Geology topics

Philip Savoy

Publications and source records attributed to Philip Savoy.

11 recordsLinked to original sources

USGS2025

Estimating indicators of cyanobacterial harmful algal blooms in New York State

Cyanobacteria harmful algal blooms (cyanoHABs) are a global concern for aquatic ecosystem and human health. Limited funding for monitoring programs and inconsistent determination of cyanoHAB occurrence present challenges for identifying commonly effective variables for characterizing cyanoHABs and the development of generalized models. We compiled a combination of water quality measurements, lake morphology, climatology, remote sensing data, and observations of cyanoHAB occurrence across New York State and used this dataset to develop two sets of predictive models. The first model predicted chlorophyll a , a common indicator of algal biomass, and assessed the importance of variables for modeled predictions. The most important variables were then used in a second set of models to classify cyanoHAB occurrence. The irradiance attenuation coefficient ( K d ), which was estimated from Secchi depth measurements, and total phosphorus were the two most important variables for predicting chlorophyll a . The second model examined several variables for their ability to classify cyanoHAB occurrence. Predicted cyanoHAB occurrence based on thresholds of chlorophyll a , K d , or total nitrogen all had moderate agreement and were able to correctly classify approximately 70% of observed cyanoHABs. Our analysis indicated that multiple data types were important for predicting chlorophyll a statewide and that simple widely available water quality parameters could classify cyanoHABs occurrence with reasonable accuracy. Identifying variables that can be monitored with increased frequency and decreased latency to detect cyanoHAB occurrence will better inform water managers and provide valuable additional data for further refining predictive models of the likelihood of cyanoHABs occurrence.

New York↗

USGS2024

Bibliography of water-quality studies in Gateway National Recreation Area, New York and New Jersey

The U.S. Geological Survey (USGS) provided technical assistance to the National Park Service (NPS) as part of the USGS-NPS Water-Quality Partnership, by gathering references related to water-quality research conducted in the three units of Gateway National Recreation Area (GATE): Jamaica Bay and Staten Island in New York, and Sandy Hook in New Jersey. As part of this effort, a literature search was performed to compile previous water-quality research conducted within the boundaries of GATE. The resulting bibliography is meant to assist GATE resource managers in understanding the extent of available data and developing plans to close data gaps.

New Jersey, New York↗

USGS2024

Evaluation of sensors for continuous monitoring of harmful algal blooms in the Finger Lakes region, New York, 2019 and 2020

In response to the increasing frequency of cyanobacterial harmful algal blooms (CyanoHABs) in the Finger Lakes region of New York State, a pilot study by the U.S. Geological Survey, in collaboration with the New York State Department of Environmental Conservation, was conducted to enhance CyanoHAB monitoring and understanding. High-frequency sensors were deployed on open water monitoring-station platforms at Seneca Lake in 2019–20, at Owasco Lake in 2019–20, and at Skaneateles Lake in 2019. One of the goals of this study was to evaluate the ability of in-place sensors to make representative measurements of dissolved organic matter, nutrients, and algal pigments (as indicators of phytoplankton biomass) while collecting routine field parameters (water temperature, specific conductance, pH, dissolved oxygen, turbidity, weather, and light) to provide additional information about environmental conditions. Despite challenges like power issues and sensor fouling, the sensors performed well overall. However, correlation analyses between sensor readings and laboratory measurements revealed variable performance. Results indicate the relation between the fluorescent dissolved organic matter sensor and laboratory-measured dissolved organic carbon was weak at all study lakes. The nitrate sensors can be sensitive to ambient temperature and have a substantial power requirement, and the relation between sensor- and laboratory-measured nitrate values differed among lakes. The orthophosphate sensors, which were complex and prone to data loss, yielded results that were difficult to interpret because orthophosphate detections are rare in the study lakes. The multichannel fluorometer was also complex to use and required several unique procedures for its operation. Chlorophyll measurements from the fluorometers correlated moderately well with laboratory-measured chlorophyll- a , although relations with total phytoplankton biovolume were weaker. Relations between phycocyanin concentration measurements from the dual-channel fluorometers and cyanobacterial biovolume were not significant; however, the cyanobacterial biovolume correlation was moderately strong with chlorophyll contribution from cyanobacteria measurements from the multichannel fluorometer. Of all collected parameters, water temperature was among the strongest correlated with chlorophyll- a , total phytoplankton biovolume, and cyanobacterial biovolume. Stepwise regression analysis was used to identify the best parameters for modeling variance in laboratory measures of phytoplankton biomass. This analysis included factors such as chlorophyll fluorescence, pH, water temperature, and others, which varied by lake. Overall, the models had limited explanatory power for chlorophyll- a and other biovolumes, possibly due to the absence of CyanoHABs at the open-water monitoring locations. Multivariate models did not outperform simple fluorescence-based models. Notably, turbidity was a more significant indicator of cyanobacterial biovolume variability than phycocyanin from dual-channel fluorometers. The study concludes that while single and multivariate models based on sensor data are useful, they did not explain any more variance than fluorescence-based models. Broader data collection, including more CyanoHAB events, is necessary to refine these models. Integrating machine learning could leverage large, complex datasets to improve CyanoHAB predictions, thereby enhancing the management and understanding of these blooms.

New York↗

USGS2024

Macroscale controls determine the recovery of river ecosystem productivity following flood disturbances

River ecosystems rely on varied flows, including regular floods, to provide food and habitat for aquatic organisms. However, flows of freshwater are becoming increasingly managed for irrigation, industry, and other human activities, and the frequency of floods is changing. Our study used time-series data of photosynthesis from 143 rivers across the United States and developed a modeling framework to examine how algae, the base of most riverine food webs, recovered following scouring disturbance during floods. We found that algae in wider rivers recovered more quickly following disturbance (e.g., removal from the bottom) but that the flow thresholds at which algae is disturbed are likely more strongly influenced by site-specific characteristics.

Proceedings of the National Academy of Sciences↗

USGS2023

Predicting daily river chlorophyll concentrations at a continental scale

Eutrophication is one of the largest threats to aquatic ecosystems and chlorophyll a measurements are relevant indicators of trophic state and algal abundance. Many studies have modeled chlorophyll a in rivers but model development and testing has largely occurred at individual sites which hampers creating generalized models capable of making broad-scale predictions. To address this gap, we compiled a large data set of chlorophyll a concentrations matched to other water quality, meteorological, and reach characteristic data for a diverse set of 82 streams and rivers across the United States. We used this data set and extreme gradient boosting, a tree-based machine learning algorithm, to predict daily chlorophyll a concentrations. Furthermore, we tested several practical considerations of broad-scale models, such as making predictions at sites not included in model training or the utility of in situ water quality data versus universally available remotely estimated model inputs. Predictions were very strongly correlated to observations when compared against a randomly withheld subset of days; however, the model had lower accuracy when applied to completely novel sites withheld from model training. Turbidity and total nitrogen were the two most important variables for predicting chlorophyll a . Although in situ variables improved modeled estimates and were identified as more important during model interpretation, using only remote inputs still resulted in highly correlated predictions with small bias. Testing a model across many sites allowed for identification of common variables relevant to chlorophyll a and highlighted several challenges for applying data-driven models to new sites or at larger spatial scales.

Water Resources Research↗

USGS2023

High potential but low achievement: Frequent disturbance constrains the light use efficiency of river ecosystems

We rarely consider light limitation in ecosystem productivity, yet light limitation is a major constraint on river autotrophy. Because the light that reaches benthic autotrophs must first pass through terrestrial vegetation and an overlying water column that can be loaded with sediments or colored organic material, there is strong selection for river autotrophs to have high light use efficiencies (LUEs), that is, the efficiency at which light energy is converted to biomass. In contrast to prior studies that have estimated river LUE on single days, we calculated continuous LUE over more than 6 full years for 64 free-flowing rivers across the United States. This dataset represents the largest compilation of continuous estimates of daily rates of gross primary productivity (GPP) and daily light inputs from which we calculated daily estimates of LUE. Early estimates of LUE in rivers found that clearwater springs with stable flows could achieve LUEs of 4%, much higher than LUEs reported for terrestrial plants. We found that 53% of the rivers in our dataset have LUEs that exceed 4% on at least one day of their time series. Because of the high variability in daily LUE, measurements taken on any given day may misrepresent a river ecosystem's annual LUE. Though most rivers share a high potential, the mean annual LUE of all rivers in our dataset is much lower, only 0.5%. We found that rivers with more variable flow regimes had lower annual LUEs, which indicates that LUE is constrained by hydrologic disturbances that remove, bury, or shade autotrophic biomass. Comparisons of LUE across ecosystems allow us to reframe our view of rivers, by recognizing the high efficiency with which they convert light to biomass compared with lentic, marine, and terrestrial ecosystems.

conterminous United States↗

USGS2022

Classifying freshwater salinity regimes in central and western U.S. streams and rivers

Freshwater salinization of rivers is occurring across the globe because of nonpoint source loading of salts from anthropogenic activities such as agriculture, urbanization, and resource extraction that accelerate weathering and release salts. Multidecadal trends in river salinity are well characterized, yet our understanding of annual regimes of salinity in rivers draining diverse central and western U.S. landscapes and their associated catchment attributes is limited. We classified annual salinity regimes in 242 stream locations through dynamic time warping and fuzzy c-medoids clustering of salinity time series. We found two dominant regimes in salinity characterized by an annual summer–fall peak or spring decline . Using random forest regression, we found that precipitation amount, stream slope, and soil salinity were the most important predictors of salinity regime classification. Advancing our understanding of salinity regimes in rivers will improve our ability to predict and mitigate the effects of salinization in freshwater ecosystems through management interventions.

Limnology and Oceanography Letters↗

USGS2022

Light and flow regimes regulate the metabolism of rivers

Mean annual temperature and mean annual precipitation drive much of the variation in productivity across Earth's terrestrial ecosystems but do not explain variation in gross primary productivity (GPP) or ecosystem respiration (ER) in flowing waters. We document substantial variation in the magnitude and seasonality of GPP and ER across 222 US rivers. In contrast to their terrestrial counterparts, most river ecosystems respire far more carbon than they fix and have less pronounced and consistent seasonality in their metabolic rates. We find that variation in annual solar energy inputs and stability of flows are the primary drivers of GPP and ER across rivers. A classification schema based on these drivers advances river science and informs management.

Proceedings of the National Academy of Sciences↗

USGS2021

Predicting light regime controls on primary productivity across CONUS river networks

Solar radiation is a fundamental driver of ecosystem productivity, but widespread estimates of light available for primary producers in rivers are lacking. We developed a model to predict light available for river primary producers and used it to estimate river primary production across the contiguous United States (CONUS). Successively accounting for riparian and water column processes improved predictions of primary production as a function of light. We calculated the ratio of river width to riparian tree height and used this metric to predict whether riparian zones or water column processes most limit productivity for over 2 million reaches. Water column processes limited productivity for 50% of the nation's river length and 80% of its surface area, with variations across ecoregions related to riparian forest cover. Our findings facilitate large-scale predictions of stream and river ecosystem productivity, as well as understanding the processes controlling productivity across networks.

Geophysical Research Letters↗

USGS2021

A seasonally dynamic model of light at the stream surface

Light is a primary constraint on primary production and drives many ecological processes in stream ecosystems, yet light regimes have received considerably less attention than other factors of the stream environment, such as hydrology or nutrient cycling. Light received by streams can be highly heterogeneous in both space and time resulting from changes in topography, channel characteristics, and riparian vegetation. Both the structure and phenology of riparian vegetation can be important determinants of the seasonality and magnitude of light reaching the stream surface, particularly in smaller forested streams. Despite the importance of riparian phenology on temporal patterns of stream light availability, existing models do not account for the seasonal dynamics of canopies. We developed a dynamic, biophysically based model ( StreamLight ) that incorporates canopy structure and phenology to predict light reaching the stream surface. We compared StreamLight to an existing model at 21 sites across the USA and found that, across sites, our biophysically based model produced light estimates that were more strongly correlated to observations and reduced the magnitude of errors in comparison to the existing model, particularly for streams that were relatively narrow compared to the height of riparian vegetation. Because smaller streams represent most global stream length, we expect that, in many smaller forested streams, the inclusion of canopy structure and phenology will enhance our ability to predict light regimes. We also used model simulations to examine the importance of controls on stream light environments and found that channel width was the strongest control on light environments. StreamLight represents an important incremental step forward in developing mechanistic models of river network productivity and in linking shifts in terrestrial vegetation structure and phenology to aquatic ecosystem productivity and thermal regimes.

Freshwater Science↗

USGS2019

Metabolic rhythms in flowing waters: An approach for classifying river productivity regimes

Although seasonal patterns of ecosystem productivity have been extensively described and analyzed with respect to their primary forcings in terrestrial and marine systems, comparatively little is known about these same processes in rivers. However, it is now possible to perform a large‐scale synthesis on the patterns and drivers of river productivity regimes because of the recent sensor advances allowing for near‐continuous estimates of river productivity. Here, we explore a dataset of 47 U.S. rivers to examine whether there are characteristic river productivity regimes. We use classification approaches to develop a typology of productivity regimes and then use these regimes to examine differences with respect to potential controls of productivity. We identified two distinct metabolic regimes, which we named Summer Peak and Spring Peak Rivers, within our dataset. These regimes meaningfully differed in both the timing and magnitude of productivity and were robust to different approaches to classification. We also found that several variables, including watershed area and characteristics of water temperature or discharge, were able to predict the class membership of these regimes with modest accuracy. Our results support the presence of characteristic metabolic regimes and suggests that these regimes may have common sets of environmental controls. We present classification as one approach to begin exploring the productivity regimes of rivers. The strength of our approach is that it fully leverages these newly available high‐frequency productivity estimates to create classes that can be used to draw inferences about how the controls of river productivity differ between or within systems.

Limnology and Oceanography↗