Continuous Water-Quality Monitor Deployment Techniques

So I’m sitting here with John, and we were asked to give an overview about our experience working with continuous water quality monitors And as John and I put this talk together, we both reflected that we work on a number of monitoring projects that often have very similar objectives Understanding what the current water quality conditions of the system are, understanding how those conditions are changing over time, and then addressing why those conditions may be changing over time But answering those questions begins with collecting data in the field But answering them really well often requires collecting frequent, high quality measurements, and that’s a task that’s really well suited for the use of continuous water quality monitors So in this talk, John and I will share the techniques that we use to most effectively deploy continuous water quality instrumentation And through the talk, we’re gonna focus on collecting, or how you can collect high quality data, while reducing site maintenance needs So before I begin, there’s just two quick disclaimers to make The first is that through this presentation, there’s gonna be a lot of pictures of different instruments being used in our network, but it doesn’t apply any endorsement for those products And secondly, more relevant to this talk, is that John and I are sharing experiences that we have working in streams and rivers So these instruments can be used in groundwater wells, in lakes But this talk doesn’t really focus on deployments in those settings Additionally, our experiences are in non-title systems So issues like bioaccumulation, we’re not really gonna focus on much And some climates were freezing is a common problem Again, that’s not really issues that we’re gonna address specifically through this talk Some structure of today’s presentation, just begin with a little background about definitions of continuous water quality monitoring, why it’s used, how it’s used We’ll talk about our opinions on what makes a deployment effective What are we trying to achieve when we put these instruments out into the stream And then talk a bit about the different techniques that we use to deploy these instruments So how can they actually get into the water? In the final section, it’s gonna be some information about techniques that have been used to advance the basic deployment options So there’s some issues at sites such as sediment fouling, working in really shallow water that you can address by taking these common approaches and expanding them beyond the basic deployment So there’s one message to this presentation that I hope becomes very clear throughout the examples and the slides that we talk through, and that is that effective continuous water quality deployments allow instruments to consistently and accurately collect measurements with minimal intervention And the method and maintenance of deployment has a large impact on data quality So begin with just a little background Continuous water quality monitors or devices that provide frequent, near real-time measurements of water quality constituents So, that contrasts from what we think of as traditional water quality measurements, which are made by having someone physically visit a location, fill up a bottle of water, bring it back, do some processing on it, submit it to a lab, wait for results And that can be done at intervals, monthly, or you can target storms, or you can tire yourself out and go out there everyday and do that kind of work And you can see the record of data that those traditional measurements produced This graph shows a time series plot of stream flow in the blue color And then if you target storms or target monthly events, you get these pink dots representing nitrate concentrations given from the lab Continuous monitors allow us to fill this record out much more completely They’re used because water quality can change frequently over time which necessitates frequent, repeated measurement to adequately characterize variations in quality So you can see now, using an instrument that measures nitrate every five minutes, every 15 minutes, at whatever interval you’d like to set, you can see a lot more of the dynamics in this time series plot of nitrate So just by the discrete samples, you lose a lot of the variability without having to monitor in stream And these monitors are being used all throughout the country for a variety of purposes This is a map of USGS spring and stream monitoring locations that are measuring at least one continuous parameter So some of these stations may simply just be collecting water temperature on a real

time basis, whereas other locations may be collecting a full suite of constituents, conductivity, pH, dissolved oxygen, turbidity But the work is being done all throughout the country in a number of different settings And the use of these monitors is increasing through time Those increases are occurring because the technology’s becoming a little cheaper, where we can measure a broader range of constituents The applications are being realized about how valuable this work is So you can see the number of USGS monitoring locations in 2006, there is about 1300 stations with continuous data And through time, that network has grown to about 2100 different stations So as I mentioned, there’s a large and a growing list of constituents that can be monitored in the field So just a little overview about what we can do with these instruments We can make measurements for constituents that are now kinda thought of as traditional water temperature, conductivity, pH, dissolved oxygen, turbidity All those sensors can be stuck on one instrument, a water quality sonde manufactured by a number of different companies There’s more probes that can be stuck on these things, dissolved organic material, ammonium, total algae There are standalone instruments just to monitor nitrate in the streams, instruments that can monitor phosphate concentrations There are more recently developed fluorometers that are packaged in a way where the instruments can be deployed and left unattended that measure constituents like oil, optical brighteners, and tryptophan, are constituents that we think of as a human waste water signature And that list is continuing to expand through time One consideration when you’re thinking about what instrument will be best suited for your project is that these instruments all have different deployment requirements and measurement limitations So you should think carefully about which is going to be best to suit your project needs and your site needs So some of these instruments can’t work in freezing temperatures, some of these instruments need to be deployed at a fixed depth or at a fixed angle And some of these instruments are more sensitive to fouling from sediment or bioaccumulation than others So all considerations to keep in mind And to add to that, although you see on the screen a variety of sensors and manufacturers, one consideration in your objective for your monitoring program, if you’re monitoring down a well, that should be a consideration, as well as what your overall water quality objectives are So as we continue to discuss the water quality installations and fouling, before you even get to the field, before you even consider your installation type, a lot of thought should be put into the consideration of your equipment type, your monitoring frequencies, and just the overall plan for maintaining and operating these sensors 08:16 S1: So why are we doing all this work? Why is it worth all this investment? It’s ultimately because these continuous monitors advance our scientific understanding of stream processes, stream conditions, there’s a lot of merging applications for these instruments Just a couple examples that we can use these instruments to improve our load estimation of various constituents by developing aggression models We can use these instruments to better characterize spatial water quality dynamics, and temporal variability in a watershed We can look at impacts from anthropogenic activities We can pair these monitors upstream and downstream and look at the effects of an activity occurring, and ultimately, the investment that’s made in these instruments, because they’re expensive They take a lot of work to maintain But we’re trying to maximize the scientific output and ultimately, that output is gonna by diminished by a reduction in data quality So we wanna make sure that we have a really strong deployment from the get-go, so we can ultimately produce these really quality products at the end of the project So what is an effective deployment, what do we mean by an effective water quality deployment? Well, firstly, an effective continuous water quality deployment prevents a loss of data, and it does that by meeting the power data collection and data relay requirements of all the instrumentation deployed in stream So most of these instruments, a commonality is that they’re powered using the solar panel and some sort of battery We often don’t have AC power at some of our remote monitoring locations They’re then hard wired from a monitoring location back to a data logger, so that the

measurements that are made on the instrument can be stored on the data logger And then the information stored on that data logger is transmitted to the internet some way via cell modem or with a satellite transmission, so that the information is provided in near real time on the internet So anybody can go in and access and look at the conditions in the stream One thing to add here, in your selection of the proper water quality instrumentation, if you feel that your site will limit your access or if you’re having different monitoring inspection time intervals, you might wanna consider an instrument that has internal logging capabilities or an instrument that is internally powered Majority of our sites we like to get the data in near real time, but to also limit some of the data losses If there is ever any, we do like to choose instrumentation that has internal logging capabilities as well So again, just to add here, in your process of thinking through your installation, thinking through your maintenance of your water quality equipment, consider how you’re gonna power your equipment from the get-go, consider how you want your data transmitted, and then what interval you want that data to be transmitted at Yeah, and ultimately, what John’s getting at there, the theme throughout these slides is that we don’t wanna lose data And that begins with making sure there’s power to the instrument, making sure you’re recording the measurements and making sure they’re transmitted appropriately at the interval that you expect them to be So another element of an effective deployment is that we prevent the loss of data by minimizing the frequency and severity of instrument fouling So we’re leaving these monitors out in a stream, we’re leaving them unattended, so they’re going to become dirty from sediment, from debris, from organic material in the environment, and they require routine servicings so that an individual can come, clean those probes off, clean those instruments off, and maintain that we are collecting high quality, accurate data We also use these servicing visits to address the calibration drift of deployed instruments So we take out our standards, we measure, is pH reading what we expect it to read If it’s not, we can recalibrate the instruments, we follow specific guidelines for these servicing visits using documentation developed by the USGS and there are different criteria for these different constituents to say, “Okay, if conductivity has drifted or has becomes fouled by a certain amount or a certain percentage, then we need to remove data from the database”, and that’s what we want to avoid The fouling or calibration drift that exceeds these criteria require our measured data to be deleted We wanna minimize how often that happens The first thing we can do to minimize the amount of fouling that an instrument experiences during its deployment is to select a deployment location that gets the instrument out into flowing water Not only does flowing water often contain the most representative water chemistry of the system, but you allow the instruments to kind of self clean There’s less chance that sediments can accumulate on the probes and foul your measurements We wanna find a location that places the instruments far away from stream banks, which can erode and foul the instruments, and we wanna place the instruments in a mechanism that keeps them off of the bottom of the stream Again, we’re trying to reduce the amount of fouling that these instruments experience from sediment build up And when you’re making these considerations, sometimes the best deployment location is not immediately adjacent to the existing stream gauge and requires a separate data logger, a separate power source, but it’s an important consideration and it’s often worthwhile to make that sacrifice to find a location that’s going to provide the maximum data quality even if it requires a little bit more effort for the installation All time that you invest at this point, is time spent on site reconnaissance, results in less data loss and site maintenance over the lifetime of the deployment So this reconnaissance is really a crucial part of an effective deployment Maybe one of the most important components of a water quality deployment is constant refinement through time This idea of trial and error, that we can make modifications to an existing deployment to reduce instrument fouling, and just a couple examples of that This is a time series graph of nitrate collected from a continuous monitor, where we have data collected through time, and the periods in red are measured data that we have deleted

from the database because the instrument was fouled for one reason or another We wanna avoid these periods of red We wanna maintain accurate, high quality data through the life of the deployment So the issue here was that our instrument was deployed in a PVC tube, which ended up trapping a lot of sediment and consistently fouling the nitrate data We made a modification to this site, and we removed the plastic case that this instrument was deployed in to reduce the frequency which it became fouled So you can see the resulting record here is much more clean Another example of a site refinement, this is a turbidity record through time Oftentimes, at some of our sites, we’ll have a nice big storm event, we’ll collect data from the storm, but then as the stream recedes back to base flow, fine sediment deposits on top of the instruments And often, we get these fouled turbidity records at the end of a storm event So in order to account for that, we’ve installed a number of bilge pumps near our instrumentation that are used to clean off the sondes by pumping ambient water across the face of the probe So that can clean off some of that fine sediment that’s accumulated on the instruments You can see some examples of storms after a bilge pump was installed and a really clean turbidity record I have a couple more examples and information about bilge pumps later in the talk Throughout our continuous water quality monitoring networks, we’re typically maintaining about greater than 90% of the data that are measured out in the field, which means we have to remove or have lost less than 10% because of fouling or instrument failure So it’s important to consider through the life of your deployment, if you find that you’re having to delete out months and months of data on an annual record, it’s time to stop and think about how that deployment can be modified and improved Because that’s not meeting the objectives of why the instrument’s out in the field in the first place In addition to refining the deployment to reduce the amount of instrument fouling that occurs, you’re also going to reduce the number of site visits that are needed to maintain that high quality deployment Another example of a turbidity record before we installed a bilge pump, not only do you see the time series data here and you see periods in red that were fouled, which required the removal of measured data from the database You also see on this plot, blue diamonds that represent the number of times an individual from the office physically had to go on site and clean off that instrument And this is about a three month window on screen here and we have technicians visiting that site about once every other week, to de-foul the sonde So we installed a bilge pump on this site, and not only did we preserve a better, cleaner turbidity record, we reduced the number of times that an individual had to go out and clean off that instrumentation So it’s really ultimately a cost savings at the end We can re-prioritize individual’s time so that they’re not making drives to go and clean off our instrumentation And it’s important to remember that there’s always going to be scheduled visits necessary to properly maintain this instrumentation, but we’re trying to minimize these unscheduled visits The times that we come into the office and see that the records don’t look accurate, so you know what, your day’s gonna be getting in the car, cleaning the instrument off We wanna have a number of days where the data look nice and clean, and we’re just going out on regular fixed frequency intervals And again, just like we’re trying to maintain a large majority of the data that are collected, we’re trying to reduce these unscheduled visits So we find that our deployments that are running really well, we have an unscheduled site visit about less than once a month, so it’s another time to stop and consider, “Can I improve my deployment? Is someone going out there every week, every few days to clean off the instruments? Maybe there’s something else that could be done to improve the data quality.” As Jimmy and I prepared for this presentation and we got to discussing our common usage of bilge pumps and air compressors to help with maintaining the fouling at locations, one of the key things that continued to arise was the need to make sure that we consider the timing The use of bilge pumps, just to be clear, is we’re using the native water to suck in and flush the in situ quality monitor Well, depending on the parameters that you’re currently monitoring, you wanna ensure that the timing of that is not affecting the water quality measurements So for instance, if you’re going to be using an air compressor to flush a sonde or even a bilge pump, you wanna ensure that the water quality monitor has enough time to equilibrate in between those measuring times So in Metro Atlanta, when we’ve used the bilge pumps, for instance, we record

our data about every 15 minutes So we’ve ensured that we have at least 10 minutes before a measurement is going to be taken to ensure that the in situ water quality sonde has enough time to get back to ambient conditions Otherwise, when you start using creativity, which please understand that we’re definitely saying sometimes you have to be creative in your installations to ensure that you’re maximizing your instrument, your data longevity What we’re really saying is that sometimes when you’re creative, you just wanna make sure that you take the time to factor in all considerations So if you’re going to consider using a bilge pump, also consider that if it’s going to effect any of your monitor parameters So all these things in mind, we talked about minimizing data loss, minimizing site visits, but there is some additional constraints on your deployments, you need to place these instruments in a location that accurately represents stream conditions A great deployment location might be right on the side of the channel, but man, if that water chemistry’s completely different from the rest of the cross section, you need to think about it Is that the right choice for your project to collect data from that location? We wanna make sure that we can collect data during all flow conditions, streams that are very shallow during base flows, we wanna make sure those instruments are underwater and that we’re not gonna lose them during high flow events We wanna make sure because there’s always routine servicings and site visits needed that personnel can safely and easily access the instrumentation, that the instance was safe from debris or from vandalism that may be possible around the sites And then there’s always this wild card of additional partner needs, right, you may be asked to put the instrumentation in a set location or the deployment needs to look a certain way, so it’s something else that needs to be kept in mind when you’re considering, “How am I gonna get these instruments in the water and collect a really high quality data record?” Now, we’d just like to step through some more specific examples of common ways that we’ve deployed our instrumentation and just some details of those deployments on how we’ve made them very effective for the work that we do John mentioned creativity earlier, and it is a really good time to be creative when deploying these instruments, ’cause you have so many options of how they can get in the water Instruments can be deployed off of a bridge, on the bottom, on a bracket, on a boom, in a bucket, in a borehole, on a buoy, in a backpack You have so many options for how you can get these monitors in the water, and it really allows you a lot of opportunities to decide what is the right choice for my site, for my project? There is no right choice There’s no single recommendation that we’re gonna make to you today Hopefully, we’re giving you some ideas of things to think about and consider, so you can make the right choice for your project needs One thing we will stress that having that consideration of what’s right for me, realize that streams are dynamic, you may install your monitor during the summertime, and it works for that season Don’t be surprised that things may occur, something may happen in your basing, your watershed may change, you may have to adapt and change that installation a little bit later in the season after you continue to watch and learn your watershed So yes, it may start off right, but don’t be surprised, sometimes they will surprise you and you may have to adapt as you continue to watch your location I wanna share more detailed examples of some of the deployments One of them that we use pretty frequently, as a suspension style approach, working off of a bridge, which may be appropriate if the bridge allows you to access a really nice cross section If the stream is maybe too deep to wade, you can put these instruments in a location that receives light recreational use The instrument can be hung vertically Well then, maybe you should think about a suspension style approach The way that this works is that we have equipment clamped to the bridge And an important part of that is that this entire deployment can be possible without installing any hardware or drilling any holes in the bridge, and that can be a really important consideration when talking to various departments of transportation who don’t want you doing anything to their bridges Well, we can clamp instruments in and keep them really safe and not disturb any of the infrastructure When we hang our instrumentation from a bridge, we hang it with a steel cable, so that the steel cable is bearing the weight of the instrument, so that there’s no strain on the instrument cable that’s also hung with this deployment The nice part about this deployment is that the instruments can be raised or lowered in

the stream, so you can keep this instrument away from the stream bed and you can change the height depending on the site characteristics In this example, we have our instrument deployed inside of a PVC case It can add some protection, so that your instrument isn’t damaged by debris Oftentime, with these PVC cases, we’ll have a number of holes drilled in the bottom, which allows sediment to fall out of the instrument and also allows flowing water to continue moving past the probes And what’s really nice about working from the bridge is that if you can access the bridge, then you can retrieve your instruments during a number of different flow events During these elevated flows, you can see, get an idea from this picture, that as flow comes up, the instruments yearly just get pulled downstream and ride on top of the stream column So debris safely passes underneath It’s been very effective in a number of different environments for us to use this type of suspension style approach But there are drawbacks and important things to consider about it The first being that you don’t wanna work from a bridge that has a really small shoulder or a really busy interstate So you wanna make sure that your personnel can safely get out to those locations Having a suspension style approach may be a bit dangerous if you have a location that receives a lot of recreation They are not very visible in the water and we don’t want anyone getting hung up on the cabling You can, during really large events, if you really get some nasty debris moving down the channel, instruments can become snagged, the cabling is typically susceptible to what gets snagged and can dislodge the instruments And one creative adaptation I’ve seen our colleagues make is to install this deployment with a PVC tube running down the length of the cabling so that debris bounces off the deployment and doesn’t have as much of a likelihood to grab a cable and pull it along with it Another deployment choice that’s worked really well in a number of our environments is placing instrumentation on a rail type system So if you don’t have a bridge to work from or if the stream is too deep to wade, maybe a rail would work really well for you This is best when your stream banks are not very erosive and conditions at the edge of the channel are representative of the water chemistry Some of our instruments can’t be hung vertically, so I’ll show you how that works really well on these rails At the bottom of these rails, there’s a cart and you can add some custom brackets and custom mounts to this cart, so that the instruments are installed on it, and then can be moved on the rail using a handle So you see in this little animation here, someone can stand at the top of a hill, and pull the cart up, and access the instrumentation These rails are all built to a custom length using different sections, different pieces, and we can therefore get the instruments far out into the water to make sure it’s in nice flowing water away from erosive stream banks We can access our instrumentation during all flow conditions, depending on how tall we build these rails, so a flow comes up a little bit, we can still get to the instruments, no problem And like I mentioned earlier, this works really well for those instruments that need to be deployed in a fixed location or position You can build these brackets, so you can add multiple instruments to a single deployment You can kinda get creative with these deployments And something I wanna mention, a few of the pictures that I’m showing, our instruments are not deployed in any PVC cases, and I’ve shown these to a number of folks, and the first thing they say is, “Well, is the instrument safe? I don’t wanna damage these instruments, they’re very expensive.” And that’s a good concern, but we’ve found that some of these instruments are just very sensitive to sediment fouling or some of our rivers just move a whole lot of sediment, so any protective case we apply is really just causing the instrument to foul, so we lose data, we have to have people come out and clean the instruments off So what’s worked really well for us is that we select a deployment location that is not right in the middle of the channel, so maybe there’s some natural protection, of some sort of boulder upstream or whatnot But these rails also place the instruments down closer to the bottom of the stream, so big debris floats over top of it, and we really had good success getting our instruments out of those cases to preserve a better quality record, while also keeping the instrumentation very safe Some drawbacks of this approach are that the rail deployments can catch debris in the stream You could see some leaves and sticks hung up on the top of these deployments These rails can be more expensive than some other deployment options to purchase and can

be difficult to install Typically, we anchor these rails into the stream bank using sign posting, so sometimes you just don’t have a lot of opportunity to drive a sign post in deep enough to secure the deployment The rails are inherently putting instrumentation near the stream edge, so you see in this location, there’s a side channel here and the main channel’s over here, so we really wouldn’t wanna get a rail deployed on the edge here because the chemistry is different than the majority of the channel Another deployment technique is putting the instrumentation in a pipe So here are photos of that type of deployment And you can see instruments are lowered into a pipe that runs the full length into the stream column So this can either be done from a bridge or from locations that are not in a bridge This can be kind of a substitute for that rail style deployment, alright? If those rails are difficult to install or expensive to purchase, using a pipe can overcome some of those obstacles, and you see how the deployment submerged the instrument in the stream and protect them in a pipe Again, I described earlier, how you can use some holes in these deployments, so that sediment falls out and flowing water moves past the instrumentation And again, we really wanna get these deployments away from the stream bank, so getting a pipe to run long enough into flowing water is really going to improve your record In Metro Atlanta, we use a lot of PVC because we have a lot of debris flow, large trees, and things that are usually floating down, but we do, just like Virginia and other water site centers, deploy sondes and monitors outside of the pipe as well One of the things when we Just kinda circling back to that creativity, you see in this picture, the holes in the PVC pipe One of the things that you’ll have to watch in your data is how many holes, the size of the holes, what type of openings really work for your location So some locations work better with holes We’ve adjusted, sometimes we put slices in the PVC, but the idea is that we wanted to still allow for some protection of the water quality monitor, while also allowing as much flow to come through to keep the sonde clean, as well as be representative of the entire cross section But then at the same time, when there is a sediment build-up, you wanna allow for that sediment and other aquatic life, leaves, sticks, and things that are gonna come by You wanna allow that trapped debris to be able to flow out as well So again, just kinda circling back to that creativity, just because we’re showing a location with holes in it, your spot may need slices, or slits, or small holes, or just remember that there’s no one correct way to install it One thing you will see as well, these PVC pipes usually have a pin or some sort of stop at the bottom of them But we’ve used a float at the top of it as well sometimes to keep the sonde so that it At a stable depth within the water column Again, it allows you to be creative, it allows you to meet your data objectives, but it also allows you to adapt to the stream conditions Yeah, those are important considerations We’ve mentioned sediment and debris accumulation in PVC cases a number of times We’re not trying to put passive sediment samplers out in the stream And sometimes, if these pipes are not deployed correctly, they can become that way So you see, this is a photo of a deployment where a pipe was laid down near a stream bank, that was actually eroding a lot of sediment during elevated flow conditions and the sonde just kinda sitting down on the bottom here, it wasn’t great flow moving past the instrument You can almost see the deployment getting buried here on the bottom So there are concerns with sediment and debris accumulation inside these pipes They can travel up into the pipe and if, as John described, if the holes or the cuts in the pipe are not made correctly, sediment won’t be able to get back out So you can consider, even if a pipe was the first attempt made at that site and the data quality aren’t meeting your objectives, what are the other options? Is there another choice to get the instruments out of that enclosed environment and allow the data to be better preserved? So this is just another example of a location where some of our colleagues had a pipe deployment, and because that deployment fouled very frequently, they’re moving to a rail deployment, which will keep the instrument a little more exposed So it’s just constant refinement, constant evaluation, and considering other improvements that we can make Another deployment technique is using the bottom of the stream bed to secure

the instrumentation So, not all sites are gonna have a bridge to work from Some of our sites are very shallow, so we can actually just wade out and retrieve the instrumentation Again, this works for those instruments that can’t be hung vertically And the nice part about these deployments is that they’re rather visually unobtrusive, so they blend in with the environment a little bit more You can see the instruments on the bottom of the stream bed here And they’re secured to signposting, which is driven into the stream bottom and bracketed to the signposts We try to keep these instruments deployed at the top or bottom of a ripple to keep them in flowing water Deploying in a pool with stagnant water wouldn’t really be your best choice We keep these instruments in line with flow so they don’t grab debris and a little more A little more streamlined so debris just flows by the instrumentation And then we usually secure the cabling in with some eye bolts or zip ties or some way, just so they’re more secure to the stream bed and they’re not being grabbed by sticks and debris that move on top of the stream Again, optionally, these deployments could be modified to use a PVC case as needed Some drawbacks of this style approach is Well, if there’s high flows, you probably shouldn’t ask anybody to go retrieve the instrumentation The signposting can trap debris, so you need to consider if there’s ways to improve that And then these stream channels can shift John talked earlier about dynamic stream environments and changing conditions Well, some of these really sandy bottoms can change dramatically after big flow events, and this whole deployment can become buried In this final section, I just want to put forward some considerations for ways to modify these deployment techniques I’ve shown you so far: The use of bilge pumps and some additional technology that can help improve the sites One challenge that could be common in working in different environments is that you’re asked to monitor in very shallow water And this could be hard Because if these probes aren’t submerged, then obviously, you’re not going in to measure the data that you’re expecting So, the first choice in this project, we are asked to monitor in storm water pipes with very little flow during base flow conditions, is that a thin instrument was deployed So think back to that early slide of all those instrument choices Well, for this project, we chose a very slim instrument to keep it in water But even that instrument choice alone didn’t result in the quality of record that we were expecting So we see, a number of times, that the instrument was popping out of water We had to delete data from the database and keep thinking about how the deployment could be improved And what we’ve done here was to use a baffle upstream of the instrument to increase depth and submerge the sonde in flowing water more consistently And that resulted in a record that we were much more happy with, required much less oversight of someone having to come out and clean off the sonde, and whatnot So this was a great adaptation for this site For those sites or environments with really frequent fouling, we talked about it a number of times, fouling occurring from sediment or algal floc or other debris, the use of bilge pumps can be really critical to maintain a quality record And so you see one of these a little bit more in action, here, of bilge pumps that are being used to de-foul the instruments John described earlier about pumping ambient water across the probes at some sort of set interval So you can see the pump in the back here, just some tubing used to direct the flow across the face of the probes, and you can see the water movement that is generated when the pump runs These have been really critical at some of our locations Some sites that have been really tricky to monitor without the use of these bilge pumps required manual oversight much more frequently than we were comfortable with The way that you can trigger these bilge pumps varies based on what you have deployed at your site You can use an existing programmable data logger or just a simple 12 volt timer We use these little simple timers at most of our sites So, we just set two or four times a day, the pump will be provided power, the pump will run for a set interval, just a minute or two And that’s often enough just to reduce the amount of fouling that’s occurring on the instrument, clean off that sediment And we’ve adapted these deployments for a number of our different deployment choices that I showed you earlier, the The on-the-bottom or on the rail Getting these pumps to blow water near the probes has been a really important part of maintaining a quality record

Another modification to some of our sites that’s been very helpful is to utilize two-way remote communication And so, two-way communication can be used with data logger programs to better inform or prevent site visits And I’ll show you a couple of these different applications, and those applications can really be modified to fit your site or project specific needs But some examples are that we can trigger the bilge pump at Whenever we want from the office So if the data appear fouled, we can dial in, access the program at a remote location, trigger the bilge pump, and then see if the resulting data record looks a lot cleaner We can turn off instruments during certain conditions We had an instrument deployed that during storm events it would foul up It would actually clog the instrument, so we would use the two-way communication to stop measuring samples during certain conditions A really nice part about these is that you can use some diagnostic information about your site So if there’s a location that isn’t measuring the way you think it should, if data aren’t showing up on the web before someone goes out to repair that site, you can log in to the location, view the error log, see if you can better diagnose the issue So if someone knows what they’re What to expect when they get to the site Another thing that’s been really, really neat, that John uses a lot, is to use site cameras to visually inspect stream conditions So I think this is great If you’re collecting data during a storm event, you can visually see the conditions at that site and see if that aligned with the data that are being collected in the field And this has been really powerful for cooperators who are interested to see what’s going on at the site, what does their investment look like So the final couple of slides I wanted to share with you is just some examples of complete deployments, just a couple slides of putting all these different pieces together in some of our locations This is a monitoring location we have on the James River, which is a very large cross section where we monitor, 800 feet wide There’s really large trees that can move down the channel So the first choice we made at this site was to deploy our instrumentation from the edge of the stream in a more protected location where the water chemistry is very representative of the cross section We used a suspension style approach here, where instruments are deployed off of an older bridge And what One different piece of this deployment was that the deployment location is different from where the existing stream gauge is currently set up So as I mentioned earlier, sometimes you need a separate power source, a separate data logger Well, that’s what we put in at this location We put in two different solar panels to provide enough power to operate all of the instruments Again, we want to make sure there’s always power, so we don’t have data loss occurring We use a data logger with cellular communication in it so that we can have remote access to this site And this is a site where we have two different deployment choices because some instruments are better suspended from the bridge, and some are Need to be fixed to a rail So we have both the suspension style and rail deployment at this location to meet the different needs of our instrumentation Some of our smaller sites, what these look like These are locations where we utilize the on-the-bottom type deployments, where we have our sonde secured to the bottom using signposting You can see a bilge pump installed on this monitor, and the monitor’s inside of a PVC pipe Again, we utilized some of that two-way communication I discussed earlier And this is a location with a really dynamic stream bottom, so it’s really sandy during certain large storm events That bottom can change and move a lot So we actually keep a couple different deployment locations installed at this site so we can quickly go out, unbury the sonde, and move it to a different deployment location without too much extra work Another one of these smaller streams Again, we’re using this on-the-bottom approach, but in this example, we’re not using the PVC tube because this instrument’s more prone to sediment fouling So this helps us preserve a higher quality record And just a final slide here is serving the example that these are just some basic Or these are just some examples of the way instrumentation can be deployed And what works best requires creativity, requires you to modify these approaches And a project that we recently started requires monitoring on headwater, very small, shallow streams So, the folks that put these instruments out took our existing deployment options of working

from the bridge and adapted a suspension style approach where there is no bridge So you see an instrument hung between cable, suspended from some large trees or off of an overhanging large tree limb So this is all part of this being creative and adapting to your specific site needs Working on the bottom of the stream bed, well, sometimes [chuckle] there’s a lot of bedrock on the bottom, and you can’t drive signposting in But you can secure the instrumentation to boulders or rock outcroppings, and this can be really useful to keep the instruments submerged in a really low profile type of deployment And again, since this is a new project, this is going to be These are gonna be deployments that we modify and see, “Was this the right choice? Do we need to modify anything? Can it be improved in any way?” And just a final add to this, one of the techniques that I’ve seen tried is socks or netting Sometimes, we’ve added a netting around your in situ water quality monitor to help keep out some aquatic life and critters and other things that like to live in and around our water quality monitors If you have great flow, these socks sometimes work very well But as we’ve shown with the PVC pipes, these socks or nettings can sometimes become a sediment trap Again, some of the tools that you utilize have to match your conditions In the coastal areas around Savannah, some of the guys Savannah, Georgia, some of the guys there have found that, in coastal areas, wrapping their probes or instruments in copper wrapping or even using some anti-fouling paints have been successful But again, we always caution that anytime you’re utilizing some other product on your water quality monitor, you want to ensure that whatever that product is, is not gonna affect the water chemistry that you’re actually measuring But again, in those coastal areas, those guys have found that the sock method or those nettings and the copper material all together have been successful In those areas, they were measuring conductors and temperature, but if you were measuring some of the other perimeters, such as turbidity, you might want to consider using something other than the socks But again, those tools all helped us in our goal of retaining as much data as possible, prolonging our maintenance schedule so that the instruments are there longer, and it reduces the amount of field visits that these instruments require The USGS has very defined techniques and methods for maintaining, operating, and calibrating water quality monitors So we wanna make sure that we’re utilizing all the tools that we have to meet those methods but also reducing the amount of energy and efforts that it’s gonna take to maintain these sensors in a very dynamic stream Yeah So I hope these examples just provided some information about the objectives of our deployments and really just reinforcing this idea that the way you maintain your deployment, the way you refine your deployment really has a large impact on the data quality that can really result in really valuable scientific information So, thank you for your attention Thank you