NTUs are a measurement of how much light can pass through a sample of liquid like the stormwater discharge from the site.  The more sediment there is in the sample, the less light can pass through and the higher the NTUs are.
 

The turbidity standard for discharge from the Marjol Battery Site was put in place at the request of Throop Borough as a real-time measurement that could be used during construction activities to determine if immediate action needed to be taken versus waiting for analytical results for lead concentrations. Since lead tends to stick to soil particles, if there was low turbidity in the discharge water, then there could be, at most, a minimal amount of lead in the water being discharged.  

There was no PADEP or USEPA turbidity standard at the time the 100% Design Plan was written. (EPA had recently proposed a standard of 280 NTUs but that standard is currently on hold; EPA suspended it in order to develop a technical basis for the value it had selected, as part of its response to a lawsuit from the National Home Builders Association.)  There also was no information available as to how turbidity levels correspond with total lead results and in turn how total lead results correspond to dissolved lead results.   In addition, we did not know what turbidity levels could be expected when measures designed to control the release of sediment like basins, skimmers, erosion control materials, etc. are used at a construction site.   

Gould researched sites where similar erosion control measures (as what would be installed at Marjol) were implemented and USEPA and PADEP approved of Gould’s proposal in the 100% Design Plan to start with 250 NTUs as a turbidity standard.  This proposal was with the understanding that once data were obtained and correlations were made between turbidity and total lead and then in turn between total lead and dissolved lead results, Gould might seek an increase in the turbidity standard if laboratory analysis showed that high-turbidity stormwater at the Site nonetheless contained very little lead.  

On the occasions when the turbidity standard for discharge from the Site was exceeded or at a minimum on a monthly basis even when the turbidity standard was not exceeded, discharge water samples were collected and sent to the laboratory for total and dissolved lead analysis.   There is no standard for total lead.  The standard for a single release of dissolved lead is 650 µg/L and 25 µg/L as an average over 3 months. The highest dissolved lead result for the discharge from the Site, even during construction, was 4.4 µg/L, significantly lower than PADEP’s standard.
 

During construction and post-construction activities at the Marjol Battery Site (July 2008 to October 27, 2010), discharge turbidity readings were collected on 99 days.  On 20 of those days, or approximately 20% of the time, the discharge exceeded the turbidity standard for the Site.  Stormwater that exceeded the Site turbidity standard of 250 NTUs left the Site on the following occasions: 

Several times from September 30 through October 16, 2010
Several times from July 19 to July 30, 2010
March 30 and 31, 2010
March 14, 2010
January 25, 2010
July 28, 2008

Every time this happened, the amount of dissolved lead in the water leaving the Site was far below the standard for lead.  Nevertheless, the level of turbidity itself exceeded the turbidity standard that had been developed for the Site.  

There are various reasons why the turbidity standard was exceeded.  Sometimes the equipment malfunctioned: In January the skimmer head separated from the skimmer arm; in late March there was a leak in the skimmer arm.  Other times, the stormwater flow was so great that the sediments did not have time to settle out.  Bottom line: Gould implemented all the Best Management Practices or BMPs (such as installation of silt socks and silt fences) it had pledged to carry out, and nonetheless the 250 NTU turbidity standard was exceeded multiple times. (The Stormwater Management page of this web site provides links for discussions on each of the dates the turbidity standard was exceeded.)

There has never been a Pennsylvania or national standard for turbidity in water leaving construction sites.  The 250 NTU standard was developed by Gould and approved by EPA and DEP specifically for the Marjol Battery Site.  It may be that this standard simply can’t be met on a consistent basis, particularly in bigger or more intense storms.   EPA had proposed a standard of 280 NTUs but the standard is currently on hold; EPA suspended it in order to develop a technical basis for the value it had selected, as part of its response to a lawsuit from the National Home Builders Association.  

Once vegetation at the Site is established, the amount of soil particles being washed away in stormwater should be much less and therefore the potential for turbid discharge should be substantially decreased.    
 

It is not uncommon for sediment to be released from active or recently active construction sites where vegetation has not been fully established during heavy rain events.  In particular, given the magnitude of the storm event that occurred on September 30 and October 1, 2010 (2% chance of occurring in any given year, and commonly referred to as a 50-year storm), sediment being discharged from construction sites was not surprising.   

The Marjol Battery Site is similar in some ways to other construction sites (where contaminated material is not involved) because it is a large construction site where there was a great deal of soil disturbance, removal of vegetation, and the need to manage stormwater during construction. However, since large quantities of contaminated materials were handled at this site, many additional precautions such as daily monitoring of turbidity during discharge, monthly sampling for total and dissolved lead, frequent use of flocculants such as Pond Clear to drop sediments out of suspension and limiting the discharge of water from the basin by flipping the skimmer to stop discharge and active pumping to closed depressions at the Site were taken during construction to protect Sulphur Creek and the Lackawanna River from a release of lead.  

According to Pennsylvania design standards, different frequency storms are used to design different parts of a stormwater system.  For example, the channels that carry water to the basin are designed to carry flow from up to a 25-year event without the water spreading outside the channel.   

The sedimentation basin at the Marjol Battery Site is designed for what is commonly referred to as a 100-year storm event, but that does not mean the basin will hold all of the water from a 100-year storm event without discharging any water.  It means that if discharge is permitted through the skimmer and as necessary, through the top of the outlet structure, the structures can handle the flow expected from storms up to a 100-year storm.  If the storm is greater than a 100-year storm or if the discharge from the outlet structure is blocked, the water level will rise until water flows over the emergency spillway.  The emergency spillway is sized and armored with stone so that it can handle all of the flow expected from a 100-year storm if the outlet structure (both skimmer and top) are fully blocked.  Water flowing over the emergency spillway would keep the unarmored sides of the basin from overtopping and possibly failing. 

Once vegetation is established across the Marjol Battery Site (possibly summer/fall of 2011), the sedimentation basin will be converted to a stormwater management basin. The bottom of the basin will be filled in so that there will no longer be ponded water, the bottom will be vegetated and the skimmer and baffle will be removed.  The stormwater that flows to the basin will then discharge freely on its own through the discharge structure.  Turbidity will no longer be measured in the stormwater leaving the site, but turbidity should  be low since vegetation will keep most of the soil from being carried away by the stormwater.

A dissolved lead sample is filtered either "in the field" (where the sample is collected) or in the laboratory to remove any solids (sediments) so that the result represents the amount of the analyte (in this case, lead) dissolved in the water. A total lead sample is not filtered, so the result represents all of the analyte in the sample (both in the sediment and what is dissolved in the water).

There is not a total lead standard for stormwater. Gould analyzed discharge samples for total lead in order to develop a correlation between turbidity and total lead and in turn between between turbidity and dissolved lead.