WATER QUALITY MODELING IN LAKE MENDOTA: DO THE BACTERIA MATTER

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: HATCH
• Reporting Frequency: Annual
• Accession No.: 0222656
• Project Start Date: Oct 1, 2010
• Project End Date: Sep 30, 2014
• Program Code: [(N/A)]- (N/A)

Recipient Organization
UNIV OF WISCONSIN
21 N PARK ST STE 6401
MADISON,WI 53715-1218

Performing Department
Bacteriology

Non Technical Summary
Eutrophication is a persistent environmental problem that impacts aquatic ecosystems worldwide. Human activities have mobilized large quantities of nutrients that are profoundly impacting water bodies ranging from small suburban ponds to the Great Lakes and from estuaries to the Gulf of Mexico. Lakes in southern Wisconsin have been struggling with this problem since the area was settled in the mid 1800's. Lake Mendota is one of most studied lakes in the world, as well as one of the now most heavily managed in the region. It is considered to be a prime example of anthropogenic eutrophication. Lake water quality is a highly visible topic of public interest, since Madison area residents consider their lakes to be a defining feature of the urban landscape. Furthermore, water quality impairment due to eutrophication has significant local economic impacts. Nonpoint source nutrient pollution (especially phosphorus) is considered to be the driving force behind accelerated eutrophication of most surface waters; the Madison area lakes are no exception. These sources include runoff transported from animal feed lots (manure), agricultural soil (fertilizer), and urban landscapes. In the lakes, microbes control the processing of nutrients, directly influencing phytoplankton (i.e. algae and cyanobacteria) growth and the resulting eutrophication. Microbial ecologists and limnologists have made great progress in understanding the fundamental mechanisms of nutrient cycling and phytoplankton growth limitation. However, the high level of ecosystem complexity demands a systems-level approach using sophisticated mechanistic models that can capture the most important physical, chemical, and biological processes occurring within the lake. Computational water quality models hold great promise as both advanced research and management tools for complex lake ecosystems. These models are able to simulate the three-dimensional fluid motion that characterizes thermally stratified lakes such as Lake Mendota, and explicitly capture important variables such as chlorophyll concentrations, nutrient concentrations, zooplankton abundance, and dissolved oxygen concentration. However, even though bacteria and phytoplankton play a critical role in mediating water quality, few models include them as discrete entities. Much could be learned by "splitting the black boxes" that modelers use to lump bacteria and phytoplankton. This research aims to determine if splitting the boxes leads to more accurate model predictions, with an emphasis on cyanobacterial biomass and cyanotoxin production as key variables to be simulated. The overarching goal of this proposed work is to improve our ability to use mechanistic water quality models to simulate complex ecosystem dynamics. We seek to develop predictive models that can be used to forecast cyanobacterial bloom formation, empower water quality managers to make regulatory decisions about nutrient loadings, and enable public health officials to issue warnings when toxin-producing cyanobacteria are expected to be present. Our work will provide the fundamental science that will be a foundation for further work towards this long-term goal.

Animal Health Component

(N/A)

Research Effort Categories

Basic

(N/A)

Applied

(N/A)

Developmental

(N/A)

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
112	0399	1190	40%
112	0599	1190	30%
112	4010	1100	30%

Knowledge Area
112 - Watershed Protection and Management;

Subject Of Investigation
0599 - Recreational resources, general/other; 4010 - Bacteria; 0399 - Watersheds and river basins, general;

Field Of Science
1100 - Bacteriology; 1190 - Limnology;

Keywords

water quality modeling, cyanobacteria, microbial ecology, community

dynamics

Goals / Objectives
The overarching goal of this proposed work is to improve our ability to use mechanistic water quality models to simulate complex ecosystem dynamics. We seek to develop predictive models that can be used to forecast cyanobacterial bloom formation, empower water quality managers to make regulatory decisions about nutrient loadings, and enable public health officials to issue warnings when toxin-producing cyanobacteria are expected to be present. Our work will provide the fundamental science that will be a foundation for further work towards this long-term goal. Specific Aims 1. Describe (cyano)bacterial community dynamics in Lake Mendota over the past decade 2. Parameterize, calibrate, and validate a water quality model for Lake Mendota that simulates microbial community dynamics 3. Evaluate the model's ability to predict (cyano)bacterial community dynamics 4. Identify the temporal and spatial scales at which model predictions are sensitive to (cyano)bacterial community parameterization Outputs The key outcome of this work will be the first effort to relate a coupled hydrodynamic-ecosystem process water quality model to molecular microbial ecology data. Microbial ecologists traditionally do not engage in this kind of mechanistic modeling and therefore are often criticized for being too descriptive in their science. We hope to discover previously unconsidered drivers of cyanobacterial blooms, particularly with respect to interactions between cyanobacteria and heterotrophic bacteria. Our ultimate goal is to provide water quality managers with a sound scientific basis for determining when the public should avoid contact with lake water due to suspected cyanobacterial presence, and to develop schemes to manage the effects of ongoing nutrient loads to the Madison-area lakes. The primary output from this work will be a water quality model that includes discrete groups of cyanobacteria and heterotrophic bacteria as components of the food web. These groups will contribute to important nutrient and carbon cycling pathways within the ecosystem process model, providing an explicit quantification of the contribution of bacteria to the provision of ecosystem services. It will also allow us to ask basic science questions about the linkages between community taxonomic composition and function, addressing an important grand challenge in applied microbial ecology. The model generates state variable values (e.g. soluble reactive phosphorus concentration, nitrate concentration, temperature) as well as fluxes (e.g. dissolved organic phosphorus mineralization rate, denitrification rate, dissolved oxygen gas exchange with atmosphere). These can be used to interpret changes in bacterial community dynamics assessed using molecular techniques.

Project Methods
We will examine the dynamics of heterotrophic bacteria and cyanobacteria in Lake Mendota using a mechanistic ecosystem process model that can be used to predict community change, process rates, and general water quality parameters. We will leverage a ten year time series of (cyano)bacterial community DNA collected by the North Temperate Lakes Microbial Observatory, water quality data collected by the NTL Long Term Ecological Research site, and data from an instrumented buoy that collects limnological and meteorological measurements at high frequency, to parameterize, calibrate, and validate the model. The model will be used to test hypotheses about potential drivers of (cyano)bacterial community composition and activity, and could ultimately be employed as a water quality management tool for Lake Mendota. We have studied bacterial community composition (BCC) in Lake Mendota and cyanobacterial community composition, both using regular sampling of the mixed surface layer biweekly or weekly during the ice-free season (usually May-November). We use a DNA fingerprinting technique known as automated ribosomal intergenic spacer analysis (ARISA), which is inexpensive and rapid, to compare BCC across time and space. The following methods and approaches will be used in this project: 1. BCC analysis by 16S rRNA gene sequencing will be conducted using pooled samples of metagenomic DNA that has been extracted from water samples collected over the past ten years from Lake Mendota. These will be used to assign taxonomic/phylogenetic identities to ARISA peaks for each individual sample. This approach has been used before by our research group and a few others, but not for such an extensive sample set. 2. Cyanobacterial community dynamics over the past ten years will be assessed using a molecular fingerprinting technique that targets the phycocyanin intergenic spacer. This approach was developed in our research group specifically for this purpose. 3. A coupled hydrodynamic-ecosystem process water quality model will be parameterized, calibrated, and validated as part of an ongoing collaboration with other scientists in the Global Lake Ecological Observatory Network. We are adapting the model to dis-aggregate phytoplankton and heterotrophic bacteria parameters specifically for this project. 4. The dynamics of distinct groups of cyanobacteria and heterotrophic bacteria will be compared to those measured as described above using a suite of multivariate statistical techniques. These techniques are commonly used to interpret and compare measured community composition data across systems, but have not been used to compare predicted and measured community dynamics. 5. We will conduct sensitivity analyses for combinations of microbial model parameters using a Monte Carlo approach, to determine the appropriate level of model complexity that most accurately predicts water quality parameters of interest and (cyano)bacterial community dynamics. Although this approach has been used previously to assess model structure, it has not been used in the context of blending mechanistic modeling with molecular microbial ecology.

Progress 10/01/10 to 09/30/14

Outputs
Target Audience: The target audience for this project includes water quality managers (local, regional, and beyond), aquatic microbial ecologists, and microbiologists. We communicate regularly with members of the broader Madison community with interest in connections between lake water quality and agricultural activities, through McMahon's service on the Dane County Lakes and Watershed Commission. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Dr. Emily Kara received her PhD in Environmental Engineering in August 2012. Emily calibrated and validated the water quality model that is the foundation for the project. Mr. Josiah Hawley received his MS in Environmental Engineering in October 2012. Josiah further calibrated and validating the water quality model that is the foundation for the project. Ms. Robin Rohwer, a PhD candidate in the Environmental Chemistry and Technology program, joined the research group in August 2012. Mr. Craig Snortheim started work on his MS in Environmental Engineering in August 2013, co-advised with Dr. Paul Hanson. Craig is contributing to the Hatch project by running the ecosystem water quality model. The project provided training and professional development opportunities to Dr. Kara, Mr. Hawley, Ms. Robin Rohwer, three Civil and Environmental Engineering undergraduate students (Douglas Chalmers, Craig Snortheim, and Aaron Besaw (URM)), and one Molecular Biology undergraduate (Yujin Lee). How have the results been disseminated to communities of interest? Results from this work have been published in peer-reviewed journals, presented at scientific conferences, and presented in public presentations focused on water quality in Madison area lakes. McMahon also has ongoing collaborations with water quality managers and public health professionals with interest in cyanbacterial toxins, and she continues to work with them to incorporate the findings of the work into actionable policy for water quality issues. What do you plan to do during the next reporting period to accomplish the goals? Nothing Reported

Impacts
What was accomplished under these goals? We used an ongoing long term dataset from Lake Mendota, Madison, WI to study microbial communities responsible for controlling water quality in this eutrophic lake. The composition and dynamics of these communities were measured using molecular techniques targeting conserved loci such as the 16S rRNA gene and the phycocyanin intergenic spacer region. A water quality model was parameterized, calibrated, and validated using long term data records and the model output was used as explanatory drivers of microbial community dynamics. We also employed advanced analytical techniques to measure cyanotoxin concentrations and phosphorus speciation in the lake. Highlights from our accomplishments and findings are provided below. Long-Term Microbial Community Dynamics With an unprecedented decade-long time series from a temperate eutrophic lake, we analyzed bacterial and environmental co-occurrence networks to gain insight into seasonal dynamics at the community level. We found that (1) bacterial co-occurrence network were non-random, (2) season explained network complexity, and (3) co-occurrence network complexity was negatively correlated with underlying community diversity across different seasons. Network complexity was not related to the variance of associated environmental factors. Temperature and productivity may drive changes in diversity across seasons in temperate aquatic systems, much as they control diversity across latitude. While the implications of bacterioplankton network structure on ecosystem function are still largely unknown, network analysis, in conjunction with traditional multivariate techniques, continues to increase our understanding of bacterioplankton temporal dynamics. Drivers of cyanobacterial community dynamics and toxicity: the role of nitrogen We combined three years of temporal data, including microcystin concentrations, 16 years of long-term ecological research, and 10 years of molecular data to investigate the potential factors leading to the selection of toxic Microcystis and microcystin production in Lake Mendota. Our analysis revealed that nitrogen speciation (specifically ammonium) drives population dynamics and that an imbalance in cellular carbon to nitrogen ratios may trigger toxin production. More specifically, precipitous declines in ammonium concentrations lead to a transitional period of nitrogen stress, even in the presence of high nitrate concentrations, that we call the "toxic phase." Following the toxic phase, temperature and cyanobacterial abundance remained elevated but microcystin concentrations drastically declined. Increases in ammonium due to lake turnover may have led to microcystin synthesis being shutdown or a shift in the community from toxic to non-toxic species. While total phosphorus to total nitrogen ratios were relatively low over the time-series, microcystin concentrations were highest when total nitrogen to total phosphorus ratios were also highest. Similarly, high carbon to nitrogen ratios were also strongly correlated to the toxic phase. We propose a metabolic model that corroborates molecular studies and reflects our ecological observations that carbon and nitrogen metabolism regulate microcystin production physiologically and ecologically. In particular, we hypothesize that an imbalance between 2-oxoglutarate and ammonium in the cell has direct bearing on microcystin synthesis in the environment. Phosphorus speciation in a eutrophic lake revealed by 31P NMR For eutrophic lakes, patterns of phosphorus (P) measured by standard methods are well documented but provide little information about the components comprising standard operational definitions. Dissolved P (DP) and particulate P (PP) represent important but rarely characterized nutrient pools. Samples from Lake Mendota were characterized using 31-phosphorus nuclear magnetic resonance spectroscopy (31P NMR) during the open water season of 2011 in this unmatched temporal study of aquatic P dynamics. A suite of organic and inorganic P forms was detected in both dissolved and particulate fractions: orthophosphate, orthophosphate monoesters, orthophosphate diesters, pyrophosphate, polyphosphate, and phosphonates. Significant transformations of P-containing molecules over time and space were evident: at all sampling locations, DP was chemically distinct from PP; inflows were distinct from in-lake P; and temporal patterns of DP and PP at the surface and hypolimnion differed. Through time, lake mixing and other biogeochemical factors were associated with changes in the relative proportion of polyphosphate and phosphonate in particulate samples. Particulate P can be used as a proxy for phytoplankton-bound P, and in this study, a high proportion of polyphosphate within particulate samples suggested P was not limiting for the dominant primary producers, cyanobacteria. Hypolimnetic particulate P samples were more variable in composition than surface samples, potentially due to varying production and transport of sinking particles. Surface dissolved samples contained less P than particulate samples, and were typically dominated by orthophosphate, but also contained monoester, diester, polyphosphate, pyrophosphate, and phosphonate. Hydrologic inflows to the lake contained more orthophosphate and orthophosphate monoesters than in-lake samples, indicating transformation of P from inflowing waters. This time series explores trends of a highly regulated nutrient in the context of other water quality metrics (chlorophyll, mixing regime, and clarity), and gives insight on the variability of the structure and occurrence of P-containing compounds in light of the phosphorus-limited paradigm. Water quality model development We evaluated the predictive ability of a one-dimensional coupled hydrodynamic-biogeochemical model across multiple temporal scales using wavelet analysis and traditional goodness-of-fit metrics. High-frequency in situ automated sensor data and long-term manual observational data from Lake Mendota were used to parameterize, calibrate, and evaluate model predictions. We focused specifically on short-term (< 1 month) predictions of phytoplankton biomass over one season. Traditional goodness-of-fit metrics indicated more accurate prediction of physics than chemical or biological variables in the time domain. This was confirmed by wavelet analysis in both the time and frequency domains. For temperature, predicted and observed global wavelet spectra were closely related, while observed dissolved oxygen and chlorophyll-a fluorescence spectral characteristics were not reproduced by the model for key time scales, indicating that processes not modelled may be important drivers of the observed signal. Although the magnitude and timing of physical and biological changes can be simulated adequately at the seasonal scale through calibration, time-scale specific dynamics, for example short-term cycles, are difficult to reproduce. Wavelet transforms and diverse observational data provide for model evaluation techniques that are complementary to traditional goodness-of-fit metrics, and are particularly well suited for assessment of temporal and spatial heterogeneity when coupled to high-frequency data from automated in situ or remote sensing platforms.

Publications

• Type: Journal Articles Status: Published Year Published: 2014 Citation: A. Eiler, K. Zaremba-Niedzwiedzka, S. G. E. Andersson, M. Martinez Garcia, K. D. McMahon, R. Stepanauskas and S. Bertilsson. (2014) Productivity and salinity structuring of the microplankton revealed by comparative freshwater metagenomics. Environmental Microbiology, 16(9):2682-2698.
• Type: Journal Articles Status: Published Year Published: 2014 Citation: T. W. Ghylin, S. L. Garcia, F. Moya, B. O. Oyserman, P. Schwientek, K. T. Forest, J. Mutschler, L.-K. Chan, M. Martinez-Garcia, A. Sczyrba, R. Stepanauskas, H.-P. Grossart, T. Woyke, F. Warnecke, R. Malmstrom, S. Bertilsson, K. D. McMahon. (2014) Comparative single-cell genomics reveals potential ecological niches for the freshwater acI Actinobacteria lineage. ISME Journal, 8:2503-2516.
• Type: Journal Articles Status: Published Year Published: 2014 Citation: E. K. Read, M. Ivancic, P. C. Hanson, B. J. Cade-Menun, K. D. McMahon. (2014) Phosphorus speciation in a eutrophic lake by 31P NMR spectroscopy. Water Research, 62:229-240.
• Type: Journal Articles Status: Under Review Year Published: 2015 Citation: L. J. Beversdorf, T. R. Miller, K. D. McMahon. Long-term monitoring reveals carbon-nitrogen metabolism key to microcystin production in eutrophic lakes In review
• Type: Journal Articles Status: Published Year Published: 2014 Citation: L. J. Beversdorf, T. R. Miller, K. D. McMahon. (2103) The role of nitrogen fixation in cyanobacterial bloom toxicity in a temperate, eutrophic lake. PLoS One. 8(2):e56103.

Progress 01/01/13 to 09/30/13

Outputs
Target Audience: The target audience for this project includes water quality managers (local, regional, and beyond), aquatic microbial ecologists, and microbiologists. We communicate regularly with members of the broader Madison community with interest in connections between lake water quality and agricultural activities, through McMahon’s service on the Dane County Lakes and Watershed Commission. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided? Dr. Emily Kara received her PhD in Environmental Engineering in August 2012. Emily calibrated and validated the water quality model that is the foundation for the project. Mr. Josiah Hawley received his MS in Environmental Engineering in October 2012. Josiah further calibrated and validating the water quality model that is the foundation for the project. Ms. Robin Rohwer, a PhD candidate in the Environmental Chemistry and Technology program, joined the research group in August 2012. Mr. Craig Snortheim started work on his MS in Environmental Engineering in August 2013, co-advised with Dr. Paul Hanson. Craig contributed to the project by running the ecosystem water quality model. Collaborators include: Dr. Paul Hanson, Scientist in the Center for Limnology; Dr. Jack Gilbert, Argonne National Laboratory. Dr. Hanson worked closely with Emily and Craig to calibrate and validate the model. He also performed spectral analysis on the model output to determine how well it captures variability at different temporal scales. Training and Professional Development: The project provided training and professional development opportunities to Dr. Kara, Mr. Hawley, Ms. Robin Rohwer, and three Civil and Environmental Engineering undergraduate students (Douglas Chalmers, Craig Snortheim, and Aaron Besaw (URM)). How have the results been disseminated to communities of interest? We have published three primary literature manuscripts and one review based on our findings. These were published in high impact journals. McMahon continues to interact with local water quality managers, private citizens with interest in lake water quality, the private-business backed Clean Lakes Alliance, farmers groups, and other interest groups, through her service on the Dane County Lakes and Watershed Commission. http://www.danewaters.com/about/ What do you plan to do during the next reporting period to accomplish the goals? We will continue to refine the water quality model calibration while using model output to interpret bacterial community dynamics measured using the 16S rRNA gene tag sequencing.

Impacts
What was accomplished under these goals? We have completed simulations using our coupled hydrodynamic-ecosystem process water quality model based on driver data from 2003-2010. The model was calibrated using data from 2008, 2009, and 2010. We are now using the model to evaluate climate change scenarios and their effects on the timing of hypoxia in the hypolimnion. We are using model output data as correlates to bacterial community dynamics. We continue to collaborate with Dr. Paul Hanson in the Center for Limnology and members of the Global Lake Ecological Observatory Network. We successfully completed a small project funded by the USGS to assess the effects of climate change on Lake Mendota, using the same datasets. We published a manuscript based on the analysis of bacterial community composition over ten years in Lake Mendota. We used automated ribosomal intergenic spacer analysis (ARISA) to create community fingerprints. Local similarity analysis revealed potential co-occurrence networks that hint at ecological interactions among species-like bacterial groups. We found that (1) bacterial co-occurrence network were non-random, (2) season explained network complexity, and (3) co-occurrence network complexity was negatively correlated with underlying community diversity across different seasons. Network complexity was not related to the variance of associated environmental factors. Temperature and productivity may drive changes in diversity across seasons in temperate aquatic systems, much as they control diversity across latitude. While the implications of bacterioplankton network structure on ecosystem function are still largely unknown, network analysis, in conjunction with traditional multivariate techniques, continues to increase our understanding of bacterioplankton temporal dynamics. Citation: E. L. Kara, P. C. Hanson, Y. H. Hu, L. Winslow, K. D. McMahon. (2013) “A decade of seasonal dynamics and interactions within freshwater bacterioplankton communities from eutrophic Lake Mendota, Wisconsin, USA”, ISME Journal, 7(3):680-684. We conducted “deep” 16S rRNA gene tag sequencing on 100 samples collected from Lake Mendota between 2000 and 2010 (many of the same sequences analyzed by ARISA), in collaboration with Dr. Jack Gilbert at Argonne National Laboratory and the Earth Microbiome Project. This represents one of the longest and most temporally rich datasets of microbial community composition ever constructed using these next-generation sequencing tools. We have performed the bioinformatics analyses required to assign phylogenetic group affiliations to operational taxonomic units (OTUs) in the dataset. Remarkably, across the whole dataset, 56% of the amplicon reads could be assigned to only 20 OTUs. We are proceeding to incorporate these 20 OTUs and a few other rarer ones into our modeling framework. We also generated multiple autoregressive models to use a subset of OTUs and environmental variables to as predictors for key OTUs of interest. We successfully secured funding from the National Science Foundation (INSPIRE program) to continue the modeling work beyond September 2014, using the findings from the Hatch project as preliminary data. This new project includes linking genome-informed metabolic modeling to key traits and embedding these trait-based populations in the ecosystem model. The INSPIRE project would not have been possible without building on the Hatch project.

Publications

• Type: Journal Articles Status: Published Year Published: 2013 Citation: E. L. Kara, P. C. Hanson, Y. H. Hu, L. Winslow, K. D. McMahon. (2013) A decade of seasonal dynamics and interactions within freshwater bacterioplankton communities from eutrophic Lake Mendota, Wisconsin, USA, ISME Journal, 7(3):680-684.
• Type: Journal Articles Status: Published Year Published: 2012 Citation: 16. E. L. Kara, P. Hanson, D. Hamilton, M. Hipsey, K. D. McMahon, J. Read, L. Winslow, J. Dedrick, K. Rose, C. Carey, S. Bertilsson, D. da Motta Marques, L. Beversdorf, T. Miller, C. Wu, Y-F. Hsieh, E. Gaiser, T. Kratz, (2012) "Time-scale dependence in numerical simulations: Assessment of physical, chemical, and biological predictions in a stratified lake at temporal from scales of hours to months". Environmental Modelling and Software. 35:104-121.
• Type: Journal Articles Status: Published Year Published: 2013 Citation: S. L. Garcia, K. D. McMahon, M. Martinez-Garcia, A. Srivastava, A. Sczyrba, R. Stepanauskas, H-P. Grossart, T. Woyke, F. Warnecke. (2013) Metabolic potential of a single cell belonging to one of the most abundant lineages of freshwater bacterioplankton ISME Journal. 7(1):137-147.
• Type: Journal Articles Status: Published Year Published: 2013 Citation: K. D. McMahon and E. K. Read. (2013) Microbial contributions to phosphorus cycling in eutrophic lakes and wastewater Annual Review of Microbiology. 67: 199-219.

Progress 01/01/12 to 12/31/12

Outputs
OUTPUTS: We have completed simulations using our coupled hydrodynamic-ecosystem process water quality model based on driver data from 2003-2010. The model was calibrated using data from 2008, 2009, and 2010. We continue to collaborate with Dr. Paul Hanson in the Center for Limnology and members of the Global Lake Ecological Observatory Network. The modeling effort will be used as a template for further modeling projects within GLEON. We are planning a workshop for October 2013 to disseminate our findings to interested GLEON members, and to initiate a similar modeling project for Lake Erken in Sweden. This demonstrates an international level of impact of our work. We published a manuscript based on the analysis of bacterial community composition over ten years in Lake Mendota. We used automated ribosomal intergenic spacer analysis (ARISA) to create community fingerprints. Local similarity analysis revealed potential co-occurrence networks that hint at ecological interactions among species-like bacterial groups. We found that (1) bacterial co-occurrence network were non-random, (2) season explained network complexity, and (3) co-occurrence network complexity was negatively correlated with underlying community diversity across different seasons. Network complexity was not related to the variance of associated environmental factors. Temperature and productivity may drive changes in diversity across seasons in temperate aquatic systems, much as they control diversity across latitude. While the implications of bacterioplankton network structure on ecosystem function are still largely unknown, network analysis, in conjunction with traditional multivariate techniques, continues to increase our understanding of bacterioplankton temporal dynamics. We conducted "deep" 16S rRNA gene tag sequencing on 100 samples collected from Lake Mendota between 2000 and 2010 (many of the same sequences analyzed by ARISA), in collaboration with Dr. Jack Gilbert at Argonne National Laboratory and the Earth Microbiome Project. This represents one of the longest and most temporally rich datasets of microbial community composition ever constructed using these next-generation sequencing tools. We have performed the bioinformatics analyses required to assign phylogenetic group affiliations to operational taxonomic units (OTUs) in the dataset. Remarkably, across the whole dataset, 56% of the amplicon reads could be assigned to only 20 OTUs. We are proceeding to incorporate these 20 OTUs and a few other rarer ones into our modeling framework. We successfully completed a small project funded by the USGS to assess the effects of climate change on Lake Mendota, using the same datasets. PARTICIPANTS: Participants in the project include: Dr. Katherine McMahon, Associate Professor. Dr. McMahon is directing the project and supervising other personnel. Dr. Emily Kara received her PhD in Environmental Engineering in August 2012. Emily calibrated and validated the water quality model that is the foundation for the project. Mr. Josiah Hawley received his MS in Environmental Engineering in October 2012. Josiah further calibrated and validating the water quality model that is the foundation for the project. Ms. Robin Rohwer, a PhD candidate in the Environmental Chemistry and Technology program, joined the research group in August 2012. Collaborators include: Dr. Paul Hanson, Scientist in the Center for Limnology. Dr. Hanson worked closely with Emily to calibrate and validate the model. He also performed spectral analysis on the model output to determine how well it captures variability at different temporal scales. Training and Professional Development: The project provided training and professional development opportunities to Dr. Kara, Mr. Hawley, Ms. Robin Rohwer, and three Civil and Environmental Engineering undergraduate students (Douglas Chalmers, Craig Snortheim, and Aaron Besaw (URM)). TARGET AUDIENCES: Nothing significant to report during this reporting period. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
We report a significant change in knowledge: Graduate students Emily Kara and Josiah Hawley have learned a significant amount about how to perform water quality modeling, including model calibration and validation. They have produced a model that performs reasonably well based on available observational data used for validation. Using sensitivity analyses we have learned which model components most significantly affect key water quality parameters such as phytoplankton biomass concentration and phosphate concentrations. Our 16S rRNA gene tag data is a landmark dataset that can be mined for interesting patterns of co-occurrence among OTUs and correlation to environmental variables, as well as process variables predicted in the model. This preliminary work has prepared us to better address subsequent project objectives.

Publications

• E. L. Kara, P. C. Hanson, Y. H. Hu, L. Winslow, K. D. McMahon. (2013) A decade of seasonal dynamics and interactions within freshwater bacterioplankton communities from eutrophic Lake Mendota, Wisconsin, USA, ISME Journal, in press.
• E. L. Kara, P. Hanson, D. Hamilton, M. Hipsey, K. D. McMahon, J. Read, L. Winslow, J. Dedrick, K. Rose, C. Carey, S. Bertilsson, D. da Motta Marques, L. Beversdorf, T. Miller, C. Wu, Y-F. Hsieh, E. Gaiser, T. Kratz, (2012) Time-scale dependence in numerical simulations: Assessment of physical, chemical, and biological predictions in a stratified lake at temporal from scales of hours to months. Environmental Modelling and Software. 35:104-121.

Progress 01/01/11 to 12/31/11

Outputs
OUTPUTS: We have finished parameterizing our coupled hydrodynamic-ecosystem process water quality model based on driver data and observations from the 2008 ice-free season. We have also gathered the necessary driver data to run the model for ten years (2001-2010) and we are now working on calibrating the model based on 2009 driver and observed data. We continue to collaborate with Dr. Paul Hanson in the Center for Limnology and members of the Global Lake Ecological Observatory Network. The modeling effort will be used as a template for further modeling projects within GLEON. We are planning a workshop for October 2012 to disseminate our findings to interested GLEON members, and to initiate a similar modeling project for Lake Erken in Sweden. This demonstrates an international level of impact of our work. We also prepared archived DNA samples for 16S rRNA gene tag sequencing to be conducted at Argonne National Laboratory as part of a collaboration related to this work. This information will be used to parameterize bacterial populations in the next model configuration. We successfully secured support for a small project with the USGS to assess the effects of climate change on Lake Mendota, using the same datasets. PARTICIPANTS: Participants in the project include: Dr. Katherine McMahon, Associate Professor. Dr. McMahon is directing the project and supervising other personnel. Ms. Emily Kara, PhD candidate in Environmental Engineering. Emily is calibrated and validated the water quality model that is the foundation for the project. She is also conducting analyses of bacterial community composition data and relating it to the project. Mr. Josiah Hawley, MS candidate in Environmental Engineering. Josiah is further calibrating and validating the water quality model that is the foundation for the project. Collaborators include: Dr. Paul Hanson, Scientist in the Center for Limnology. Dr. Hanson is working closely with Emily to calibrate and validate the model. He is also performing spectral analysis on the model output to determine how well it captures variability at different temporal scales. Training and Professional Development: The project is providing training and professional development opportunities to Ms. Kara, Mr. Hawley and one undergraduate student (Douglas Chalmers, sophomore in Civil and Environmental Engineering). TARGET AUDIENCES: Water quality professionals interested in the Yahara Lakes (e.g. WI DNR, Madison City Engineering, Dane County Lakes and Watershed) PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
We report a significant change in knowledge: Graduate students Emily Kara and Josiah Hawley have learned a significant amount about how to perform water quality modeling, including model calibration and validation. They have produced a model that performs reasonably well based on available observational data used for validation. Using sensitivity analyses, we have learned which model components most significantly affect key water quality parameters such as phytoplankton biomass concentration and phosphate concentrations. This preliminary work has prepared us to better address subsequent project objectives.

Publications

• E. L. Kara, P. Hanson, D. Hamilton, M. Hipsey, K. D. McMahon, J. Read, L. Winslow, J. Dedrick, K. Rose, C. Carey, S. Bertilsson, D. da Motta Marques, L. Beversdorf, T. Miller, C. Wu, Y-F. Hsieh, E. Gaiser, T. Kratz, 2012. "Time-scale dependence in numerical simulations: Assessment of physical, chemical, and biological predictions in a stratified lake at temporal from scales of hours to months". In press.

Progress 10/01/10 to 12/31/10

Outputs
OUTPUTS: We have finished parameterizing our coupled hydrodynamic-ecosystem process water quality model based on driver data and observations from the 2008 ice-free season. Model calibration was performed using historical water quality data from the North Temperate Lakes Long Term Ecological Research site. Mean seasonal predictions of physical, chemical, and biological variables were accurate for key water quality variables. Broad spatial and temporal patterns were represented. Model simulations reproduced the general magnitude of phytoplankton biomass; however, some observed patterns were not reproduced. We continue to explore why this is the case. We used the model to evaluate several scenarios related to altered nutrient loadings and precipitation events. The results of our scenario testing indicate that future changes in nutrient loading may be more important for phytoplankton biomass than changes in climate. Seasonal phytoplankton biomass and productivity were influenced by P initial conditions more than spring water temperature; direct effects on respiration were more difficult to interpret. These results indicate that effects of future climate change on hydrology and thermal stability may have greater effect on phytoplankton biomass than direct temperature effects on algal physiology. This model response is consistent with literature on controls of biomass and primary production, particularly at higher, summertime temperatures. We continue to collaborate with Dr. Paul Hanson in the Center for Limnology and members of the Global Lake Ecological Observatory Network. The modeling effort will be used as a template for further modeling projects within GLEON. We are planning a workshop for summer 2011 to disseminate our findings to interested GLEON members, and to initiate a similar modeling project for Lake Erken in Sweden. This demonstrates an international level of impact of our work. We also prepared archived DNA samples for 16S rRNA pyrotag sequencing to be conducted at Argonne National Laboratory as part of a collaboration related to this work. This information will be used to parameterize bacterial populations in the next model configuration. We also used the preliminary results from this work as the basis for another proposal to the USGS to evaluate climate change scenarios using the model. If the project is funded, it will synergize well with this project. PARTICIPANTS: Participants in the project include: Dr. Katherine McMahon, Associate Professor. Dr. McMahon is directing the project and supervising other personnel. Ms. Emily Kara, PhD candidate in Environmental Engineering. Emily is calibrating and validating the water quality model that is the foundation for the project. Collaborators include: Dr. Paul Hanson, Scientist in the Center for Limnology. Dr. Hanson is working closely with Emily to calibrate and validate the model. He is also performing spectral analysis on the model output to determine how well it captures variability at different temporal scales. Training and Professional Development: The project is providing training and professional development opportunities to Ms. Kara and one undergraduate student (Peter Bergquist, junior in Civil and Environmental Engineering). TARGET AUDIENCES: Nothing significant to report during this reporting period. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.

Impacts
We report a significant change in knowledge: Graduate student Emily Kara has learned a significant amount about how to perform water quality modeling, including model calibration and validation. She has produced a model that performs reasonably well based on available observational data used for validation. Using sensitivity analyses we have learned which model components most significantly affect key water quality parameters such as phytoplankton biomass concentration and phosphate concentrations. This preliminary work has prepared us to better address subsequent project objectives.

Publications

• No publications reported this period