Thuman, Andrew

Andy has over 30 years of experience conducting water quality modeling studies in all types of water bodies for permit and TMDL development. These studies include evaluating the effects of nutrient loading on algal blooms, dissolved oxygen and transparency. At HDR, he is the water quality practice lead and manages the Water Quality Modeling Group in Mahwah NJ. He grew up on the south shore of Long Island New York, attended Manhattan College and enjoys being outdoors as much as possible.

Titles from this speaker

Nutrient Permitting in Newport Bay Using a Simplified Modeling Approach

Abstract

The objective of this paper is to provide the technical background (data analysis and modeling) that led to the Irvine Ranch Water District (IRWD) successfully obtaining a NPDES permit for a new diversion to San Diego Creek from the Michelson Water Recycling Plant (MWRP) during wet weather conditions. Data analyses provided the background on existing water quality issues and trends, time periods of sensitive water quality, and current baseline water quality conditions. The modeling effort used a hydrodynamic model to calculate tidal circulation in the downstream Newport Bay and provided estimates of nutrient effects due to the diversion and bay flushing times during wet weather periods.

SpeakerThuman, Andrew

Presentation time

09:10:00

09:30:00

Session time

08:30:00

10:10:00

SessionResiliency Runway: New Trends in Modeling

Session number526

TopicResilience, Disaster Planning and Recovery, Utility Management and Leadership, Watershed Management, Water Quality, and Groundwater

Author(s)

A. ThumanA. ThumanG. EstradaN. KimD. RapantP. Cook

Author(s)A. Thuman1; A. Thuman1; G. Estrada2; N. Kim1; D. Rapant1; P. Cook3;

Author affiliation(s)HDR, NJ1; HDR, Inc., CA2; Irvine Ranch Water District, CA3

SourceProceedings of the Water Environment Federation

Document typeConference Paper

PublisherWater Environment Federation

Print publication date Oct 2020

DOI10.2175/193864718825157494

Volume / Issue

Content sourceWEFTEC

Word count11

St. Albans Bay Phosphorus Study: A Simplified Modeling Approach

Abstract

Objectives
The objective of the presentation is to describe the modeling used to evaluate phosphorus management alternatives in St. Albans Bay, Vermont. The modeling used a simple spreadsheet Lake Loading Response Model (LLRM) to estimate watershed phosphorus loads and response in the bay along with a stand-alone sediment flux model (SFM) to estimate internal sediment phosphorus fluxes. However, the simple modeling approaches required a few major assumptions that included a phosphorus dispersive exchange between the bay and Lake Champlain. This presentation will also describe how model sensitivity runs allowed evaluation of the assumptions used. A number of management scenarios were analyzed with LLRM to evaluate the effect of watershed and sediment load reduction alternatives to reduce phosphorus and algal blooms in the bay. Results of the modeling will help guide future phosphorus management strategies in the bay.
Background
St. Albans Bay is located off of Lake Champlain and has a long history of eutrophic conditions and excessive algal blooms during the summer. Figure 1 presents the St. Albans Bay watershed and land uses. Major sources of nutrients to the bay include the City of St. Albans Wastewater Treatment Facility (WWTF) as well as runoff from the agricultural dominated watershed. In 1987, the WWTF was upgraded to reduce phosphorus loading by 90%. However, algal blooms and phosphorus levels in the bay did not improve as expected. The USACE Phase II Feasibility Study for Wetland and Phosphorus Management funded the effort and in coordination with the Vermont Department of Environmental Conservation (VTDEC) is planning for short- and long-term phosphorus management that include sediment phosphorus control in addition to potential watershed and wetland controls. This project was completed in 2021 and implementation of management alternatives is under discussion.
Approach
To evaluate watershed phosphorus loading and response in the bay, we integrated two simple models: an Excel-based steady-state Lake Loading Response Model (LLRM) for watershed loads and bay response; and a time-varying sediment flux model (SFM) to provide sediment phosphorus fluxes and response time for management actions. Setup for LLRM involved evaluation of the three major rivers that discharge to the bay (Mill River, Jewett Brook and Stevens Brook), runoff loads and flows based on land use, WWTF effluent loads, and the exchange of phosphorus between the bay and open waters of Lake Champlain. LLRM assumes a completely mixed volume with inflows and outflows. The bay is not a fully enclosed lake, but has a downstream connection with Lake Champlain that includes dispersive exchange in addition to outflow. One of the major challenges was incorporating the exchange of phosphorus between the bay and open waters of Lake Champlain. Setup for the SFM involved analysis of data, estimates of settled loads for inputs along with overlying water column dissolved oxygen and temperature to calculate the time-variable response in sediment phosphorus fluxes and sediment oxygen demand (SOD). Results of the SFM were then incorporated into LLRM. LLRM was calibrated to available runoff flows, watershed loads, bay total phosphorus (TP) and chlorophyll-a for the years 2016 through 2019 that represented low, average and high flow years. The SFM was calibrated to historic sediment data that included sediment phosphorus fluxes, SOD, sediment pore water and particulate concentrations. Both models reasonably reproduced the available data and were considered useful the intended project use. Thirteen different load reduction scenarios in addition to the baseline conditions were analyzed using the LLRM-SFM integrated models. The scenarios included reductions in watershed loads, reductions in internal sediment loads, and various combinations of sediment and watershed load reductions. These model scenarios were used to estimate the water quality benefits associated with the various load reduction alternatives. As part of the LLRM modeling, the dispersive exchange between the bay and lake was needed. This dispersive exchange was calculated from available chloride and phosphorus data with the concentration gradient between the bay/lake proving to be a very important factor affecting calculated bay phosphorus levels. For the model projection scenarios, specification of the lake phosphorus concentration in calculating the dispersive exchange was a major factor in determining the magnitude of water quality improvements in the bay. To account for the lake concentration in the projection scenarios, multiple model scenarios were run assuming different lake TP concentrations for the dispersive exchange rate calculation. These lake TP concentrations ranged from the proposed lake TMDL TP concentration to existing lake TP concentrations. Figure 2 presents the significant impact that the assigned lake TP concentration has on the resulting bay TP reduction for a number of different load reduction scenarios.
Findings/Results
Although there were many assumptions that needed to be made for application of these simplified models, the LLRM and SFM reproduced the data well in all years except 2019, which happened to be a very high flow year. This is likely due to the limitation of the model with regards to the bay/lake dispersive exchange and availability data.
The following main findings are provided from this modeling study.
1. Model projection scenario and sensitivity results showed the significant impact that the assigned lake TP concentration (used to calculate bay/lake dispersive exchange) has on the LLRM model results. For example, the calculated bay TP concentration increases from 23 Âµg/L (using a lake TP of 14 Âµg/L) to 30 Âµg/L (using a lake TP of 23 Âµg/L). This equates to a percent TP reduction that decreases from 32% to 7%.
2. As phosphorus load reductions in the watershed are implemented and ultimately achieved, their effectiveness in reducing bay TP and chl-a levels need to be viewed together with actual reductions in lake TP levels. Without concurrent reductions in lake TP levels, expected improvements in bay water quality due to load reductions may not be realized because lake TP levels are not changing.
3. There are benefits to applying simple models to evaluate load reductions on water quality improvements that include reduced data needs, shorter time (and cost) for model development and ease of use. Application of these simple models can provide guidance for management strategies but may require various assumptions to complete the analysis.
4. In this case, the bay/lake dispersive exchange could be better represented with use of more complex hydrodynamic/transport models; and the estimation of sediment phosphorus flux could be better represented with a coupled eutrophication/SFM model.
5. Although there are limitations in the use of simple modeling tools, as long as the limitations and assumptions are transparent, their utility in evaluating watershed management strategies can be effective.

This paper describes the modeling used to evaluate phosphorus management in St. Albans Bay, Vermont. The modeling used a simple spreadsheet lake loading response model to estimate watershed phosphorus loads and response in the bay and a stand-alone sediment flux model to estimate internal sediment phosphorus fluxes. A number of management scenarios were analyzed to evaluate the effect of watershed and sediment load reduction alternatives to reduce phosphorus and algal blooms in the bay.

SpeakerThuman, Andrew

Presentation time

16:00:00

16:25:00

Session time

15:30:00

17:00:00

TopicIntermediate Level, Watershed Management, Water Quality, and Groundwater, Wet Weather

Author(s)

Thuman, Andrew

Author(s)Andrew J. Thuman1; Mikayla Reichard2

Author affiliation(s)HDR Engineering, Inc., Mahwah, NJ1; HDR Engineering, Inc., Mahwah, NJ2