Stormwater Management Facility Sediment Efficiency Monitoring Using Innovative Forecasting and Sampling Techniques (Part 2)

In part one, the ways in which stormwater management facility sediment efficiency monitoring could be improved were examined. Part one covered:

• Flow monitoring
• Rainfall monitoring
• TSS Sampling
• TSS Removal Efficiency Analysis
• Sample Size
• Sampling Locations

Part two will examine the grab sampling methodology and the benefits for municipal wastewater management when using a semi-automated process for collecting total suspended solids (TSS).

Grab Sampling Methodology

There were three options for TSS water sampling:

Manual Grab Sampling:

Using field staff to obtain grab samples during a 25 mm storm event. This would incur many manhours to achieve the required sampling protocol. For example, a 36-hour sampling period would require 72 manhours of sampling time plus travel time to the site multiplied by six events. The feasibility of this option as well as budgetary limitations were reasons for not selecting this option.

Fully Automated Samplers:

This type of equipment has various options for measuring flow/level in addition to taking samples during storm events. This option was certainly considered, due to the nature of the wet-weather sampling program. However, the purchase of this type of equipment has a high capital cost that would lead to high equipment rental costs for the client. Budgetary limitations were the main reason why this option was not selected.

Simple In-house Auto-Samplers:

The auto-samplers used for this project were built in-house by Civica’s sister company, Smart City Water Inc. (SCW), for an approximate cost of $2-3k. This equipment utilized automatic controls to operate a pump and valves to collect six (6) composite one (1) litre samples through a secured sample hose and coarse strainer at predetermined sample times. The autosampler equipment reports the sample ID and sample time, which were compared to the flow monitoring data for preparation of flow-weighted composite samples.

Using the RDA software, qualified storms were forecasted 48 hours in advance. A detailed sampling plan was prepared as per the forecasted rainfall hyetographs before each qualified event to identify the best time to begin taking samples and the sampling interval. The start time and sampling interval were preprogramed into the eighteen (18) autosamplers and were deployed in the field, at the inlet and outlet of each pond, prior to the start of the event.

During the event, the autosamplers will turn on and take samples at the predetermined times, with a backwash in between each sample. Following the conclusion of the storm event, the autosamplers were retrieved from the field, and the samples and sampling logs were recovered from the units for analysis.

Benefits of This Semi-Automated Process for Collecting TSS Samples

The implementation of this semi-automated process for collecting TSS samples provided the following benefits when compared to manual grab sampling:

• Lower Cost: When compared to having two field technicians in the field for a 36-hour drawdown sampling period, this option has a much lower cost.

• Consistency in Sampling Method: For each monitoring point, the sampling location and procedure will be consistent across all events, ensuring that the results of one storm can reliably be compared to another.

• Consistency in Sampling Times: Because the samplers are pre-programmed, the sampling start times and intervals were consistent across all ponds, unless modifications needed to be made for site-specific conditions noted through previous event monitoring.

• Access to Sampling Locations: The equipment was placed prior to the start of each storm, so there is no risk that heavy flows will limit access to the ideal sampling locations.

• Staff Safety: There are greater hazards associated with accessing these ponds due to their proximity to the Red Hill Valley Parkway. These traffic hazards would be exacerbated by large rain events occurring at night. This methodology maintains the same quality of data collection while minimizing the risk to field staff.

The implementation of this semi-automated process also had benefits when compared to a fully automated sampler in that it reduced the overall cost of the sampling project. The in-house autosamplers, area velocity flow meters, and need to perform a site visit prior to the rain event to deploy the equipment do come with their own costs. However, the savings for a sampling program that requires that capture of six storm events were estimated at approximately 10-20%.

Key Takeaways

Through this project, a number of lessons have been learned, including:

1. Innovation and ingenuity have helped the project team save time and money for this project. The in-house autosamplers saved many hours of staff time compared to a manual grab sampling program and reduced cost when compared to traditional automated samplers.

2. Requiring a storm size of 25 mm for sampling will likely require more than 12 months to complete as these storms only occur on average 4-5 times per year for this study location. Environment Canada historical climate and rainfall data can be leveraged to help make informed decisions for how long the sampling program may need to last to obtain the desired number of events. At this time, five sampling events have been captured between June 2020 to July 2021. Winter shutdown of operations occurred between December and the end of March, when temperatures were not reliably above zero Celsius.

3. Thunderstorms are difficult to predict both in their timing, amount of rainfall and track of the storm. One event was missed entirely due to a pop-up thunderstorm. However manual sampling would have missed this as well.

4. One rainfall event large than 25 mm was missed due to accidental programming of the autosamplers for the wrong date/time. Operations manuals and code QA/QC are crucial for programming autosamplers. 

5. Incorporation of global climate models and using forecasts are an effective way to assess and plan an auto sampling schedule. However, global climate forecasts are prone to high levels of uncertainty, especially when trying to plan your sampling schedule well in advance (e.g., 12-24+ hours).  

6. Monitoring programs, especially those for a sized-event trigger, are difficult to predict as a lack of rainfall can occur during dry years. Suggestion is these programs should be priced on a monthly basis, with an upset limit rather than a set number of events (in this case six events). 

Contact Civica Infrastructure for Stormwater Management Services in Ontario 

Civica leads the industry in municipal wastewater management solutions and water flow monitoring systems. We offer stormwater management pond analysis services, storm water drainage analysis, and more. Contact us today for a free consultation.

Learn More At:

Stormwater Management Facility Sediment Efficiency Monitoring (Part 1)

Impacts of Existing Storm Drainage Design Standards in Ontario on Sanitary System Capacity (Part 1)

Impacts of Existing Storm Drainage Design Standards in Ontario on Sanitary System Capacity (Part 2)

Methodology for the Identification and Quantification of Sanitary Maintenance Hole Inflow and Infiltration (I&I)

Key Insights Into The New Systemwide Stormwater ECA Regulations