Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
47 Once the monitoring method, manufacturer, and instrument model have been selected, the monitoring system must be implemented. The monitoring system includes the monitoring instrument plus the equip- ment and protocols needed to support its function. Typical aspects of the implementation phase as related to the type of monitoring methods are represented in Table 6.1. In many ways, the implementation phase for on-site monitoring sys- tems is as important as the monitor selection when it comes to obtaining accurate and reliable monitoring data. The criticality of the monitoring system implementation phase is related to effects of the environment in which the monitors are used and the means by which many on-site instruments produce output. On-site monitoring methods for water-based systems typically work by measuring a response to a water characteristic in a particular sample and correlating the response to a standard. If the monitor is not exposed to a representative sample, if the sample is not properly pretreated, if the instrument is not set up to properly relate the actual measured charac- teristic to desired output parameters, or if the instrument is not func- tioning properly because of maintenance issues, inaccurate results may be obtained. The sections that follow provide guidance on the aspects of implementation. 6.1 Guidance on Design and Installation Following selection of the specific instrument, design and installation of the monitoring sys- tem make up the next critical step in obtaining accurate and representative data. It is not uncom- mon for perfectly suitable monitoring methods and instruments to produce unreliable data because the design and installation of the systems associated with the instrument are insufficient. Because the installation is integral to the proper functioning of the monitor system, this section is included to aid in the successful application of the monitoring method. 6.1.1 Sampling Location Determining locations for sampling for handheld monitors or collection of stormwater sam- ples for test kits requires consideration of access, safety, and sample representation. Implementation of On-Site Monitoring Systems C h a p t e r 6
48 Guidebook for Selecting Methods to Monitor airport and aircraft Deicing Materials 6.1.1.1 Obtaining Representative Samples Representative sample measurement is achieved by designing a sampling system and sampling protocols with verifiable results as well as implementing a representative maintenance program to ensure that the sampling collection system is operating properly. Sample collection is most representative where the sample stream is well mixed. A well-mixed stream will avoid stratification and preferential sampling that may concentrate or limit the con- stituents in the collected sample. A well-mixed sample occurs when the sample stream is tur- bulent or flow is concentrated into a small area. Examples of where turbulent and concentrated flows occur are a flume or rapids, following a bend in the flow, or downstream of a weir. Online instruments that monitor water-soluble parameters (deicer and ammonia) have filters or other solid-rejecting features. Solids in the sample stream sent to these monitors will increase the amount of maintenance required for operation. It is recommended that the samples be collected from the upper portion of a flow stream to avoid collection of solids. Water-soluble parameters mix quickly in water streams and will not stratify like solids; therefore, sample results will be accurate with reduced maintenance of the monitor. Locations where backflow can occur should be avoided for sampling, espe- cially for online monitors since there may be no way to determine that a back- flow condition occurred when samples are collected automatically. Backflow can occur near outfalls where the surface waters are susceptible to flooding, where high flows in a branch cause temporary backflow into the sampled stream, or at discharges with tidal influence. Under backflow conditions, water other than the intended sample stream is being sampled, and the results of the monitor- ing will be invalid or not representative of the discharge. Backflow can also give inaccurate flow measurements for weirs and flumes since these devices assume a forward flow. Care should also be taken not to collect a sample at the location where two flows join. Sample eddies may cause a condition where the flows do not mix well and the sample is collected preferentially from only one of the streams rather than both. Samples should be collected downstream of the junction of two flows, after the flows have had a chance to mix. Sample collection for online monitors requires special consideration. Many online moni- tors require use of a sample loop. The sample loop is a system that collects the sample from the stormwater stream and delivers it to an online monitor. If the discharge from the moni- TSS Sample Collection A sample collected from the bottom of a slow-moving stream will tend to have a higher TSS concentration because the solids will concentrate at the stream bottom. Implementation Phase Activity Monitoring Type Handheld Test Kits Online Sampling l ocation X X X Site p reparation (1) (1) X Design and i nstallation Sample p reparation X X X Utility s upply X Communication and c ontrol X Shelters X Calibration X X X Setup, operation, and maintenance Correlation X X X Maintenance X (1) Some site preparation may be required for personnel access to sample stream. Table 6.1. Implementation activities by monitoring type.
Implementation of On-Site Monitoring Systems 49 tor is returned just downstream of the sample location, downstream stormwater handling is simplifiedâparticularly if the monitoring is part of a diversion system. In this way, sample loop water can be diverted with the rest of the stormwater instead of a separate system being implemented for the sample loop flow. The sample loop usually includes a pump station structure (manhole) where pumps are located to lift the sample stream from a storm sewer to the online monitor. Flow rates for moni- tors that require filtering are typically in the 10-gpm to 20-gpm range. Flow rates for moni- tors that do not require filtering are lower. Piping of the loop should be as short as possible to reduce the time delay between sample collection and sample analysis. The piping should be sized for high velocities so that biofouling does not occur within the piping. One airport surveyed installed sample pumps that can reverse direction and eliminated check valves so that the sample lines can be back-flushed if they plug. Allowing the sample line to back-flush each time the pump stops has reduced the maintenance and failure of the sample loop systems. Some TOC monitors use a compressed air blast to clean the filters in the sampling loop. One airport surveyed was experiencing corrosion issues in the sample shelters in which TOC moni- tors were installed. The airport theorized that the compressed air blast was stripping chemicals (including sulfides) from the sample water and the air was escaping into the TOC monitor cabi- net, causing the corrosion. Airport personnel installed air-release valves on the sample loop and discharged the air to the outside of the sampling shelter. Corrosion in the sample shelter appears to have decreased since the change was made. Handheld meter accuracy is most reliable if the probe can be fully submerged directly in the flowing stream. 6.1.1.2 Access to Sample Locations Sampling of deicer-affected stormwater will typically occur during winter, and snow or ice may make access difficult. An appropriate health and safety plan describing appropriate appara- tus and conditions should be provided for any sampling staff. Typical probe cables are 4 ft to 6 ft in length, and this length should be considered in selecting the sampling location. Grab samples may be collected and probes inserted into the sample if close proximity to the water surface is not possible, such as when sampling from a deep storm sewer. Grab samples for test kit methods are easier to collect from above a sample stream than from a horizontal position. A bailer, dipper, or swing sampler is typically used to collect a sample and transfer it to sample bottles. Bridges or sewer pipe discharges readily provide access above the flow stream. Safety regulations may require that railing be installed at locations where an elevation drop of more than 4 ft exists. Safety experts should be consulted for specific location requirements. For locations where online monitor probes are installed, access similar to that needed for handheld monitors should be included. The online monitor probes require periodic cleaning and calibration, and personnel will need access to perform these functions. Cable lengths for online monitors can be specified longer than the 4 ft to 6 ft typical of handheld meters, so the platform elevation relative to the water level is not as restrictive as for a handheld meter. Online monitors should be visited daily, so easy access to the sampling location by airport personnel, particularly in winter weather conditions, is a must. The sampling locations may be inside or outside of the security fence and remote from other airport access. Therefore, consid- erations for routes to be traveled, gate access, and plowing of roads should be addressed for each sampling location.
50 Guidebook for Selecting Methods to Monitor airport and aircraft Deicing Materials 6.1.2 Sample Preparation Samples measured using on-site devices may need proper preparation or conditioning to increase the reliability of the measurement and to protect the instrument. Guidelines for sample holding times for test kit analyses should be followed. If guidelines do not exist, holding times for similar standard methods should be adopted. Samples entering online meters can create conditions that affect the instrument and hence the accuracy of the instrumentâs measurement. Deicer compounds provide a food source for nuisance bacteria in stormwater streams. Biogrowth in the sampling system is a particular prob- lem for online monitors. The biogrowth occurs on the pipes and tubing walls and plugs the small-diameter sample tubing inside the monitors. Ongoing maintenance of the instrument to prevent biogrowth can be time-consuming and expensive. An innovative approach to preventing biogrowth in online monitors has been developed by a surveyed airport for their two TOC monitors used in the deicer management system. A dilute chlorine solution is continuously injected into the sample stream to the online monitor. The chlorine solution (fed as a 0.8% bleach solution) is sufficient to prevent biological buildup on the online monitor filter and limit biogrowth in the tubing of the monitor. Split-sample testing results indicate that the TOC monitor measurements are not affected by the low chlorine dosage. The sample loop flow rate is much smaller than the average stormwater flow, and the chlorine is diluted below active level in the stormwater discharge. Chlorination of the sample system cannot be used for BOD monitors because the chlorine will harm the bacteria used for the measurement. If chlorination of the sample system is used for preventing biogrowth in ammonia monitors, split samples should be collected to verify that the chlorine is not oxidizing the ammonia and affecting the results. Automatic cleaning systems can be installed in monitoring systems that analyze samples internally or on probe systems. Probe systems such as for pH or temperature can be equipped with water-jet systems that remove biogrowth from the probe face. Potable water must be avail- able at the sampling site for jet cleaning systems. DO probes are equipped with wiper systems; however, in conditions with biogrowth issues such as deicer stormwater streams, wiper systems are not as effective because the biogrowth will coat the wiper. DO monitor systems with optical/ fluorescence probes may be equipped with water-jet cleaning by some manufacturers. If potable water is not available at the sampling site, compressed air may be substituted for the water jet. 6.1.3 Utility Supply All of the online monitors require electricity, and many require a water connection. Providing utilities to the sampling site may be as costly and complex as the installation of the monitoring instruments. Locations where online monitoring occurs, near stormwater manholes or discharges, may not have existing electric utilities nearby. Although many of the online instruments only require 120-volt power, ancillary equipment such as sample pumps and shelter heaters may require 240-volt or greater power. Because of the power requirements for the ancillary systems and because of the long distances the power may be fed, it is recommended that the power be sup- plied at a higher voltage (480 volts or 240 volts) and transformed at the sample location. Some jurisdictions require electronic equipment to be UL listed prior to installation. Some monitoring units may not have UL listings, especially those manufactured in Europe. It is rec- ommended that electrical code requirements be confirmed and the applicable certification be verified with the manufacturer.
Implementation of On-Site Monitoring Systems 51 Potable water supply is required for some online monitors and is required for most automatic cleaning systems. Some monitors do not require a continuous supply of potable water, and the water requirement may be supplied by a local tank that is refilled by airport personnel as needed. If the sampling location is remote from the terminal or other airport buildings, getting potable water to the sampling site may be challenging. If water cannot be supplied to the sample site, choices of monitoring instruments that require a water supply would be eliminated. Ancillary cleaning units that use water may also have options that use compressed air. For the systems that use compressed air for cleaning, an air compressor and additional electrical power replace the water connection. 6.1.4 Communications and Controls The ability of online monitors to provide significant amounts of real-time data is a key to their usefulness in stormwater and deicer management. Full utilization of the real-time data capabili- ties may require development of systems and protocols for communicating the data to airport staff as well as stormwater and deicer management control systems. Real-time communication of the results can be required for: ⢠Control of equipment that performs stormwater diversion, ⢠Remote observation by personnel to track data and diagnose issues, and ⢠Control of loading of the stream flow. 6.1.4.1 Communication Method Considerations For users to take advantage of the data provided by online monitors, the monitors will need to communicate with a local airport communication network. The most common method to transfer information from a monitor to a control system is via 4-mA (milliamp) to 20-mA control wire, referred to as analog communication. Analog communication is limited in the resolution of the data that can be transmitted. The resolution for most communication systems is approximately 4,000 units. So for a monitor that has a 0-mg/L to 20,000-mg/L range, the control system will only have a resolution of approximately 5 mg/L (20,000 divided by 4,000). Electrical noise in the system will decrease the resolution even further. If the control system has to record the instrument reading with more precision, more expensive digital communication must be used. Other than data results, online monitors can communicate alarms or other events through various types of communication protocols. The protocols for this kind of communication should be considered based on the complexity and needs of the control system. Monitoring and record- ing of events such as loss of sample, calibration error, or other events will alert airport personnel to perform maintenance and diagnose potential issues with sample results. Some airports have issues with lightning inducing high voltages in copper wires buried near the ground surface and destroying communication equipment. Fiber optic communication wires are recommended for locations that have potential for lightning strikes. 6.1.4.2 Control Method Considerations If the monitoring data are used for control of stormwater or deicer management system equipment, the following considerations should be assessed. 6.1.4.2.1 Diversion Considerations. Systems that divert stormwater rely on real-time data to make control decisions. There may be an inherent assumption that the sample results are continuous and that diversion can be made very quickly. Some online monitors have an adjust- able parameter that allows the time between sample collections to be selected by the operator. Most online monitors do not provide instantaneous readings but rather require some time
52 Guidebook for Selecting Methods to Monitor airport and aircraft Deicing Materials for the analysis to occur. The analysis period can range from a few seconds to as much as 20 min. The analysis period may be a defined interval or a variable period necessary to reach a stable measurement. With some instruments, the output of the monitor may appear to be continuous, but users are cautioned to understand the analysis interval since this may affect interpretation of results. Both sample collection times and sample analysis times will delay stormwater concentra- tion information from being used to make a diversion control decision. Increasing the sampling frequency may increase the reagent usage or power usage from that quoted by manufacturers. 6.1.4.2.2 Considerations for Treatment Influent or Effluent Monitoring. When consid- ering the method for monitoring treatment system influent and effluent flows for BOD, COD, or TOC, the effects of the treatment process on stormwater should be considered. The treat- ment process may affect both calibration and correlation factors, including the relationships among PG, BOD, COD, and TOC. Therefore, instruments at untreated streams need to be set up differently from instruments at treated streams. The specific settings are determined on a site-specific basis. In cases where the stream characteristics do not affect the calibration and correlation setup for an instrument, a single online instrument could be used to analyze samples from multiple locations. Collection of multiple samples is accomplished by manifolding the instrument intake to alternate between treatment system influent and effluent samples. One unit may not be appro- priate for BOD monitors. Alternating between influent and effluent samples may not yield accu- rate results for BOD monitors and biological treatment systems because different compounds are present in the influent and effluent samples. The effluent sample will contain compounds that are resistive to biological degradation. The short contact time of the BOD monitors may not allow sufficient time for the bacteria in the instrument to acclimate and degrade the effluent compounds. When alternated with easily degraded influent samples containing deicer, the bac- teria may not adjust to less easily degraded compounds in the effluent. No airports are known to have tested alternating influent and effluent samples using a BOD monitor. 6.1.5 Combining Sampling with Flow Rates for Accurate Loads Load is determined by multiplying the parameter concentration by the flow rate and a con- version factor. If load information is required, both accurate flow and concentration data are required. Ideally, flow and concentration data are multiplied together in short time steps to accu- rately track the load. Data from most types of flow monitors are available nearly continuously, so the short time duration is not an issue for flow data. Because sample collection and analysis times vary significantly for online monitors, the frequency of the concentration data is usually the limiting factor. The flow data should be averaged over the same frequency as the interval between concentration results to achieve the best accuracy. It is generally more accurate to monitor loadings of stormwater downstream of a storage tank or basin. The concentrations will not vary as greatly as upstream of the storage. The concentra- tion will not change as significantly during the sampling interval, and therefore the loading esti- mate will be more accurate. Sample intervals may also be extended for large storage basins with permanent storage because the basins will provide continuous equalization. 6.1.6 Recommendations for Equipment Shelters Equipment shelters are recommended for online monitors that will not be installed inside existing buildings. Preconstructed shelters can be purchased with integrated lighting, ventila- tion, and power. Venting of the stormwater lines should be directed outside the shelter to avoid potential corrosion and issues with health and safety. Potable water, if available, should be pro-
Implementation of On-Site Monitoring Systems 53 vided to aid in equipment cleaning or preparation of solutions at the site. Heating should be provided to provide freeze protection since some solutions may have minimum recommended storage temperatures. In warm climates, air-conditioning or temperature-controlled ventilation may be necessary in the summer to prevent damage to electronic equipment. Local occupancy regulations should be consulted to determine additional applicable requirements. It is recommended that the shelter be sized slightly larger than what may seem necessary in order to account for needs such as reagent storage or possible additional equipment such as automated samplers. 6.2 Guidance on Setup, Operation, and Maintenance Proper selection of the monitoring method in addition to proper design and installation of the instrument and support systems will greatly increase the chances of accurate and reliable measurements. These steps alone are not sufficient, however. It is critical that the on-site moni- tors be properly set up, operated, and maintained. Proper setup, operation, and maintenance of handheld and test kit monitors can typically be accomplished by following the guidance in the instrument manuals and consulting with the instrument manufacturer. Setup, operation, and maintenance of online instruments, however, often requires going beyond the guidance given by manufacturers and making adaptations to procedures that are specific to the individual airportâs stormwater and deicing environments. 6.2.1 Calibration Calibration is defined as the process by which the monitoring method is adjusted to achieve accurate measurements of the parameter within the expected range. It is important to remember that the response the method uses to determine concentration is only indirectly related to the concentration of the parameter(s) of interest. Since the actual chemical is not positively identified and concentration is not directly measured, as they are in analytical laboratory measurements, it is important to minimize issues that may interfere with the monitoring process. During operation, calibration and calibration checks will determine if the monitoring method will accurately measure the parameter concentrations. Operators should not take for granted that calibration solutions are correct or that automatic calibrations are being per- formed properly. The steps to be considered for proper calibration of online monitors are listed in the following. 6.2.1.1 Selection of Calibration Compound The selection of the calibration compound is very important to the calibration process. The calibration compound does not necessarily need to be the chemical of interest, but the calibra- tion compound should give a response similar to the chemical of interest. Manufacturers will recommend compounds based on response and stability. For example, TOC monitor manufac- turers typically recommend potassium hydrogen phthalate (KHP) because this compound is not biodegradable and the solutions therefore have good long-term stability. KHP will not work for calibrating BOD monitors because the compound is not biodegradable. Selection of the calibration concentration should be made based on the expected range of measurements. Laboratory instruments will use the calibration result and a blank (or automatic zero point) to determine a correlation line. If the calibration concentration is higher than the typical sample concentrations (i.e., a pure propylene glycol solution), the instrument response
54 Guidebook for Selecting Methods to Monitor airport and aircraft Deicing Materials may not be the same as in the concentration range where samples are typically measured. The same is true if the calibration concentra- tion is lower than the typical sample concentrations. If a calibration compound other than one of the primary deicer compounds (e.g., propylene glycol) is used, the concentration of the calibration com- pound in solution should be determined so that the resulting param- eter concentration (COD or TOC) is within the range of expected measurements. Diluted deicer and propylene glycol solutions may biodegrade if stored in open containers at room temperature. If deicer or propyl- ene glycol is used as the calibration solution, the calibration solution should be tested in an analytical lab periodically to confirm its con- centration. Solutions that have degraded should be replaced. Even if a different calibration compound is used, it is good practice to peri- odically have a laboratory analyze calibration solutions to confirm their concentrations. The frequency of calibration solution analyses can be decreased for compounds that have demonstrated long-term stability. If the monitor performs automatic calibrations, the raw calibration data should be reviewed to determine the accurate functioning of the instrument. Most monitoring instruments will provide a response value, which is the unadjusted measurement made by the instrument. If the response value is typically within a narrow range, the instrument is consistently measuring the calibration solution in the same way. If the response value changes, the instrument is drifting and the instrument response is inconsistent with the correlated measurement. A drifting response value is usually an early indication of instrument failure, and maintenance procedures should be performed to determine the cause of the drift. Operators should also periodically perform a manual calibration check on an independent sample to verify that the instrument is measuring accurately. The calibration check is performed on a sample with a known concentration, created separately from the calibration solution. The calibration check sample may be a single compound in laboratory-grade water, but it is recom- mended that the sample be created by spiking deicer (or pure propylene glycol) into a storm- water sample that is known to not contain deicer. 6.2.2 Maintenance Maintenance of online monitors and their support systems in a deicing environment is critical to successful operation and requires a commitment of time and knowledgeable personnel. Pre- ventative maintenance should be performed in the non-deicing season to ensure proper opera- tion during the deicing season. Preventative maintenance should include examination for worn parts and replacement of parts that are near the end of their useful lives. Typical maintenance activities include: ⢠Checking for functioning of equipment (pumps, valves), ⢠Replacing worn or broken tubing, ⢠Cleaning of biofouling, ⢠Checking and replacing calibration standard fluids, ⢠Performing off-season preventative maintenance, ⢠Replacing of malfunctioning probes, and ⢠Preventative maintenance for extended non-deicing season shutdown. Steps for Proper Calibration of Online Monitors 1. Select calibration compound based on method and parameter of interest. 2. Select the calibration solution concen- tration based on the instrument range. 3. Analyze the calibration solution peri- odically by a laboratory. 4. Review the automatic calibration data to determine if the unit is drifting. 5. Periodically analyze calibration check samples.
Implementation of On-Site Monitoring Systems 55 6.2.3 Monitor Site Visits Online monitors should be visited once per day to verify proper operation. Since monitors will be visited frequently during the deicing season, easy access should be provided. During a site visit, the following items should be verified: ⢠The monitor has power and the system is functioning; ⢠The previous daily sample results agree with expectations (deicing events have higher concentrations); ⢠Automatic calibrations have occurred, and the response value is typical; ⢠Auxiliary systems are functioning; ⢠Sample loop flows are within an acceptable range; and ⢠Reagent or solution levels are okay for the next operating period. In addition, calibration checks should be periodically performed.
