Productivity of lentic ecosystems is well studied, and it is widely accepted that as nutrient inputs increase, productivity increases and lakes transition from lower trophic state (e.g., oligotrophic) to higher trophic states (e.g., eutrophic). These broad trophic state classifications are good predictors of ecosystem condition, services (e.g., recreation and esthetics), and disservices (e.g., harmful algal blooms). While the relationship between nutrients and trophic state provides reliable predictions, it requires in situ water quality data to parameterize the model. This limits the application of these models to lakes with existing and, more importantly, available water quality data. To address this, we take advantage of the availability of a large national lakes water quality database (i.e., the National Lakes Assessment), land‐use/land‐cover data, lake morphometry data, and other universally available data, and we apply data‐mining approaches to predict trophic state. Using these data and random forests, we first model chlorophyll a and then classify the resultant predictions into trophic states. The full model estimates chlorophyll a with both in situ and universally available data. The mean‐squared error and adjusted R2 of this model was 0.09 and 0.8, respectively. The second model uses universally available GIS data only. The mean‐squared error was 0.22, and the adjusted R2 was 0.48. The Kappa coefficients of the trophic state classifications derived from the chlorophyll a predictions were 0.57 for the full model and 0.29 for the “GIS‐only” model. Random forests extend the usefulness of the class predictions by providing prediction probabilities for each lake. This allows us to make trophic state predictions and also indicate the level of uncertainty around those predictions. For the full model, these predicted class probabilities ranged from 0.42 to 1. For the GIS‐only model, they ranged from 0.33 to 0.96. It is our conclusion that in situ data are required for better predictions, yet GIS and universally available data provide trophic state predictions, with estimated uncertainty, that still have the potential for a broad array of applications. The source code and data for this manuscript are available from https://github.com/USEPA/LakeTrophicModelling.

No abstract

Migratory stopover habitats are often not part of planning for conservation or new development projects. We identified potential stopover habitats within an avian migratory flyway and demonstrated how this information can guide the site-selection process for new development. We used the random forests modeling approach to map the distribution of predicted stopover habitat for the Whooping Crane (Grus americana), an endangered species whose migratory flyway overlaps with an area where wind energy development is expected to become increasingly important. We then used this information to identify areas for potential wind power development in a U.S. state within the flyway (Nebraska) that minimize conflicts between Whooping Crane stopover habitat and the development of clean, renewable energy sources. Up to 54% of our study area was predicted to be unsuitable as Whooping Crane stopover habitat and could be considered relatively low risk for conflicts between Whooping Cranes and wind energy development. We suggest that this type of analysis be incorporated into the habitat conservation planning process in areas where incidental take permits are being considered for Whooping Cranes or other species of concern. Field surveys should always be conducted prior to construction to verify model predictions and understand baseline conditions.

In addition to being used as a tool for ecological understanding, management and conservation of migratory waterfowl rely heavily on distribution models; yet these models have poor accuracy when compared to models of other bird groups. The goal of this study is to offer methods to enhance our ability to accurately model the spatial distributions of six migratory waterfowl species. This goal is accomplished by creating models based on species-specific annual cycles and introducing a depth to water table (DWT) data set. The DWT data set, a wetland proxy, is a simulated long-term measure of the point either at or below the surface where climate and geological/topographic water fluxes balance. For species occurrences, the USGS' banding bird data for six relatively common species was used. Distribution models are constructed using Random Forest and MaxEnt. Random Forest classification of habitat and non-habitat provided a measure of DWT variable importance, which indicated that DWT is as important, and often more important, to model accuracy as temperature, precipitation, elevation, and an alternative wetland measure. MaxEnt models that included DWT in addition to traditional predictor variables had a considerable increase in classification accuracy. Also, MaxEnt models created with DWT often had higher accuracy when compared with models created with an alternative measure of wetland habitat. By comparing maps of predicted probability of occurrence and response curves, it is possible to explore how different species respond to water table depth and how a species responds in different seasons. The results of this analysis also illustrate that, as expected, all waterfowl species are tightly affiliated with shallow water table habitat. However, this study illustrates that the intensity of affiliation is not constant between seasons for a species, nor is it consistent between species.

Modeling the magnitude and distribution of sediment-bound pollutants in estuaries is often limited by incomplete knowledge of the site and inadequate sample density. To address these modeling limitations, a decision-support tool framework was conceived that predicts sediment contamination from the sub-estuary to broader estuary extent. For this study, a Random Forest (RF) model was implemented to predict the distribution of a model contaminant, triclosan (5-chloro-2-(2,4-dichlorophenoxy)phenol) (TCS), in Narragansett Bay, Rhode Island, USA. TCS is an unregulated contaminant used in many personal care products. The RF explanatory variables were associated with TCS transport and fate (proxies) and direct and indirect environmental entry. The continuous RF TCS concentration predictions were discretized into three levels of contamination (low, medium, and high) for three different quantile thresholds. The RF model explained 63% of the variance with a minimum number of variables. Total organic carbon (TOC) (transport and fate proxy) was a strong predictor of TCS contamination causing a mean squared error increase of 59% when compared to permutations of randomized values of TOC. Additionally, combined sewer overflow discharge (environmental entry) and sand (transport and fate proxy) were strong predictors. The discretization models identified a TCS area of greatest concern in the northern reach of Narragansett Bay (Providence River sub-estuary), which was validated with independent test samples. This decision-support tool performed well at the sub-estuary extent and provided the means to identify areas of concern and prioritize bay-wide sampling.