Infrared measurements
Network Dike Monitoring
Workshop on Infrared measurements for flood defence management
On 6 June, the Workshop on Infrared Measurements for Flood Defence Management took place in Tiel. This workshop covered the technical principles of infrared measurements, how measurement results can be linked to dike assessments in the context of the WBI and presented the results of the measurement campaign during the high water in early 2018. This report looks back on the workshop and reports the results in outline.
Welcome and introduction
Everyone is welcomed by Wouter Zomer. He briefly explains what the Dike Monitoring Network does. The network works as a facilitator and agenda-setting body, among other things by organising meetings like this one. The aim is to hold about eight such meetings a year, each with a different topic. On 27 June, for instance, the next workshop is already scheduled, with the topic of fibre optic monitoring. The afternoon's programme will be presented next:
13:00 Reception at Waterschap Rivierenland (De Blomboogerd 1, Tiel)
13:15 - 13:30 Welcome and presentation programme by Wouter Zomer
13:30 - 14:00 Presentation feasibility study infrared measurements by Onne Rösingh
14:00 - 14:30 hrs Presentation results infrared measurements by Linda Klein & Nelle Jan van Veen
14:30 - 14:45 hrs Introduction case
14:45 - 15:00 Pause
15:00 - 15:45h Elaboration of case
15:45 - 16:30 hrs Discussion
16:30 Closure and drinks
Feasibility study Infrared
The reason to start monitoring has been due to several factors, but one of them has been the dike breach at Wilnis. Also from that came the IJkdijk projects. One of the tests carried out there are the piping tests. In those tests, controlled piping was caused and many different measurement and monitoring techniques were applied by way of testing. Those measurement techniques were also used to determine predictability.
A feasibility study was conducted by Intech, Arcadis, Deltares and BZIM in 2017 on the use of infrared monitoring in dykes.
Infrared measurements are based on temperature differences. Seepage water has a different temperature than the water in the ditch. With IR, incipient seepage can be detected and high-risk areas can be picked out. Here, in summer seepage water is colder than ditch water and in winter seepage water is warmer than ditch water. This is caused by the varying temperature of surface water and the relatively constant temperature of groundwater. This, of course, includes a tipping point where groundwater and surface water have almost the same temperature. At that point, the difference between day and night is the saving grace, causing a few tenths of degrees difference as yet. Measuring this, however, requires a highly accurate infrared sensor.
There are several ways to measure with infrared, both from the air and from the ground, using, for example, a 4x4, plane or drone. In the case of using a drone, however, you are strongly bound by laws and regulations regarding drone flying.
As for sensors, there is a lot of choice. Infrared sensors are available everywhere these days. However, the biggest difference is in accuracy, which is very important for seepage detection. Quell detection requires high accuracy and high resolution. Infrared measurements are sensitive to the outside temperature, especially when measuring absolute temperature. It is therefore important, even during measurements, to record the outside temperature.
The application of infrared is actually in every step of the WBI, Onne shows. First, there is the dike section classification, where the seepage behaviour says something about the correctness of the dike section classification. If excessive seepage occurs at one location and not at all at another, while both locations are in the same dike section, one can question whether the assumptions regarding the dike section classification are correct.
In the simple test, Infrared then has an application for hard-to-reach places. In the detailed test, the seepage behaviour can be monitored and something can be said about the entry and exit points. The entry point cannot be determined exactly, but the degree of temperature exchange does say something about the length of seepage.
In the detailed test, infrared has another application in the case of performing stress tests.
The infrared camera can be used to measure 50 to 60 km per day, depending on various factors. The interpretation of the data is then done with software developed in-house. In the detailed test, there will always be a combination with other measurements, such as measuring the flow in a well, determining the residual resistance of the overburden, etc. Infrared is a start and then additional measurements are made.
What is striking when using infrared is that the passing of a ship immediately gives a result in the seepage ditch. There is first a pressure drop because water is sucked away towards the ship, but then there is no pressure increase. One explanation for this could be that there is a thick layer of silt in the fairway that causes the hydraulic short circuit to be closed.
Currently, the problem is that at high tide, all the water boards suddenly want to measure. This can be done better and more efficiently by making a national approach for this. This can then formulate at which locations measurements must be taken at which water levels. By setting up a 'flying brigade' for this, for instance by means of a framework contract, measurements can then be taken quickly at the moment they are needed.
Results IR measurements high water January 2018
The locations of burrows are often, for various reasons, unknown. For example, the wel may be in a place that is difficult to reach, or the location of the wel may be difficult to see due to, for example, bushes. During the last high water, in January, several water boards used infrared measurements. A drone with an 'ordinary' high-resolution camera and an infrared camera were used for this purpose. The advantage of using a drone is that hard-to-reach places become accessible. For example, it is possible to look further into a wet meadow where it is normally impossible to walk or drive.
At Herxen, for example, a very large difference was noticed between the number of wellen observed with the naked eye and the number of wellen observed with the infrared camera. However, a combination between infrared and ordinary photographs is always needed to check that the seepage is not something else. For example, measurements were taken where a number of seeps were present, but one of these 'seeps' turned out to be a horse. In addition, a well is far from carrying sand. This cannot be seen with the infrared images, so an inspection with the regular photos or an on-site inspection will always remain necessary to determine whether a wel is actually a risk.
Using infrared has both advantages and disadvantages. The advantages are (1) infrared complements visual observations; (2) one can watch the measurements live, even from a distance; (3) the combination of infrared and normal photos makes it possible to validate afterwards; (4) the technique can be deployed quickly, also in case of calamities. The first disadvantage is that measurements can no longer be taken in extreme weather (heavy rainfall, from wind force 7). In particular, rain is a problem for moisture on the camera lens. In addition, laws and regulations can be restrictive. In city and village centres, for example, flying is normally not allowed. However, this was never really a problem during the measurements of the last high water in January, mainly because of the location of most dykes outside urban areas. However, it is expected that an exception will be made in the event of an emergency situation in extreme floods. There will be new European legislation in 2019, which will change some things. Currently, the rules for flying drones commercially are much stricter than the rules for private individuals. That will change with the new rules, which could broaden the deployment.
Besides using infrared, there are a number of other additions that could be applied. The first is the use of satellite imagery, where near infra red can be used to see how the river runs. Near infra red looks at the light spectrum that is near infrared. Thus, during last high water, it was easy to see that for one river the floodplains were flooded and the water was therefore against the barrier, while for another river the water still followed the main course.
In addition, Aveco de Bondt has been developing a risk map for piping. In combination with previously unknown seepage locations, this could be a very valuable application. The infrared images do not show whether a well is carrying sand, but that is possible on the ordinary photograph, by zooming in enough. Subsequently, a field crew can also be on standby and sent to the vulnerable spots. It is further noted that wheels are very difficult to size. They are generally of such a size and depth that seepage water cannot actually be seen.
The audience asked what the measuring range of a drone with an infrared camera was. This was answered that it measures from 40 metres high with a ground resolution of a few centimetres for the infrared camera. The camera is accurate to within a few tenths of degrees.
Toekomstbeeld
Nelle Jan van Veen explains from the POV piping where they ultimately want to go. The goal is to have a scenario/monitoring plan in place by the next high water, allowing rapid measurements to be taken if necessary. The emphasis here is also on learning to deal with high water and calamities. In the case of the observation of a well, there is a second registration after the normal observation to learn about the behaviour of the well. Then there is also a third registration in case after the 2nd registration not everything is clear about the wel. In that third registration, after the flood, drilling and the like can be done to learn more about the wel. The manager uses its own database for this, but it is also important to have a national database. That way, the available information and data can be handled more efficiently and better.
In addition, Deltares does have a registration of sand entrained dams. However, this lacks the properties. It is therefore very important to properly record observations. After a high tide, observations are often left unrecorded because they are not a priority. However, a lot of information can be extracted from them. Simultaneous measurements certainly add value. We also need to consider what needs to be measured ad hoc and what needs to be measured continuously.
Case
The case deals with the Pleijdijk location near Westervoort. Eight maps with information are provided of this location:
1. Infrared drone image taken on 10 January 2018 showing the locations of wellen
2. Orthophoto (corrected aerial photograph) from the drone of 10 January 2018
3. Grain size analysis of two voids and location of sand samples
4. Soil data (drilling and probing) and the location of those measurements
5. Sand track map including depths of the surrounding area
6. Elevation map of the surrounding area, AHN2
7. Monitoring well data from Water Board Rijn and IJssel
8. Infrared image of the inner embankment, dated 30 January 2018 (second flood wave)
Using these data, three questions are posed:
1. What does the data say and what questions does it raise for you?
2. Can you estimate the risk based on this data?
3. What data is needed to better answer the above questions?
It is noted from the floor that historical data would add value here. How did the wheel originate. Probably by a dike breach and how was it filled in at the time. Despite the fact that no measures were taken and it went well, it does not seem wrong to box here given the large amount of seepage.
It is also noted that we are now only looking at a snapshot. Repetitive measurements could give a better insight into seepage behaviour. One possibility for this could be a stress test on the dike.
It is also noted that a visual inspection would not hurt here and that a cross-sectional profile of the section in question would also help in making a decision. It could be the case that the wells close to the dike act as a kind of valve.
Finally, it is also questioned what happens in the wheel. Indeed, the water seems to flow towards the wheel, but the water inside the wheel does not seem to have a different temperature. What is this caused by? Does the water flow back out of the wheel on the other side? Or is the wheel so deep that the temperature doesn't just change? Moreover, the wheel itself does not seem to have any seepage, which could indicate a thick layer of peat at the bottom of the wheel. To find out, an underwater camera would be an interesting addition to the measurement data. With that underwater camera, it is easier to find out what is happening inside the wheel and what is happening to the water (and sand) that seems to be flowing into the wheel.
Ultimately, nothing was done during the flood, because this is how it happens at every flood. So, based on historical data, it was decided to leave it as it is. However, it was also noted from the audience that this could be dangerous. If it has happened a few times before and sand is washed out with it, it could easily become a pipe after a few times. Especially so close to the dike, this is quite a big risk. It should also be noted that a small support berm of a few metres was recently installed. This will most probably have been constructed because of a stability problem. The historical data on which this decision was based are therefore no longer applicable due to the change in the situation.
It is also noted from the audience that the data from the borings show that the overburden is about 1.5 to 2 metres thick. However, these are measurements on the outside of the dike. If we assume that this is the same on the inside and the ground level on the inside of the dike is about 1 to 1.5 metres lower, the overburden there is about half a metre to a maximum of one metre. If a farmer then starts ploughing the field there, there is only a few tens of centimetres of overburden left. As a result, backfilling makes little sense, because the overburden will break through somewhere else and the seepage will come out through there. This means that other solutions must be devised to counter this seepage. It almost seems better to plough over the top layer completely, so it can no longer burst. This will also prevent piping from occurring, as the seepage does not concentrate in one location.
Conclusion
In conclusion, the audience, and especially the water authorities, were asked what action perspective it actually gives water authorities.
The answer was that using proven strength is very difficult and that it rather leads to an unknown sensitivity. It is also responded that a good information plan is crucial. You have too much rather than too little information. Of the data supplied, actually the infrared photo is enough to see that it is not good. The photo of the embankment, for example, is of little or no use in this matter.
In addition, it is pointed out that you receive a lot of information, but it is unknown what you do with it. After all, you can't respond to every single one. This makes capturing and monitoring sensitive areas even more valuable. To keep it manageable, you will have to filter to really sensitive spots. The size of the temperature difference still gives an indication of the size of the wel. This could then also help in filtering the sensitive locations.
After the discussion of the case, this afternoon's speakers will be thanked by means of a Deventer Cake. Afterwards, everyone is invited to discuss things over drinks and snacks.
Description
On Wednesday, 6 June 2018, from 13:00 to 17:00 at Waterschap Rivierenland in Tiel, a workshop on Infrared measurements on dykes for the detection of swells and piping will be organised in cooperation with the Project Overstijgende Verkenning Piping (POV Piping).
This workshop will focus on the results of the infrared measurements carried out during the high water in early 2018. It also discusses the different ways in which infrared measurements can be carried out. Finally, the further use of infrared measurements as a regular application in dyke management and the link with the WBI will be discussed. We have the time, so can go in-depth on the technical content aspects of infrared measurements/monitoring and the relationship with management/assessment/reinforcement.
The presentations will be given by Nelle Jan van Veen (POV Piping), Linda Klein (Aveco de Bondt) and Onne Rösingh (Intech Dike Security/Dijkmonitoring Nederland). Ger Vergeer of Waterschap Rivierenland will be your host.
You can register to participate in this meeting via netwerk@dijkmonitoring.nl. After registering, you will be informed about your participation and kept up to date on current developments about this meeting by e-mail
Location
De Blomboogerd 1, 4003 BX Tiel
Organiser
Land Use and Water Management
POV Piping, Network Dike Monitoring
Name and contact details for information
Wouter Zomer
Apply via
Network Dike Monitoring
netwerk@dijkmonitoring.nl