Ecosystem effects of a tropical cyclone on a network of lakes in northeastern North America

Trophic state resilience to hurricane disturbance of Lake Yojoa, Honduras

Cyclones are a poorly described disturbance in tropical lakes, with the potential to alter ecosystems and compromise the services they provide. In November 2020, Hurricanes Eta and Iota made landfall near the Nicaragua-Honduras border, inundating the region with a large amount of late-season precipitation. To understand the impact of these storms on Lake Yojoa, Honduras, we compared 2020 and 2021 conditions using continuous (every 16 days) data collected from five pelagic locations. The storms resulted in increased Secchi depth and decreased algal abundance in December 2020, and January and February 2021, and lower-than-average accumulation of hypolimnetic nutrients from the onset of stratification (April 2021) until mixus in November 2021. Despite the reduced hypolimnetic nutrient concentrations, epilimnetic nutrient concentrations returned to (and in some cases exceeded) pre-hurricane levels following annual water column turnover in 2021. This response suggests that Lake Yojoa's trophic state had only an ephemeral response to the disturbance imposed by the two hurricanes, likely due to internal input of sediment derived nutrients. These aseasonal storms acted as a large-scale experiment that resulted in nutrient dilution and demonstrated the resilience of Lake Yojoa's trophic state to temporary nutrient reductions.

Modeling the sensitivity of cyanobacteria blooms to plausible changes in precipitation and air temperature variability

Many recent studies have attributed the observed variability of cyanobacteria blooms to meteorological drivers and have projected blooms with worsening societal and ecological impacts under future climate scenarios. Nonetheless, few studies have jointly examined their sensitivity to projected changes in both precipitation and temperature variability. Using an Integrated Assessment Model (IAM) of Lake Champlain's eutrophic Missisquoi Bay, we demonstrate a factorial design approach for evaluating the sensitivity of concentrations of chlorophyll a (chl-a), a cyanobacteria surrogate, to global climate model-informed changes in the central tendency and variability of daily precipitation and air temperature. An Analysis of Variance (ANOVA) and multivariate contour plots highlight synergistic effects of these climatic changes on exceedances of the World Health Organization's moderate 50 μg/L concentration threshold for recreational contact. Although increased precipitation produces greater riverine total phosphorus loads, warmer and drier scenarios produce the most severe blooms due to the greater mobilization and cyanobacteria uptake of legacy phosphorus under these conditions. Increases in daily precipitation variability aggravate blooms most under warmer and wetter scenarios. Greater temperature variability raises exceedances under current air temperatures but reduces them under more severe warming when water temperatures exceed optimal values for cyanobacteria growth more often. Our experiments, controlled for wind-induced changes to lake water quality, signal the importance of larger summer runoff events for curtailing bloom growth through reductions of water temperature, sunlight penetration and stratification. Finally, the importance of sequences of wet and dry periods in generating cyanobacteria blooms motivates future research on bloom responses to changes in interannual climate persistence.

Bacterioplankton Zonation Does Exist in High Elevation, Polymictic Lakes

The assessment of distribution patterns or zonation of planktonic microbes along the water column is a crucial step to interpret their function in the ecosystem. In lakes without seasonal thermal stratification or polymictic systems such as high elevation tropical lakes, planktonic bacterial taxa are probably homogeneously distributed in the water column in contrast to what is known for thermally stratified lakes. However, we know little about bacterial distribution patterns in polymictic lakes and their relation to environmental gradients other than temperature. Here we assessed the diversity, microdiversity, and bacterial community composition at different discrete depths in three high elevation lakes (4,400-4,550 m above sea level) from the Andean plateau to test whether bacterial zonation patterns exist along the water column. For this objective, we analyzed bulk DNA and the putatively active fraction (cDNA) of the 16S rRNA gene. Although a clear gradient of temperature and oxygen was not detected along the water column, a significant vertical spatial zonation of the bacterial communities was present in two out of the three lakes, with microdiversity contributing to such pattern. Our results provide a reference for understanding how changing environmental conditions could affect high elevation aquatic ecosystems, particularly when warming is amplified with elevation, accelerating changes in hydrological regimes and biodiversity. Finally, our results highlight the importance of incorporating the whole water column in ecological studies of aquatic ecosystems lacking temporal or permanent thermal stratification.

The impacts of the hydraulic retention effect and typhoon disturbance on the carbon flux in shallow subtropical mountain lakes

A typhoon is extreme weather that flushes terrestrial carbon (C) loads and temporally mixes the entire water columns of lakes in subtropical regions. A C flux varies based on the trophic level associated with the ecological cycle related to hydraulic retention time (residence time). Herein, we sought to clarify how the hydraulic retention time and the disturbance from a typhoon affect the C flux regimes in two subtropical mountain lakes in a humid region of Taiwan with different trophic levels-oligotrophic and mesotrophic. We investigated the meteorological data and vertical profiles of the water temperature, dissolved inorganic carbon (DIC), dissolved organic C (DOC), and chlorophyll a (Chl. a) during the pre-typhoon period (April-July), during the typhoon period (August-November), and the post-typhoon period (December-March) for five years (2009-2010 and 2015-2017). We applied a three-dimensional environmental model (Fantom) to investigate the hydraulic retention effect on the net ecosystem production (NEP) using the residence time in stratified lakes. The results demonstrate that typhoon-induced mixing associated with the hydraulic retention effect plays one of the critical roles in controlling the NEP and C flux in shallow subtropical lakes.

OUP accepted manuscript

Extreme events have increased in frequency globally, with a simultaneous surge in scientific interest about their ecological responses, particularly in sensitive freshwater, coastal, and marine ecosystems. We synthesized observational studies of extreme events in these aquatic ecosystems, finding that many studies do not use consistent definitions of extreme events. Furthermore, many studies do not capture ecological responses across the full spatial scale of the events. In contrast, sampling often extends across longer temporal scales than the event itself, highlighting the usefulness of long-term monitoring. Many ecological studies of extreme events measure biological responses but exclude chemical and physical responses, underscoring the need for integrative and multidisciplinary approaches. To advance extreme event research, we suggest prioritizing pre- and postevent data collection, including leveraging long-term monitoring; making intersite and cross-scale comparisons; adopting novel empirical and statistical approaches; and developing funding streams to support flexible and responsive data collection.

Hydrologic Setting Dictates the Sensitivity of Ecosystem Metabolism to Climate Variability in Lakes

Global change is influencing production and respiration in ecosystems across the globe. Lakes in particular are changing in response to climatic variability and cultural eutrophication, resulting in changes in ecosystem metabolism. Although the primary drivers of production and respiration such as the availability of nutrients, light, and carbon are well known, heterogeneity in hydrologic setting (for example, hydrological connectivity, morphometry, and residence) across and within regions may lead to highly variable responses to the same drivers of change, complicating our efforts to predict these responses. We explored how differences in hydrologic setting among lakes influenced spatial and inter annual variability in ecosystem metabolism, using high-frequency oxygen sensor data from 11 lakes over 8 years. Trends in mean metabolic rates of lakes generally followed gradients of nutrient and carbon concentrations, which were lowest in seepage lakes, followed by drainage lakes, and higher in bog lakes. We found that while ecosystem respiration (ER) was consistently higher in wet years in all hydrologic settings, gross primary production (GPP) only increased in tandem in drainage lakes. However, interannual rates of ER and GPP were relatively stable in drainage lakes, in contrast to seepage and bog lakes which had coefficients of variation in metabolism between 22–32%. We explored how the geospatial context of lakes, including hydrologic residence time, watershed area to lake area, and landscape position influenced the sensitivity of individual lake responses to climatic variation. We propose a conceptual framework to help steer future investigations of how hydrologic setting mediates the response of metabolism to climatic variability.

Effect of Extreme Climate Events on Lake Ecosystems

The Earth is facing a major change in climate due to ongoing global warming [...]

Effects of Consecutive Extreme Weather Events on a Temperate Dystrophic Lake: A Detailed Insight into Physical, Chemical and Biological Responses

Between May and July 2018, Ireland experienced an exceptional heat wave, which broke long-term temperature and drought records. These calm, stable conditions were abruptly interrupted by a second extreme weather event, Atlantic Storm Hector, in late June. Using high-frequency monitoring data, coupled with fortnightly biological sampling, we show that the storm directly affected the stratification pattern of Lough Feeagh, resulting in an intense mixing event. The lake restabilised quickly after the storm as the heatwave continued. During the storm there was a three-fold reduction in Schmidt stability, with a mixed layer deepening of 9.5 m coinciding with a two-fold reduction in chlorophyll a but a three-fold increase in total zooplankton biomass. Epilimnetic respiration increased and net ecosystem productivity decreased. The ratio of total nitrogen:total phosphorus from in-lake versus inflow rivers was decoupled, leading to a cascade effect on higher trophic levels. A step change in nitrogen:phosphorus imbalances suggested that the zooplankton community shifted from phosphorus to nitrogen nutrient constraints. Such characterisations of both lake thermal and ecological responses to extreme weather events are relatively rare but are crucial to our understanding of how lakes are changing as the impacts of global climate change accelerate.

A Research Framework to Integrate Cross-Ecosystem Responses to Tropical Cyclones

Tropical cyclones play an increasingly important role in shaping ecosystems. Understanding and generalizing their responses is challenging because of meteorological variability among storms and its interaction with ecosystems. We present a research framework designed to compare tropical cyclone effects within and across ecosystems that: a) uses a disaggregating approach that measures the responses of individual ecosystem components, b) links the response of ecosystem components at fine temporal scales to meteorology and antecedent conditions, and c) examines responses of ecosystem using a resistance–resilience perspective by quantifying the magnitude of change and recovery time. We demonstrate the utility of the framework using three examples of ecosystem response: gross primary productivity, stream biogeochemical export, and organismal abundances. Finally, we present the case for a network of sentinel sites with consistent monitoring to measure and compare ecosystem responses to cyclones across the United States, which could help improve coastal ecosystem resilience.

Storm impacts on phytoplankton community dynamics in lakes

In many regions across the globe, extreme weather events such as storms have increased in frequency, intensity, and duration due to climate change. Ecological theory predicts that such extreme events should have large impacts on ecosystem structure and function. High winds and precipitation associated with storms can affect lakes via short-term runoff events from watersheds and physical mixing of the water column. In addition, lakes connected to rivers and streams will also experience flushing due to high flow rates. Although we have a well-developed understanding of how wind and precipitation events can alter lake physical processes and some aspects of biogeochemical cycling, our mechanistic understanding of the emergent responses of phytoplankton communities is poor. Here we provide a comprehensive synthesis that identifies how storms interact with lake and watershed attributes and their antecedent conditions to generate changes in lake physical and chemical environments. Such changes can restructure phytoplankton communities and their dynamics, as well as result in altered ecological function (e.g., carbon, nutrient and energy cycling) in the short- and long-term. We summarize the current understanding of storm-induced phytoplankton dynamics, identify knowledge gaps with a systematic review of the literature, and suggest future research directions across a gradient of lake types and environmental conditions.

Modeling response of water quality parameters to land-use and climate change in a temperate, mesotrophic lake

Lake Auburn, Maine, USA, is a historically unproductive lake that has experienced multiple algal blooms since 2011. The lake is the water supply source for a population of ~60,000. We modeled past temperature, and concentrations of dissolved oxygen (DO) and phosphorus (P) in Lake Auburn by considering the catchment and internal contributions of P as well as atmospheric factors, and predicted the change in lake water quality in response to future climate and land-use changes. A stream hydrology and P-loading model (SimplyP) was used to generate input from two major tributaries into a lake model (MyLake-Sediment) to simulate physical mixing, chemical dynamics, and sediment geochemistry in Lake Auburn from 2013 to 2017. Simulations of future lake water quality were conducted using meteorological boundary conditions derived from recent historical data and climate model projections for high greenhouse-gas emission cases. The effects of future land development on lake water quality for the 2046 to 2055 time period under different land-use and climate change scenarios were also simulated. Our results indicate that lake P enrichment is more responsive to extreme storm events than increasing air temperatures, mean precipitation, or windstorms; loss of fish habitat is driven by windstorms, and to a lesser extent an increasing water temperature; and catchment development further leads to water quality decline. All simulations also show that the lake is susceptible to both internal and external P loadings. Simulation of temperature, DO, and P proved to be an effective means for predicting the loss of water quality under changing land-use and climate scenarios.

Response of prokaryotic communities to extreme precipitation events in an urban coastal lagoon: A case study of Yundang lagoon, China

Increasing extreme precipitation events (EPEs) can induce biogeochemical disturbances in the coastal lagoon ecosystems. Very little is known about the response of prokaryotic communities to such influences, which are the key components mediating the biogeochemical cycling in lagoons. Here 16S rRNA gene amplicon sequencing and high-through quantitative PCR (HT-qPCR) were employed to investigate the distribution of prokaryotic communities and fecal indicator genes in the surface waters of Yundang lagoon, Xiamen, China during EPEs, respectively. Prokaryotic communities from rainwaters, influents (IFs) and effluents (EFs) from a nearby wastewater treatment plant were also characterised. The results indicated a significant variation in the composition of lagoon prokaryotic communities compared with rainwaters, IFs and EFs. Multivariate and phylogenetic signal analyses revealed that environmental filtering, mainly controlled by salinity, was the major ecological process responsible for the temporal succession of lagoon prokaryotic communities during EPEs. Moreover, the pollution indicator taxa (based on amplicon sequencing) and fecal indicator genes (based on HT-qPCR) demonstrated that EPEs may induce sewage overflows and fecal pollution (mainly from humans and dogs), resulting in an increase in the relative abundance of pollution indicator taxa and genes in Yundang lagoon. Network analysis illustrated that the number of network edges and keystone species decreased along the sampling times, implying that EPEs-induced disturbances may affect prokaryotic species associations. Taken together, this study provides an enhanced understanding of the responses of lagoon prokaryotic communities to EPEs-induced disturbances.

13 Years of Storms: An Analysis of the Effects of Storms on Lake Physics on the Atlantic Fringe of Europe

While winter storms are generally common in western Europe, the rarer summer storms may result in more pronounced impacts on lake physics. Using long-term, high frequency datasets of weather and lake thermal structure from the west of Ireland (2005 to 2017), we quantified the effects of storms on the physical conditions in a monomictic, deep lake close to the Atlantic Ocean. We analysed a total of 227 storms during the stratified (May to September, n = 51) and non-stratified (November to March, n = 176) periods. In winter, as might be expected, changes were distributed over the entire water column, whereas in summer, when the lake was stratified, storms only impacted the smaller volume above the thermocline. During an average summer (May–September) storm, the lake number dropped by an order of magnitude, the thermocline deepened by an average of 2.8 m, water column stability decreased by an average of 60.4 j m−2 and the epilimnion temperature decreased by a factor of five compared to the average change in winter (0.5 °C vs. 0.1 °C). Projected increases in summer storm frequency will have important implications for lake physics and biological pathways.

Geographic patterns of the climate sensitivity of lakes

Climate change is a well-recognized threat to lake ecosystems and, although there likely exists geographic variation in the sensitivity of lakes to climate, broad-scale, long-term studies are needed to understand this variation. Further, the potential mediating role of local to regional ecological context on these responses is not well documented. In this study, we examined relationships between climate and water clarity in 365 lakes from 1981 to 2010 in two distinct regions in the northeastern and midwestern United States. We asked (1) How do climate-water-clarity relationships vary across watersheds and between two geographic regions? and (2) Do certain characteristics make some lakes more climate sensitive than others? We found strong differences in climate-water-clarity relationships both within and across the two regions. For example, in the northeastern region, water clarity was often negatively correlated with summer precipitation (median correlation = -0.32, n = 160 lakes), but was not correlated with summer average maximum temperature (median correlation = 0.09, n = 205 lakes). In the midwestern region, water clarity was not related to summer precipitation (median correlation = -0.04), but was often negatively correlated with summer average maximum temperature (median correlation = -0.18). There were few strong relationships between local and sub-regional ecological context and a lake's sensitivity to climate. For example, ecological context variables explained just 16-18% of variation in summer precipitation sensitivity, which was most related to total phosphorus, chlorophyll a, lake depth, and hydrology in both regions. Sensitivity to summer maximum temperature was even less predictable in both regions, with 4% or less of variation explained using all ecological context variables. Overall, we identified differences in the climate sensitivity of lakes across regions and found that local and sub-regional ecological context weakly influences the sensitivity of lakes to climate. Our findings suggest that local to regional drivers may combine to influence the sensitivity of lake ecosystems to climate change, and that sensitivities among lakes are highly variable within and across regions. This variability suggests that lakes are sensitive to different aspects of climate change (temperature vs. precipitation) and that responses of lakes to climate are heterogeneous and complex.

Physical and chemical impacts of a major storm on a temperate lake: a taste of things to come?

Extreme weather can have a substantial influence on lakes and is expected to become more frequent with climate change. We explored the influence of one particular extreme event, Storm Ophelia, on the physical and chemical environment of England's largest lake, Windermere. We found that the substantial influence of Ophelia on meteorological conditions at Windermere, in particular wind speed, resulted in a 25-fold increase (relative to the study-period average) in the wind energy flux at the lake-air interface. Following Ophelia, there was a short-lived mixing event in which the Schmidt stability decreased by over 100 Jm^-2 and the thermocline deepened by over 10 m during a 12-h period. As a result of changes to the strength of stratification, Ophelia also changed the internal seiche regime of Windermere with the dominant seiche period increasing from ~ 17 h pre-storm to ~ 21 h post-storm. Following Ophelia, there was an upwelling of cold and low-oxygenated waters at the southern-end of the lake. This had a substantial influence on the main outflow of Windermere, the River Leven, where dissolved oxygen concentrations decreased by ~ 48%, from 9.3 to 4.8 mg L^-1, while at the mid-lake monitoring station in Windermere, it decreased by only ~ 3%. This study illustrates that the response of a lake to extreme weather can cause important effects downstream, the influence of which may not be evident at the lake surface. To understand the impact of future extreme events fully, the whole lake and downstream-river system need to be studied together.

Storm impacts on alpine lakes: Antecedent weather conditions matter more than the event intensity

Extreme weather events may be just as important as gradual trends for the long-term trajectories of ecosystems. For alpine lakes, which are exposed to both exacerbated atmospheric warming and intense episodic weather events, future conditions might not be appropriately forecast by only climate change trends, i.e. warming, if extreme events have the potential to deflect their thermal and metabolic states from their seasonal ranges. We used high-frequency monitoring data over three open-water seasons with a one-dimensional hydrodynamic model of the high-altitude Lake Muzelle (France) to show that rainstorms or windstorms, notwithstanding their intensity, did not trigger long-lasting consequences to the lake characteristics when light penetration into the lake was not modified. In contrast, storms associated with high turbidity input from the watershed ("turbid storms") strongly modified the lacustrine hydrodynamics and metabolism for the rest of the open-water season through reduced light penetration. The long-lasting effects of turbid storms were related to the inputs and in-lake persistence of very light glacial suspensoids from the watershed. The occurrence of the observed turbid storms was not related to the wind or rain intensities during the events. Instead, the turbid storms occurred after dry and atypically warm spells, i.e. meteorological conditions expected to be more frequent in this alpine region in the upcoming decades. Consequently, storm events, notwithstanding their intensity, are expected to strongly imprint the future ecological status of alpine lakes under climate warming.

Drivers of Ecosystem Metabolism in Two Managed Shallow Lakes with Different Salinity and Trophic Conditions: The Sauce Grande and La Salada Lakes (Argentina)

Understanding the drivers and how they affect ecosystem metabolism is essential for developing effective management policy and plans. In this study, net ecosystem production (NEP), ecosystem respiration (R), and gross primary production (GPP) rates were estimated in relation to physicochemical, hydrological, and meteorological variables in La Salada (LS) and Sauce Grande (SG), two shallow lakes located in an important agricultural region with water management. LS is a mesosaline, mesotrophic-eutrophic lake, whereas SG is a hyposaline and eutrophic lake. GPP and R showed daily and seasonal variations, with R exceeding GPP during most of the study period in both lakes. Net heterotrophic conditions prevailed during the study period (NEP LS: −1.1 mmol O2 m−2 day−1 and NEP SG: −1.25 mmol O2 m−2 day−1). From data analysis, the temperature, wind speed, and lake volume are the main drivers of ecosystem metabolism for both lakes. Despite the significant differences between the two lakes, the NEP values were similar. The different hydrological characteristics (endorheic vs. flushing lake) were crucial in explaining why the two different systems presented similar ecosystem metabolic rates, emphasizing the importance of water management.

Thermocline deepening boosts ecosystem metabolism: evidence from a large-scale lake enclosure experiment simulating a summer storm

Extreme weather events can pervasively influence ecosystems. Observations in lakes indicate that severe storms in particular can have pronounced ecosystem-scale consequences, but the underlying mechanisms have not been rigorously assessed in experiments. One major effect of storms on lakes is the redistribution of mineral resources and plankton communities as a result of abrupt thermocline deepening. We aimed at elucidating the importance of this effect by mimicking in replicated large enclosures (each 9 m in diameter, ca. 20 m deep, ca. 1300 m³ in volume) a mixing event caused by a severe natural storm that was previously observed in a deep clear-water lake. Metabolic rates were derived from diel changes in vertical profiles of dissolved oxygen concentrations using a Bayesian modelling approach, based on high-frequency measurements. Experimental thermocline deepening stimulated daily gross primary production (GPP) in surface waters by an average of 63% for >4 weeks even though thermal stratification re-established within 5 days. Ecosystem respiration (ER) was tightly coupled to GPP, exceeding that in control enclosures by 53% over the same period. As GPP responded more strongly than ER, net ecosystem productivity (NEP) of the entire water column was also increased. These protracted increases in ecosystem metabolism and autotrophy were driven by a proliferation of inedible filamentous cyanobacteria released from light and nutrient limitation after they were entrained from below the thermocline into the surface water. Thus, thermocline deepening by a single severe storm can induce prolonged responses of lake ecosystem metabolism independent of other storm-induced effects, such as inputs of terrestrial materials by increased catchment run-off. This highlights that future shifts in frequency, severity or timing of storms are an important component of climate change, whose impacts on lake thermal structure will superimpose upon climate trends to influence algal dynamics and organic matter cycling in clear-water lakes.

Automatic High Frequency Monitoring for Improved Lake and Reservoir Management

Recent technological developments have increased the number of variables being monitored in lakes and reservoirs using automatic high frequency monitoring (AHFM). However, design of AHFM systems and posterior data handling and interpretation are currently being developed on a site-by-site and issue-by-issue basis with minimal standardization of protocols or knowledge sharing. As a result, many deployments become short-lived or underutilized, and many new scientific developments that are potentially useful for water management and environmental legislation remain underexplored. This Critical Review bridges scientific uses of AHFM with their applications by providing an overview of the current AHFM capabilities, together with examples of successful applications. We review the use of AHFM for maximizing the provision of ecosystem services supplied by lakes and reservoirs (consumptive and non consumptive uses, food production, and recreation), and for reporting lake status in the EU Water Framework Directive. We also highlight critical issues to enhance the application of AHFM, and suggest the establishment of appropriate networks to facilitate knowledge sharing and technological transfer between potential users. Finally, we give advice on how modern sensor technology can successfully be applied on a larger scale to the management of lakes and reservoirs and maximize the ecosystem services they provide.

Dynamics of dissolved organic matter during four storm events in two forest streams: source, export, and implications for harmful disinfection byproduct formation

Dynamics of river dissolved organic matter (DOM) during storm events have profound influences on the downstream aquatic ecosystem and drinking water safety. This study investigated temporal variations in DOM during four storm events in two forest headwater streams (the EH and JH brooks, South Korea) and the impacts on the disinfection byproducts (DBPs) formation potential. The within-event variations of most DOM quantity parameters were similar to the flow rate in the EH but not in the larger JH brook. The dissolved organic carbon (DOC) showed clockwise and counterclockwise hysteresis with the flow rate in the EH and JH brooks, respectively, indicating the importance of both flow path and DOM source pool size in determining the effects of storm events. The stream DOM became less aromatic/humified from the first to the last event in both brooks, probably due to the increasing fresh plant pool and the decreasing leaf litter pool during the course of rainy season. The DOC export during each event increased 1.3-2.7- and 1.1-7.0-fold by stormflows in the EH and JH brooks, respectively. The leaf litter and soil together was the major DOM source, particularly during early events. The enhanced DOM export probably increases the risks of DBPs formation in disinfection, as indicated by a strong correlation observed between DOC and trihalomethanes formation potential (THMFP). High correlations between two humic-like fluorescent components and THMFP further suggested the potential of assessing THMFP with in situ fluorescence sensors during storms.

Predicting trihalomethanes in the new york city water supply

Chlorine, a commonly used disinfectant in most water supply systems, can combine with organic carbon to form disinfectant byproducts, including carcinogenic trihalomethanes. We used water quality data from 24 monitoring sites within the New York City water supply distribution system, measured between January 2009 and April 2012, to develop an empirical model for predicting total trihalomethane (TTHM) levels. Terms in the model included the following water quality parameters: total organic carbon, pH, water age (reaction time), and water temperature. Reasonable estimates of TTHM levels were achieved with overall of about 0.75, and predicted values on average were within 6 μg L of measured values. A sensitivity analysis indicated that total organic carbon and water age are the most important factors for TTHM formation, followed by water temperature; pH was the least important factor within the boundary conditions of observed water quality. Although never out of compliance in 2011, the TTHM levels in the water supply increased after tropical storms Irene and Lee, with 45% of the samples exceeding the 80 μg L maximum contaminant level in October and November. This increase was explained by changes in water quality parameters, particularly by the increase in total organic carbon concentration during this period. This study demonstrates the use of an empirical model to understand TTHM formative factors and their relative importance in a drinking water supply. This has implications for simulating management scenarios and real-time estimation of TTHMs in water supply systems under changing environmental conditions.