Why TMDLs are needed
Water quality monitoring programs, like the Buzzards Bay Coalition Baywatchers Program, surveys of eelgrass, and other studies have shown that many areas of Buzzards Bay are adversely impacted by excess inputs of nitrogen from human sources. As required under the section 303(d) of the US Clean Water Act, the Massachusetts Department of Environmental Protection (DEP) lists these areas as impaired on their Integrated List of Waters. For those systems (“assessed units”) that have been documented as impaired (Category 5 of the Integrated List), the state is required to establish a Total Maximum Daily Load (TMDL) limit for nitrogen. These TMDLs are then approved by the US Environmental Protection Agency (their list is posted on this page).
How nitrogen TMDLs are developed
There are various ways in which a state could establish a nitrogen TMDL limit for an estuary, and the EPA developed these Nutrient TMDL protocols in 1999 to guide the states in their efforts. This document includes the conceptual formula in the box to the right. In general terms the process involves characterizing existing impairments, defining the contributing watershed, quantifying each source of nitrogen in the watershed and its contribution to the receiving waters, identifying water quality targets or living resources goals for the estuary, linking these loading sources to water quality, then establishing loading targets in a TMDL to achieve water quality targets or living resources goals.
In terms of establishing a TMDL, one of the most important and complex tasks is to link existing and future nitrogen loading estimates to water quality targets. EPA acknowledged the complexity and evolving scientific understanding of ecosystem response to nutrient pollution, and they did not prescribe a specific modeling approach or methodology for establishing TMDLs. They also acknowledged the role of “best professional judgment” in the process. In recognizing the importance of the use of simulation models for establishing TMDLs, EPA also cautioned, “If an appropriate concentration-response relationship cannot link indicators and sources, an appropriate simulation model can be used. A key aspect of model identification is the complexity, cost, and effort of implementation, which must be balanced against the benefits achieved by using the model to estimate the TMDL…. Public understanding and communication also can be crucial to choosing an analytical technique. This is particularly important for TMDLs that must rely on voluntary management measures to control nonpoint loads. Using a model that is overly complex, poorly documented, not peer reviewed, proprietary, or not well known will increase the difficulty of understanding, communicating, and gaining acceptance of the results.” (page 6-5 of EPA’s 1999 “Protocol for Developing Nutrient TMDLs“).
The Massachusetts DEP administers the TMDL program in Massachusetts. Any questions about the program or its implementation or approval of TMDLs should be directed to the contacts of this DEP TMDL page.
Nitrogen TMDLs in Massachusetts and the MEP
In 2001, the Massachusetts DEP implemented the Massachusetts Estuaries Project (MEP) to develop TMDLs for 89 embayments in Massachusetts. This effort represents fulfillment of one of the goals of the Buzzards Bay CCMP, to identify embayment specific nitrogen loading limits based on sound science. The Massachusetts DEP prepared this fact sheet on implementing TMDLs. Additional information on the MEP can be found on this BBNEP MEP page, the MEP Oceanscience.net site, UMass Dartmouth SMAST site, and this DEP MEP case studies page.
The core foundation of setting a TMDL for an estuary is to select a sentinel monitoring station and to establish a threshold target total nitrogen concentration for that station. Typically the MEP selects a sentinel station of around 0.4 ppm or less total nitrogen, if the water quality goal is to restore or protect eelgrass, and 0.5 ppm or higher, if the goal is only to protect or restore benthic habitat like shellfish beds. The Buzzards Bay NEP had proposed similar water quality standards in this 2000 report. The actual selection of the exact TN at a sentinel station in an estuary, and the location of that station, are subjective decisions based on best professional judgment, and involves consideration of conditions and loading in nearby estuaries that are used as reference systems. The selection of a target total nitrogen concentration and the placement of the sentinel station affect the estimated total maximum daily load (TMDL) of nitrogen goal for that estuary, and how much watershed nitrogen loading must be reduced. The MEP makes these decisions based on best professional judgment using eelgrass distribution and nitrogen concentrations in nearby estuaries, and modeled nitrogen concentration and eelgrass distributions in the estuary being studied. They then run their water quality models to determine what amount of watershed nitrogen loading reduction would result in the target concentration at the sentinel station.
The MEP approach uses parcel specific town data and other GIS coverages to estimate septic systems, impervious areas, and other features relevant to a nitrogen loading analysis, including property specific water use data to estimate septic loading. This data is incorporated into a conventional loading model which is then linked to a water quality model that incorporates embayment tidal circulation. The model is calibrated so that loading matches existing conditions, then altered to predict what level of nitrogen reduction is necessary to meet the Massachusetts Surface Water Quality Standards. More specifically, the MEP recommends a specific total nitrogen concentration targets at a sentinel station in the receiving waters for their nitrogen TMDL development. For Buzzards Bay embayments, these targets are typically around 0.4 ppm total nitrogen for eelgrass protection, 0.5 ppm total nitrogen for fish, shellfish, and benthic habitat protection, but the actual values depend on nitrogen concentrations outside the embayment, and how representative the sentinel station is of average conditions in the estuary. Individual elements of this approach are highlighted below, but a more thorough explanation of the MEP methodology is contained in this 2003 MEP report appendix.
After a draft TMDL report is prepared, municipalities should review the MEP loading spreadsheets to ensure that loading model used reasonable estimates of the number of septic systems, and areas of impervious surfaces, golf courses, and agricultural lands were incorporated into the loading model, and that all known point source discharges, including any open or closed landfills were considered in the analysis. After municipalities and DEP review the TMDL reports, DEP prepares the TMDL for submission to and approval by the EPA.
A TMDL essentially represents an estimate of the assimilative capacity of an ecosystem to handle a pollutant (with an appropriate margin of safety). It can be revised or replaced by DEP and EPA as new data, information, or models become available.
Once a TMDL is in place, EPA is required to issue any water discharge permits (including discharges from permitted stormwater networks) so they achieve maximum feasible compliance with the “Waste Load Allocation” (WLA) of the TMDL. The US EPA has less direct control over mandating reductions in the non-point sources of pollution included in the “Load Allocation” component of the TMDL, including onsite septic systems. This limitation in the regulations promulgated by EPA has become subject of a legal challenge by the Buzzards Bay Coalition and the Conservation Law Foundation.
MEP Loading Model
The MEP watershed loading methodology is contained in this 2003 MEP report appendix and key loading coefficients are summarized on our Nitrogen Management Tools and our Nitrogen Loading Assumptions page. The Buzzards Bay NEP now uses the MEP’s watershed loading assumptions in any project analyses we conduct for the towns, and we encourage municipalities to also use these loading assumptions in local decision making to better direct and track progress toward achieving nitrogen TMDL limits.
In plain terms, no matter how complex the watershed groundwater and surface pathways are, watershed nitrogen loading to estuaries is expressed by this formula:
Septic system loading is often the largest nitrogen source reaching an estuary. However, one of the most difficult challenges in assessing septic loading is to estimate the number of persons using the septic system and for what portions of the year. This is especially challenging on Cape Cod, where a large number of residences are used only during the summer, or only on weekends during a portion of the year. Populations in some towns or coastal villages may increase by several factors during the summer. The MEP’s solution to this problem was to use actual water bill records for each property. To account for lawn irrigation use, the MEP assumes that 10% of the billed water use is used for lawns. To estimate septic loading in a watershed, the MEP’s loading can be expressed by this formula:
[In actually, in MEP reports and loading spreadsheets, the 90% water use coefficient is applied to the nitrogen concentration, so septic loading is calculated as 23.625 mg per liter times total water use.]
In the case of watershed loading from residential lawns, the MEP typically uses 5,000 square feet times the number of dwellings in the watershed. For certain urban watersheds, where average lot may be only 7,000 square feet, this value would need to be adjusted. Similarly, for roof and driveway impervious areas on individual parcels, the MEP typically uses 1,500 square feet for each component. In some analyses, more subwatershed specific data may be used. This value may not be appropriate for urbanized watersheds, or where commercial properties with parking lots are significant, and in these cases the BBNEP has proposed the use of Mass GIS impervious coverage for stormwater planning. The best estimates of impervious are those based municipal GIS data sets if available (for example, the Town of Falmouth has sidewalks, driveways, parking lots, and road areas as GIS coverages).
Estimating Attenuation Coefficients
Modelers of watershed nitrogen loading to coastal watersheds have long recognized that while nitrate in groundwater can travel great distances without appreciable loss (also called attenuation), nitrogen losses often occur when groundwater enters streams, ponds, and bordering salt and freshwater wetlands. To calculate these losses or attenuation factors, the MEP may monitor, over a period of time, both flow and nitrogen concentration in a stream leaving a pond. Using their loading model, they will calculate loading from septic systems, precipitation, and other sources in this subwatershed, and compare this estimate to the loading calculated from measurements discharging from the pond downstream. If only 55% of the subwatershed loading flows downstream of the pond to the estuary, the attenuation factor for this subwatershed will likely be assigned a value 0.55, which can also be expressed as a nitrogen transmission factor of 0.45.
The MEP may not monitor all subwatersheds in this way, and subwatershed attenuation factors may be assumed based on past studies (usually 0.5 is used). An example of a complex watershed attenuation analysis can be found on our Wareham Subwatershed Loading Analysis page. On Cape Cod, many MEP subwatershed attenuation factors were close to 0.5 if groundwater fed streams entered a pond system, but there is some variability in the actual values determined by the MEP for each of their studies. In large riverine systems, attenuation values of 0.3 or less may be assigned.
MEP linkage of loading to water quality
The MEP collects tidal data, bathymetry, salinity, and other field data, and incorporates this information into a two-dimensional water circulation mode called RMA2. This circulation model and nitrogen loading spreadsheet data are then added to the water quality model called RMA4. It is unclear if the MEP adds any customization to RMA2 and RMA4, other than the estuary specific conditions. In various MEP TMDL reports they write, “The Linked Watershed-Embayment Model when properly parameterized, calibrated and validated for a given embayment becomes a nitrogen management planning tool, which fully supports TMDL analysis…. In addition, since the Model uses a holistic approach (the entire watershed, embayment and tidal source waters), it can be used to evaluate all projects as they relate directly or indirectly to water quality conditions within its geographic boundaries.”
Information about RMA2 and RMA4 models is at the Aquaveo-wiki.
Setting Total Nitrogen Thresholds
In developing a TMDL for an estuary, the MEP will review data on dissolved oxygen, benthic infauna, and benthic macrophytes (seaweeds and eelgrass). Often this data is collected in surveys undertaken by the MEP. While the MEP model is sophisticated with respect to modeling nitrogen concentrations within an estuary, the actual threshold total nitrogen concentrations at selected sentinel stations is based on empirical observations and best professional judgment based on the above information. Thus for setting a TMDL for the protection of eelgrass beds, the MEP adopted a rule of thumb that total nitrogen concentrations of 0.4 ppm TN or below are necessary, but specific sentinel nitrogen target concentrations may be lower than 0.4 ppm TN based on the distribution of eelgrass and monitoring of total nitrogen concentrations in nearby studied embayments. The placement of the sentinel station is also based on best professional judgment, and in the case of eelgrass coverage, areas seaward of the sentinel station can be construed as being able to support eelgrass beds if the target total nitrogen concentration is met, whereas areas up-estuary of the sentinel station will less likely be able to support it. In some cases, the placement and concentration of the sentinel station may be based on achieving some specific maximum TN concentration across a broad area of the estuary.
Similarly, for estuaries where eelgrass restoration is not the goal, the protection of benthic population from hypoxia, and the water quality target is generally selected to be 0.5-ppm total nitrogen, except for salt marsh tidal creek systems, where the target nitrogen concentration at the sentinel station is generally chosen to be 1.0 ppm Total Nitrogen. The MEP may also deviate from these targets in their model depending upon observations of oxygen concentrations and total concentrations specific to that system or similar nearby studied areas, and in the Oyster Pond, Falmouth report they identified scenarios for 3, 4, and 5 ppm oxygen concentrations.
Draft and final TMDL reports are published at the MEP, DEP, EPA, or other publicly available websites at these links:
EPA Approved TMDLs in MA
Town of Chatham Page
Town of Orleans Page
Edgartown Report at the Martha Vineyard Commission
Reports posted by Town of Barnstable
Lagoon Pond Study posted on the MV Times
Comments about this page should be directed to Joe Costa.