Prediction of stream nitrogen and phosphorus concentrations from high-frequency sensors using Random Forests Regression.

Stream nutrient concentrations exhibit marked temporal variation due to hydrology and other factors such as the seasonality of biological processes. Many water quality monitoring programs sample too infrequently (i.e., weekly or monthly) to fully characterize lotic nutrient conditions and to accurately estimate nutrient loadings. A popular solution to this problem is the surrogate-regression approach, a method by which nutrient concentrations are estimated from related parameters (e.g., conductivity or turbidity) that can easily be measured in situ at high frequency using sensors. However, stream water quality data often exhibit skewed distributions, nonlinear relationships, and multicollinearity, all of which can be problematic for linear-regression models. Here, we use a flexible and robust machine learning technique, Random Forests Regression (RFR), to estimate stream nitrogen (N) and phosphorus (P) concentrations from sensor data within a forested, mountainous drainage area in upstate New York. When compared to actual nutrient data from samples tested in the laboratory, this approach explained much of the variation in nitrate (89%), total N (85%), particulate P (76%), and total P (74%). The models were less accurate for total soluble P (47%) and soluble reactive P (32%), though concentrations of these latter parameters were in a relatively low range. Although soil moisture and fluorescent dissolved organic matter are not commonly used as surrogates in nutrient-regression models, they were important predictors in this study. We conclude that RFR shows great promise as a tool for modeling instantaneous stream nutrient concentrations from high-frequency sensor data, and encourage others to evaluate this approach for supplementing traditional (laboratory-determined) nutrient datasets.

Long-term dataset on aquatic responses to concurrent climate change and recovery from acidification.

Concurrent regional and global environmental changes are affecting freshwater ecosystems. Decadal-scale data on lake ecosystems that can describe processes affected by these changes are important as multiple stressors often interact to alter the trajectory of key ecological phenomena in complex ways. Due to the practical challenges associated with long-term data collections, the majority of existing long-term data sets focus on only a small number of lakes or few response variables. Here we present physical, chemical, and biological data from 28 lakes in the Adirondack Mountains of northern New York State. These data span the period from 1994-2012 and harmonize multiple open and as-yet unpublished data sources. The dataset creation is reproducible and transparent; R code and all original files used to create the dataset are provided in an appendix. This dataset will be useful for examining ecological change in lakes undergoing multiple stressors.

Brooktrout Lake case study: biotic recovery from acid deposition 20 years after the 1990 Clean Air Act Amendments.

The Adirondack Mountain region is an extensive geographic area (26,305 km(2)) in upstate New York where acid deposition has negatively affected water resources for decades and caused the extirpation of local fish populations. The water quality decline and loss of an established brook trout (Salvelinus fontinalis [Mitchill]) population in Brooktrout Lake were reconstructed from historical information dating back to the late 1880s. Water quality and biotic recovery were documented in Brooktrout Lake in response to reductions of S deposition during the 1980s, 1990s, and 2000s and provided a unique scientific opportunity to re-introduce fish in 2005 and examine their critical role in the recovery of food webs affected by acid deposition. Using C and N isotope analysis of fish collagen and state hatchery feed as well as Bayesian assignment tests of microsatellite genotypes, we document in situ brook trout reproduction, which is the initial phase in the restoration of a preacidification food web structure in Brooktrout Lake. Combined with sulfur dioxide emissions reductions promulgated by the 1990 Clean Air Act Amendments, our results suggest that other acid-affected Adirondack waters could benefit from careful fish re-introduction protocols to initiate the ecosystem reconstruction of important components of food web dimensionality and functionality.

Acidification in the Adirondacks: defining the biota in trophic levels of 30 chemically diverse acid-impacted lakes.

The Adirondack Mountains in New York State have a varied surficial geology and chemically diverse surface waters that are among the most impacted by acid deposition in the U.S. No single Adirondack investigation has been comprehensive in defining the effects of acidification on species diversity, from bacteria through fish, essential for understanding the full impact of acidification on biota. Baseline midsummer chemistry and community composition are presented for a group of chemically diverse Adirondack lakes. Species richness of all trophic levels except bacteria is significantly correlated with lake acid-base chemistry. The loss of taxa observed per unit pH was similar: bacterial genera (2.50), bacterial classes (1.43), phytoplankton (3.97), rotifers (3.56), crustaceans (1.75), macrophytes (3.96), and fish (3.72). Specific pH criteria were applied to the communities to define and identify acid-tolerant (pH<5.0), acid-resistant (pH 5.0-5.6), and acid-sensitive (pH>5.6) species which could serve as indicators. Acid-tolerant and acid-sensitive categories are at end-points along the pH scale, significantly different at P<0.05; the acid-resistant category is the range of pH between these end-points, where community changes continually occur as the ecosystem moves in one direction or another. The biota acid tolerance classification (batc) system described herein provides a clear distinction between the taxonomic groups identified in these subcategories and can be used to evaluate the impact of acid deposition on different trophic levels of biological communities.