable among island groups. Our findings demonstrate that inshore coral reef communities are typically structured by broadscale climatic perturbations, superimposed upon unique sets of local-scale drivers. Although rapidly escalating climate change impacts are the largest threat to coral reefs of the GBRMP and globally, our findings suggest that proactive management actions that effectively reduce chronic stressors at local scales should contribute to improved reef resistance and recovery potential following acute climatic disturbances.

Complex interactions among successional trajectories and climate govern spatial resilience after severe windstorms in central Wisconsin, USA

Resilience is a concept central to the field of ecology, but our understanding of resilience is not sufficient to predict when and where large changes in species composition might occur following disturbances, particularly under climate change. Our objective was to estimate how wind disturbance shapes landscape-level patterns of engineering resilience, defined as the recovery of total biomass and species composition after a windstorm, under climate change in central Wisconsin. We used a spatially-explicit, forest simulation model (LANDIS-II) to simulate how windstorms and climate change affect forest succession and used boosted regression tree analysis to isolate the important drivers of resilience. At mid-century, biomass fully recovered to current conditions, but neither biomass nor species composition completely recovered at the end of the century. As expected, resilience was lower in the south, but by the end of the century, resilience was low throughout the landscape. Disturbance and species’ characteristics (e.g., the amount of area disturbed and the number of species) explained half of the variation in resilience, while temperature and soil moisture comprised only 17% collectively. Our results illustrate substantial spatial patterns of resilience at landscape scales, while documenting the potential for overall declines in resilience through time. Species diversity and windstorm size were far more important than temperature and soil moisture in driving long term trends in resilience. Finally, our research highlights the utility of using machine learning (e.g., boosted regression trees) to discern the underlying mechanisms of landscape-scale processes when using complex spatially-interactive and non-deterministic simulation models.

Long-term dynamics and drivers of coral and macroalgal cover on inshore reefs of the Great Barrier Reef Marine Park.

Quantifying the role of biophysical and anthropogenic drivers of coral reef ecosystem processes can inform management strategies that aim to maintain or restore ecosystem structure and productivity. However, few studies have examined the combined effects of multiple drivers, partitioned their impacts, or established threshold values that may trigger shifts in benthic cover. Inshore fringing reefs of the Great Barrier Reef Marine Park (GBRMP) occur in high-sediment, high-nutrient environments and are under increasing pressure from multiple acute and chronic stressors. Despite world-leading management, including networks of no-take marine reserves, relative declines in hard coral cover of 40-50% have occurred in recent years, with localized but persistent shifts from coral to macroalgal dominance on some reefs. Here we use boosted regression tree analyses to test the relative importance of multiple biophysical drivers on coral and macroalgal cover using a long-term (12-18yr) data set collected from reefs at four island groups. Coral and macroalgal cover were negatively correlated at all island groups, and particularly when macroalgal cover was above 20%. Although reefs at each island group had different disturbance-and-recovery histories, degree heating weeks (DHW) and routine wave exposure consistently emerged as common drivers of coral and macroalgal cover. In addition, different combinations of sea-surface temperature, nutrient and turbidity parameters, exposure to high turbidity (primary) floodwater, depth, grazing fish density, farming damselfish density, and management zoning variously contributed to changes in coral and macroalgal cover at each island group. Clear threshold values were apparent for multiple drivers including wave exposure, depth, and degree heating weeks for coral cover, and depth, degree heating weeks, chlorophyll a, and cyclone exposure for macroalgal cover, however, all threshold values were variable among island groups. Our findings demonstrate that inshore coral reef communities are typically structured by broadscale climatic perturbations, superimposed upon unique sets of local-scale drivers. Although rapidly escalating climate change impacts are the largest threat to coral reefs of the GBRMP and globally, our findings suggest that proactive management actions that effectively reduce chronic stressors at local scales should contribute to improved reef resistance and recovery potential following acute climatic disturbances.

Environmental factors and river network position allow prediction of benthic community assemblies: A model of nematode metacommunities

The field of metacommunity studies is growing rapidly, including recent applications to river networks. Most of these studies have targeted a single river network but whether their findings are relevant to other river systems is unknown. This study investigated the influence of environmental, spatial and temporal parameters on the community structure of nematodes in the river networks of the Elbe and Rhine. We asked whether the variance in community structure was better explained by spatial variables representing the watercourse than by overland distances. After determining the patterns in the Elbe river network, we tested whether they also explained the Rhine data. The Elbe data were evaluated using a boosted regression tree analysis. The predictive ability of the model was then assessed using the Rhine data. In addition to strong temporal dynamics, environmental factors were more important than spatial factors in structuring riverine nematode communities. Community structure was more strongly influenced by watercourse than by Euclidean distances. Application of the model’s predictions to the Rhine data correlated significantly with field observations. Our model shows that the consequences of changes in environmental factors or habitat connectivity for aquatic communities across different river networks are quantifiable.

Relative contribution of biotic and abiotic factors to stemflow production and funneling efficiency: A long-term field study on a xerophytic shrub species in Tengger Desert of northern China

Stemflow production has been reported to be influenced by a suite of biotic and abiotic factors. Nevertheless, relative contributions of biotic and abiotic factors to stemflow production and funneling efficiency were largely unclear due to complex interactions among those factors. In this study, stemflow of nine xerophytic shrubs of Caragana korshinskii were measured in nearly nine growing seasons from 2010 to 2018 within a desert area of northern China, accompanying with observing on six biotic variables (shrub morphological attributes) and ten abiotic variables (meteorological conditions). We performed boosted regression trees (BRT) model to evaluate the relative contribution (θ) of each biotic and abiotic variable to stemflow volume (SFv), stemflow percentage (SFp), and funneling ratio (FR), associating with partial dependence plots (PDPs) to visualize the effects of individual explanatory variables on SFv, SFp, and FR, respectively. Generally, we observed that larger shrubs generated more SFv, while had lower SFp and FR. BRT analysis demonstrated that biotic variables outweighed abiotic variables by 1.5-fold as to their contribution to SFv, whereas abiotic variables prevailed for SFp and FR, respectively. Differences in θ between variables for SFv (CV = 146%) were much pronounced than for SFp (CV = 57%) and FR (CV = 26%), with the foremost three influential variables for SFv ranking in descending orders of θ by rainfall amount (33.2%), basal area (20.9%), projected canopy area (15.1%), and that for SFp by projected canopy area (13.2%), rainfall amount (13.0%), and air temperature (11.1%), and that for FR by antecedent dry period (9.1%), rainfall amount (8.4%), and wind speed (8.3%). The quantitative and mechanistic explanations regarding the effects of biotic and abiotic variables on three stemflow parameters from the present study are expected to be applicable to other shrub species within arid and semi-arid ecosystems.