 Okay. This is Alan Marr. Welcome to our webinar this afternoon. Thank you for joining us. We got a very interesting topic this afternoon on corrosion of buried steel. This is a new report out in draft form that we were going to hear about today. Thanks for joining us. I am chair of Coggy and served in that role for a few years now. Coggy is a standing committee of the National Academies of Science, Engineering and Medicine. And it's a part of the Board on Earth Sciences and Resources. Coggy was established to be a focal point within the National Academies for government, industry and academia on technical and public policy issues related to earth processes and materials. Soil and rock mechanics, responsible human development and mitigation of natural and human hazards. As we go through this, if you have questions about Coggy or the, the, the, you can get a hold of Sam Massano. She's on the National Academies staff. You can get her on the website. This webinar is part of a quarterly webinar series that is produced by Coggy through the support of the National Science Foundation, which we appreciate. We are also supported by NIOSH and Federal Highway Administration and some of our undertaking. This webinar will be posted on YouTube shortly after we ended here. Announcement will be sent out to you when it becomes available. So you should watch for that. You can use the chat line to monitor messages that are going on and to get more information about our speakers. And you can also use that to input. There's a Q and A part down at the bottom of the screen. You'll see that you can submit questions as we go along and we'll try to answer as many as we can. Assuming we have time. Before we begin, I'd like to thank Sam Massano and Emily Bermudas for helping put this together. They've done a great job pulling this together and getting it organized. Here you'll see that it will run smoothly due to their efforts. With that, we've got a lot packed into this hour and so I'll get going with it right away by introducing to you Scott Anderson, who will lead off the presentations. Scott actually was the leader and had a very difficult job of making this project a success. I'm sure you'll see that as you hear the presentation. He's been a member of Kagi for some time now and serves as a principal geotechnical engineer at BGC engineering in Golden Colorado. He has over 35 years of experience in geotechnical design and construction involving transportation, water resources, mining and pipeline industries. In addition to working as a consultant, Scott is a former researcher, professor and governmental agency leader with 15 years of leadership positions at the Federal Highway Administration. So with that, Scott teaches some things about buried steel. All right, Alan. Thank you for that introduction and thank you everyone for joining us today. As you can see on the slide, there's a few different things on here. One's the cover of our report, which we're very proud to be releasing. And next to me up on the top on the right there are a couple of co-speakers today, Susan Burns and Gerald Frank are, and I'm very happy to have them as part of this presentation today. They're going to help with some of the content here and also be available for some Q&A at the end. You'll also note that the title of this webinar is different than the title of our report, Improving Engineering to Manage Corrosion of Buried Steel. Corrosion might be something like, you know, gravity or weather. I mean, it happens, right? But there is something we can do and we do see room for improvement in our engineering practice on how we can manage corrosion. And that's what we'll talk about here today. Next slide, please. So I want to introduce just by one word or so, we had a very great committee with very diverse experience and they're listed on the screen and more about them as our background is in our report. And Mercedes brought experience from a practice in the pipeline industry. Susan is soil mechanics and soil behavior kind of experience. So Merrow is works in the laboratory doing corrosion modeling and testing and Ken Fishman is, has a wealth of experience in geosivil industries and looking at corrosion and Gerald frankly who you'll hear from here today again is a material science perspective on corrosion and as you can see she brought excellent perspectives from soil and site characterization electrical resistivity geophysical methods and so on. And Brenda, you know, brought a lot to the table in terms of microbial influence corrosion and Randy Poston brought experience with forensic designs for geosivil infrastructure and and Liz Rutherford she brought in the experience as a pipeline operator so there's a wealth of experience and different kinds of perspectives come to the table through our committee here. There's almost as many nasum staff over there on the right hand side I want to really thank them for all that they did to being kind of a diverse group of cats here we did need some hurting at time to time and they were awesome at doing that. Next slide please. Thank our study sponsors, as you can see on the screen here we had several of them, you know, without without their sponsorship, we couldn't make this advancement that we're sharing here today so thank you. Next slide. Okay, the statement of task that we're operating under and again this is explained in the report more but I'll just summarize it real quickly. The first is kind of a summary of the mechanisms so what what what is defining the problem. And that sounds rather simple, but it was anything but simple and I think the less simple it became the more valuable. It became and it became clear that this was a really important contribution of this work to have such a diverse group of people being able to describe this problem similarly and that's what we did is the first part of the statement of task. The second two bullets or assessment bullets and one is about assessing the practice for characterizing the earth, the ground where is the steel buried. And the second assessment bullet is is about assessing modeling predicting it brings in the element of time into our work and recognizing that we are again back to the title of our webinar we're managing something so it's for a period of time. And we made these kind of assessment actions as part of our task. Next slide please. So, as part of our work we did a bit of information gathering and we did this through a number of different ways different engagements with people and so on but I would say we have we held a two day workshop which was extremely formative for the committee. I can't thank the people who contributed to this workshop enough we had people who prepared material in advance for the committee members to look at some that contributed during the two day workshop it was international we had many time zones engaged. Maybe some of you on the on the call that were participants and asked the right questions and led in different ways there and so that was extremely formative for our work I kind of liken it to a, you know, getting the roller coaster up to the top of the first grade and now we had lots of lots of energy and we went up to the top of the first grade and while you might not see the actual contributions or the sections or the outline of the workshop in our report. I can assure you that the deliberations happened as soon as the, the weapon that webinar was turned off everything here was remote of course this is an entirely within a coven 19 kind of environment and everything was remote, including that workshop. Yeah, so next slide please. And this for the workshop and so on and the participant lists are all in the report by the way and then headaches of the report, and the report is organized as shown on the screen right here pretty logical organization of things. And so, with that kind of general introduction of why we're here and what we're doing. I want to pass the microphone over to Gerald Frankl who will talk about some of our initial recommendations. Yeah, thanks Scott. So, the next slide please we're going to start with some definitions. We're dealing with steel that's buried in soils and different types of steel infrastructure is given the bottom in the asterisk but soils needs to be defined also for those of you who aren't soil people soils in organic mineral grains with void spaces occupied by gases and liquids those liquids in particular water, then with soluble ionic species becomes an electrolyte and together with organisms that might be present in the in the soil, create an aggressive electrolyte that that can cause corrosion. And the important thing is that there can be spatial variations in the aggressiveness of that environment from centimeters to kilometers that become important and time variations that might be seasonal or much longer associate for instance with with global warming. And all of these factors are important in assessing corrosion. So the next slide please we. In the report we have a short tutorial about the fundamentals of corrosion and the forms of corrosion, but suffice it to say for today's purposes that corrosion involves oxidation which is an anodic process and reduction which is a cathodic process those are half reactions that together are the phenomenon and it's the anodic process that creates the loss of metal, as shown in the schematic above and, in particular, this pipe that shown on the left corroded to form pits that you can see there that's one form of corrosion. In the next slide. We know that we know about steel corrosion and in fact, the different grades of steel don't have vastly different rates of corrosion in soil in the same soil environment. The complexities of that environment, the variations the details of the environment and the variations in space and time, make the prediction of the corrosion rate, in particular over long times, very difficult so we don't know where in a large structure corrosion will be worse necessarily and when failure might occur. So it becomes really complicated as a result of this environment. So next slide we described two different approaches that are taken in in industry so there are geo civil industries that build structures, and they deal with corrosion with the approach that we have defined as corrosion allowance. So they, they design with extra thickness to allow for the expected loss of material by corrosion. They also have redundancy built into the design, such that a corroded member doesn't necessarily cause failure of the whole structure. On the other hand, there are other industries such as pipeline industries and storage tanks that take what we called a corrosion avoidance approach. So in this case there's little or no redundancy a failure such as a whole could result in in a catastrophic environmental disaster, actually. And so they in that in these industries they prevent or minimize corrosion using coatings and cathodic protection, and they're often required by regulations or even law to to monitor the, the effectiveness of that protection. The dichotomy and approach is is quite important. The next slide we have some overarching conclusions. And I think we found that practices are not always informed by the best knowledge to deal with corrosion, and they don't account for all of the factors that can impact corrosion. Data that are used to generate the models that that people use for for prediction of the amount for instance the amount of material that would be quite required to last the expected lifetime of a structure. Well those data were collected in the mid 20th century. And there are some issues with the with the, that data set will say, I'll talk more about vocabulary shortly. So we found, as is not a typical some silos that prevent people from adopting new ideas or even learning about them. All right, so our first recommendation is on the next slide it's actually could be recommendations zero, I would say, because it's really about communication. So, we found that there are different vocabularies in these different fields that are listed in that sub bullet on the right. And we encountered this early and often in our discussions in the on the committee, because indeed we came from, you know, a number of And this impacted our communication and we believe will impact the any interaction moving forward so our recommendation is that perhaps standards making bodies could lead an effort in collaboration with other interested parties to develop a common lexicon to deal with steel in the corrosion of steel in soil. And in fact in the next slide, we have a statement, a common lexicon with more technically precise terminology will increase effectiveness of communication and collaboration between and within disciplines and industries. So this is a sort of obvious statement or a tautology. In fact, and you might think that this is an issue that could sort of simple, but actually addressing it is hard and complex so each community has its own vocabulary and is happy with it. So, you know, I can tell you that in my community, which is corrosion, the term corrosion potential. Scott describes a very specific scientific term. And, you know, when Scott in would talk about corrosion potential of a certain situation being high. You know my interpretation would be much different than his, and it took a long time to train Scott to use different words, such as the potential as sorry the possibility for corrosion, or the corrosivity of the of the condition. And so this is not a simple straightforward matter, but we believe is actually critical to accomplish, for instance, the next recommendation, which in the next slide please is that yeah we need, we need more data and and better models. Okay, so we are recommending multi disciplinary longitudinal research and experimentation to identify, you know, the key aspects of the environment that I discussed, you know the synergies between them, and to be able to come up with better predictive models. And so this would involve multi disciplinary researchers, researchers funded from various means, and to address the multiple variables in steel corrosion in in control and characterize conditions. And had come up with we have some thoughts about how this is done, and we have detailed those in the report. And in fact, in, I'll describe a few of them such as in the next slide. We suggest that there would be long term multi varied experiments that would deal with all the different important properties that would use monitoring electrochemical testing and removal of coupons for destructive testing. And to deal with issues that could be impacted by climate change. And in particular, this aspect of microbial influence corrosion, which can be, which can have a huge effect in the corrosion of buried steel. We believe that this approach will allow the development of models that will be very useful for designers operators owners of structures, etc. More detail in the next slide is that, you know, time wise, we have to make observations more rapidly at the beginning but then have long term experiments. We, we recommend lab based experiments with very controlled conditions, as well as field based experiments that that use extensive characterization and monitoring. Okay, so this is difficult and long term work. In the next slide we talk about how this would be done. So, multiple, you know, I, in my own personal experience, I had the, if you could go back to the previous slide please. Sorry, that one right. I've had the pleasure to be involved in multiple, multi disciplinary projects, and they're hard, but when things come together, they can be very productive and, and, you know, with great results and impact. And it's really required here because of the range of expertise that's needed. And we believe that partnerships between various entities will be required to set this up and get it in motion. So, you know, it seems to be actually an ideal topic for this type of approach, multi disciplinary collaborative research and development. So with that, what I would like to do is hand this over to another committee member, Susan Burns, her bio has just appeared in the chat, and she is a professor of civil engineering at Georgia Tech. Excellent. Thank you, Jerry. So Emily, if you could go to the next slide will move to recommendation three. And from the committee we spent a lot of time looking at the data that was available. The range of data as Jerry and Scott have alluded to some of which was gathered in in the early 1900s, even so some of the most reliable data sets we have can be quite old. We also have new data which is being collected to the to be used today, and we've got some good studies which have looked at statistical analysis of these data. However, we need to continue to expand this work, and the committee recommends that we apply techniques such as cluster analysis or Bayesian theory to both the existing data that that has been statistically analyzed in some studies say by NIST but we have all the new data that are being gathered, looking at the influence of different types of soil properties on corrosion, especially so that we can better understand the relationships between specific individual individual components as well as their their synergies that come together. We believe that this will allow enhanced site characterization and better monitoring because we can better target problem sites. We also bring insights into relationships between physical properties and those that are causing corrosion in specific locations. We also believe that this is an area which is rich for understanding microbially influenced corrosion as well. So next slide please. So essentially what we know is that we need improved estimates of corrosion rates that's clear and clear from the work, the review work that we perform. These improved estimates are going to come about if we are able to consider all relevant subsurface properties. This also will give us the chance to use data driven approaches to apply weighting factors as opposed to intuition. And that is what will help us come in also and identify synergies between these relevant and related subsurface properties. Next slide. In addition, as we worked through the year or two that we spent on this looking in depth at the different ways that we make decisions, both in terms of site characterization and design as well. We came to think broadly about decision support systems. So we know here that when we look at parties who are responsible for managing buried steel. There is a lack of a framework that will tie together these multivariate properties and also allow us to guide prioritizations of repair work right and we especially want to do this using risk informed approaches. There are protocols that are specific to some and some specific industries but those are relatively few also in order to guide our site characterization data collection oftentimes site characterization is driven by say mechanical properties of strength or deformability, and not necessarily driven by gathering information about corrosivity. Now we're going to see here that we use simplified and empirical methods for modeling corrosion, corrosion rates as well. We know that these are limited and effective only in some specific conditions. So what we want to do is develop a larger framework in order to do some of these assessments. Next slide please. The support systems are nice for this these are tools that will guide us through a variety of different alternatives and let us make decisions based on fairly complex input parameters. The chart that you see on the right hand side is an example of a relatively simplified 2d flow chart so this was developed by Fishman at all in 2021. And you can see here that the decisions are being made in terms of corrosivity based on soil properties such as whether or not it's a well graded or upper open graded soil, looking at things like pH resistivity common parameters that we think about when we're looking at the corrosivity of soils. We realize that we need more complex systems these more complex systems or databases that can handle multiple inputs and we believe that this can help mitigate some of these uncertainties that occur when we're making decisions about implementation for buried steel design and monitoring. We know these are particularly helpful when we've got large volumes of data and we can combine them with predictive models, which is the ultimate goal we're looking at here. The example of an active decision support system available today is geotech tools which helps you make decisions on ground modification. Next slide please. So as we moved into here what we would suggest for decision support systems is that these be developed based on standards for multiple industries. So here that there are multiple points along the line of design for any kind of buried steel. First of all, a decision support system that would that would help sort out site characterization questions we feel is very important. So let's just look at a look through a comprehensive set of characterization tests, and then look at the ones that are temporarily and spatially variable in order to do the characterization that's needed over time and also over depth. The decision support system which is focused on risk informed decision making to help us guide the corrosion management processes. In all of these cases will have multiple inputs, those can be selected and uniquely specified for the sites which which are being designed or being monitored, and then looking at a range of comprehensive monitoring tech techniques as well. So the circling back to Jerry's point about the lexicon. Again, this is another location where the need for a common lexicon across industries is very great. Next slide please. So the site characterization here we want to highlight the things that are important we need to capture lateral and vertical temporal spatial variability here and let us do combined site specific subsurface characterization. And then we need to produce a minimum field and laboratory characterization program right and this is this can be tailored to changes in time take changes in land use, any kind of climate changes, those types of issues. And then on the corrosion management side here we're interested in what actions can managers take one investments, how do we do this in a risk informed model. Next slide please. So we understand that these decision support systems can start small and to date we've gathered a great deal of information, although not consistently but we do have studies that have measured moisture content saturation resistivity pH, and a variety of practical parameters in addition. So we can start with an initial focus on properties that are currently commonly used, and then in the future expand these to more promising, or hopefully more promising technologies as well so things that, for example could be helpful to us when we're looking at microbial the influence corrosion as well. So we want to include this both in terms of the lab work and the field based methods as well we understand that some lab methods are designed to replicate field conditions, while other lab based methods are designed to be more of an index test. So we want to capture all of this and the related usefulness in the decision support system. Next slide please. So with that we feel that there are lots and lots of questions with monitoring and how to best use the resources we have in order to gain sufficient monitoring data. We understand that it's infeasible to directly monitor all buried steel right it's just not possible economically or logistically. And so we recommend looking at changes in land surface so for example, have there been changes that might affect surface water groundwater flow. So what kind of chemistry do we have some influences there. And we feel this could be a cost effective early indicator of detrimental changes within the subsurface. Next slide please. Surface monitoring, for example, can be done from from relatively simple desktop considerations so changes in land use for example, we can look at changes in land cover as well are we changing from say forested regions into paved regions. It will affect surface water run off it will take affect moisture content of the soil, all of which in turn then will affect the crossivity of the soil. So it's an indirect indicator. Anything that alters surface groundwater or geochemistry. We oftentimes can monitor through a change in land use. And that concerns about any type of installation that could produce stray currents are an issue as well, or any type of issue that might feed the microbial potential for corrosion as well to so changes in fertilizing practices things like that. And as always, we know that climate change here will come in and produce changes in temperature and precipitation which will also influencing to have influence in these monitoring decisions. Next slide please. We encourage again the use of publicly available resources, we can see that there will be other opportunities as well so anytime when we install a retrofit infrastructure. We have sensors that could be installed in order to measure things relatively inexpensive sensors exists these days for measuring moisture resist resistivity pH for example. We also understand that a systems management approach is needed to be able to track these relevant practices across infrastructures buried steel infrastructure is often separately managed by different agencies. Next slide please. So then we, following on from this we also feel like there is an opportunity for what we'll call opportunistic data collection. And that is that when we are at a point of partially or completely decommissioning buried steel infrastructure say for example during maintenance, or replacement or say in the case of a ruptured water pipe if we're doing a repair. So here that we've got the asset uncovered, we can apply standardized protocols to collect this opportunistic subsurface property and infrastructure data. This, this type of monitoring we think could be a source of information but we acknowledge that it needs to be systematically saved to in order to properly inform longitudinal research so believe also here that our decision support systems can can take some of this as well and help influence future decisions. And with that, I will turn it back over to Scott. Thank you Susan. Next slide, please. The observation that led that led to our, our last specific recommendation was that as much as we're recommending new data be collected and collected in a systematic way with common terminologies and so on and so forth. There are a lot of data out there, and the data aren't reaching their value potential because they're not reachable. And so we came to this idea that some sort of a data clearinghouse would be a valuable recommendation. Next slide please. The clearinghouse terminology may not be all that common or familiar but this basically a way of placing data publishing data it's publishing data is becoming a more common practice and in other parts of the country that uses engineering and sciences and medicine and other other disciplines as well. You know, in fact there's this idea that's expressed in the first bullet there of findable accessible interoperable and reusable data. So these are data that they can be reached by a number of people as needed. We found that even among the committee that there were some folks who had access to a lot of data that could be used in various ways of be valuable to others, but it really wasn't shareable. And we recognize that in a lot of the industries that we work in that that there is some privacy concern with with some of the data and these are regulated industries these are industries that are in the business of business. And they're also government entities as well. And not all data can be shared but there are ways of keeping private private and proprietary data can be we believe could be shared in a private kind of way. As a case in point I'll remind folks that yesterday was Cyber Monday it turns out and there were 11.3 billion dollars of transactions made yesterday all in a private way so we're kind of thinking that there are ways and we don't need to just sort of dismiss this one so easily. Next slide please. So the report gives some ideas on on how we might implement this data clearinghouse. And that's what's kind of listed on these these bullets here. And that's probably sufficient for here the next slide please. And the value. You know this is a value slide so you know data aren't information data need to be transformed into information so so a lot of what the value of having a data clearinghouse will be is for research, but not solely research and you know but but the data would be used in a lot of different ways to produce information that's useful for specific projects and that's where we see the value of the clearinghouse next slide please. So that's our seventh recommendation now just some some some concluding thoughts here and we'll go to a question and answer period, but but you know really. I think it's the committee's opinion that we need to be cautious about complacency, the corrosion does happen but there are things that we can do that would greatly improve our practice and make us better engineers and and if we take advantage of them we know there's no doubt that we have some really, you know, fairly rare but very dramatic and very costly and impactful failures that result from corrosion buried steel, whether it's a pipe or a bridge foundation or so on. And there's also, you know, things that are less visible to us but there's a lot of buried resource in the ground where perhaps we're being over conservative with our designs and we're not taking full advantage of the kinds of things that we've talked about here today to to provide a more accurate design. And so with those concluding thoughts we'll move on to the last slide which is kind of a visionary way forward and I think that our committee found that the collaboration was the key really. And it's always the case of course but this is a corrosion study has always been a very interdisciplinary kind of area. It's an area that maybe gets left behind a little bit because of that fact but we found it in our committee the importance of collaboration and I think to a broader extent in the industries we all work within that if we can find ways to collaborate and these input these recommendations that we've provided here today, almost all require that kind of collaboration. We can we can make some great advances and and and save lots of money building safe infrastructure for for a long period of time. That lasts a long period of time. So with that I think that's the last slide that we have here and Alan I think I'll turn it back over to you for some moderation of the questions that we have. Thanks very much. Scott, Jerry and Susan, and thank you for being concise here. It's a big, big topic. Well done to stay within our time limits. I'm going to start kind of with a question, maybe. You know, how might I mean what your recommendations are really pointed at more work is needed. How might that work be organized and how might it be funded. How might one get this kicked off. I think that that are if I'll start on that and I welcome any contributions from others there too but I think that one of our recommendations and Jerry maybe called it recommendation zero is a place that does need to happen early in the process right so this could mean that we need to communicate clearly with one another we need to have a common lexicon we need to be talking the same language and and and that is not an easy task but it's not an expensive task either I suppose so it opens it up to lots of forums where these kinds of discussions could be could be had initially and get that that kind of thing rolling we all in our respective more or less to be doing that kind of thing and and and and bringing and recognizing through through this work recognizing the importance of doing it and then bringing it together later on. Many of the other recommendations can happen in parallel they're not they're not sequential type things and any bodies that you know found a particular interest in those could pick them up and start there. Any comments from others where we might might find sympathetic partner and funding this kind of long term work. Yeah so we were trained to not specify where funding would come from exactly, but it does seem that that you the government industry partnerships would be useful for this type of study. You know, there are I think there are various mechanisms that exist already funding mechanisms that could be applied to this topic. I'll second that I think that the the cost to the US GDP is somewhere on the order of about three to 5% every year for metal corrosion right so that's about a two and a half trillion dollar global GDP so this is not. It's not a by not funding it, we are already spending the money as well too. It gets down to whose pocket is that coming out of it's hard to sometimes allocate those out and get somebody to fund it. But I think your work will certainly be helpful to us as we talked to various folks about our agencies or groups about where to go with this. Kind of related to that might be. What would you all recommend as the next big step here. What are your recommendations but what would be the next big step. Well, I think you know what we what we plan to do with this work and maybe this could get the the next big step started. Alan is that we will have a workshop in the coming year to describe this work with the sponsors of the work who have already sort of expressed an interest in it by way of their sponsorship and we'll have. Sponsors and other participants in these workshops details of which will be released, you know, shortly but we do plan that in 2023. And these may be ways to float, you know, sort of pilot ideas for how to get started on some of these recommendations. That's good, because that might flush out more clearly where support for this might might go. I'm curious. In your work. Have you been able to identify immediate steps that you would take into your own engineering practice. Different than what you might have done before you started this study. I have but I'll let someone else go first if they want to. Yeah, I'll jump in I think to me I think I before we started this study I appreciated the complexity of the process. But I did not understand the depth of the complexity here right so looking at the range of parameters that were currently measuring right which may or may not give us good results. I guess for me the next focus is more clearly identifying the role of the soil parameters and how they interact in order to cause corrosion so and by soil parameters I'm including the saturation the fluids the geochemistry the microbiology. I think there's lots of places where we can see conditions where we would expect corrosion to occur but it does not. And so understanding the synergies in those relationships I guess for me as a researcher would be would be the next step that that I find very interesting. Yeah, that's that's a great lead in Jerry I don't know if you have something you want to say but I'm going to go off of Susan's lead in here because it goes to one of the things that I've been thinking about it and that is as much as I like to think about time and life cycle and so on in the work that we do. I've always thought of corrosion is fairly static, you know, there is a certain. responsibility of the environment Jerry I got it right there you know that I can characterize but but I don't typically think about how that might change your time whereas you know like, and other things I'm doing right now maybe working on a dam that's not intended to hold water except for an extreme events but I know it has the potential to hold water so I designed it as though it's full water and and I don't I don't think that way with respect to I didn't any have before working on this for the last couple years. I don't think that way with respect to corrosion but I do think that's really an important thing is that the environments change and that if it has the potential to hold water so to speak it might hold water and that's true with the buried steel environment as well as you know a reservoir. Well, I think you know Susan and I are professors and we do research that gets funded, and I can guarantee everyone that if funding is made available that there will be great interest in the academic research committees to join in and participate. And, you know, the communities is sort of lemmings you know once a group goes off in the in the direction, often other people follow behind, you know, and find ways to get involved. So, I think we need a, you know, a spearhead of activity to initiate interest. The point you made was the need for central data repositories. I think you use a different word but you know I worry a lot about these aggregations of data, having all types of data quality from high to very low. Have you thought, or do you provide any guidance as to how, how that gets managed and dealt with, or is that something that data science people have already sorted out for us. I just think for example that a lot of the testing that gets done today might be itself suspect. Any response from anyone. No curating data is is complicated. But you know what we talk in in the, in the report about, for instance, machine learning approaches to addressing this issue, and you need large data, you know so if we have large data sets, which we think could be created in this area. Large data sets can have varying quality and, you know, there are ways to deal with that. You know, if there, if there aren't large data sets then bad data can be impactful. Yeah, it's a good point to write because one of the primary data sets which is used to guide decisions today was based was developed over 100 years ago or on the order of 100 years ago before statistical design of experimentation was really a thing So, so I think we've had the data issue for a very long time we relied on data sets that aren't necessarily ideal for a very long time as well. Yeah, well we have mechanisms journey volumes more data, maybe including continuing volumes of it being. Yeah, I think the down to all that we did to address that really goes back to this, you know, common lexicon common terminology I mean a lot of quality, you know, could be attributed to misinterpretation misunderstanding of what's being reported and what the metadata are on that you know so so I think is as we build clarity there we're doing something to improve the quality and consistency of the data we collect, but it's a good point though. Yeah. And I keep wondering are we collecting all the right data for example, fundamental to corrosion is oxygen. That word wasn't mentioned in any of this presentation, nor in any of the measurements suggested so are we missing something during maybe that's a question for you. Yeah right I think well, you know we understand what the critical parameters are that will impact corrosion so oxygen content is critical, the degree of saturation of the soil is critical right so resistivity is a. A property that's easily measurable and commonly used in design of structures but is not really sufficient because resistivity does not equal corrosivity. So, you know, pH is important temperature, all of these things are important in, and of course the ionic content in the electrolyte, as well as the local microbes that are existing. So all those things can be controlling, and you know are important to characterize and to understand. And that's again part of the challenge is that those, those conditions can change over space and time. Yeah, to your point to Alan we do mention redox potential right but that it's a tricky thing to measure and that's part of the right that's part of the, the place where we run up against what can we measure well and what what is still fairly challenging to measure and then in situ condition. Yeah I was going to ask my next question of you just kind of on that point. Given what our practices are today, primarily to measure cross measure conductivity, pH, maybe sulfate content. What are we missing, what would be the next thing to add to that list of common laboratory tests that we should consider. I think it's a hard question right, I think what we're missing is the synergies honestly, I think that we probably have, we have the soup. I'm not sure how, how the taste is yet though right how did those parameters come together and mix. And what is it that that identifies where corrosion will occur versus where corrosion will not occur. That's really well said, we should get that in the report, I don't think that's in the report. I was about to say that's my personal opinion, going off the point. Yeah, that's that's that is a very good point. I think it's a good one worth capturing. Kind of related to that as a question popping up here is there a need for revising current best practices for assessing corrosion that can be implemented by why we wait the results of this recommended longer term research. Well, I'm not sure. I'm not sure that can be done in the interim. I'm not sure that's a shorter step than getting started on some of our recommendations here. I think revising current best practices in my mind is a is a long term goal, and not an interim step, but I think that the kinds of ideas that we put forward in this report do lead to to how we would want to revise those practices and, and I think, you know, just getting getting right at it is I think we're probably measuring the right things more or less but but we're not looking at how they work together or work opposed to one another in a sufficient way like these tests you listed out there Alan. I think we just look at them based against some reference values individually and make in the geo civil space anyhow make some kind of assessments of the aggressiveness of a site or a ground based on them individually. And that versus some kind of combination which I think the practice is moving in that direction, but moving there faster could be, you know, something we should, we could pursue. If you look at all those data that are presently out there in a given plot correlation plot. There's like two orders of magnitude at least on corrosion rate scatter in the relationships. That's, that's, that's a lot. Maybe us being able to close that with more and better data models, or does more data just add more uncertainty. I think we could gather better data for sure. So we can run controlled statistically designed experiments and get better data which will help us generate better models. I think we can close that. If your report does address that importance of better data. If it doesn't you might want to drop a couple of words in on that. It does include that you know it does specifically address that. I think even some of the details in the slides I presented, talk about how to go about collecting good data. Very good I look forward I have to admit I haven't read the draft yet but I'm looking forward to it. So we're about out of time here. We had a very good participation here. 180 people listening in on this that's that's quite significant webinar on such a focused topic. I want to thank each of the three of you and pass our thanks back to your committee members as well for a job well done, and especially Scott for shepherding this group of diverse folks through very difficult times with COVID. Getting us to a result that I think we can be proud of. So again there is a link available to you on a, where you can freely download this report just as a guest, and this webinar will be posted online you should get an email soon to tell you where, where to log on to find it. I want to remind you that we have a very interesting presentation coming from December 15 by Professor Lord Meyer from England. A fascinating man who's had a lot of involvement in the development of the Elizabeth line, which was called Crossrail in its earlier days. He'll describe some of his experiences working in that, but maybe also a little bit of comment on the politics in England as he is currently serving as a member of the House of Lords. So, I really look forward to to that. And with that, I think we're pretty well done I think our speakers again for excellent job presenting this material and staying within our time limits. Any other comments from anyone we do have a couple of minutes. But I think we're in good shape. Any of the panelists want to or speakers want to add anything as a close out. No, just once again, thanks to everyone for their participation, you know, just in the Q&A is here today and also for the helping the committee through deliberations over the last couple of years and. Yeah, it's been a pleasure. Thank you so much. Thank you so much. Thank you. Thank you, Emily has posted a couple of links in the chat that you can go chase down more. And thanks again, everyone. Thanks for your participation and look forward to you in our next weapon. Thanks Alan for moderating.