 Okay, so Anne and I have been asked to give a high-level summary from our first three sessions today. And so we will do our best to give a quick sort of our high-level takeaways. I'm sure they won't match with everyone. But from session 1.1 where we talked to people who work in operations and do risk communication on the day-to-day, one of the big takeaways I saw was a big challenge in this risk communication is the noise. And the noise might not mean what we all think it, or at least what I thought immediately when I heard noise. The noise here was the kind of GWIS statements or these TC vitals which are these messages about the category of a tropical cyclone or the intensity, things that don't directly necessarily relate to local experienced impacts by people. And how do we sort of work towards making sure people are hearing the actual impacts and personalizing this information versus these really cool but maybe irrelevant in the moment historical facts. Like this is the first time this has happened or this is the strongest tropical cyclone since X or Y. Another element was a really important part of this risk communication, especially dealing with less typical types of storms is laying this base foundation for communication among partners and how important that is. I think that's probably something that has been studied pretty well to date but could always use more research. I know a lot of the work and done with mental models risk communication and Julia DeMuth and Rebecca Morse. But understanding the communication flow with partners, if you have a good base communication, you can sort of pivot when you have these sort of unusual events and make those messages a little clearer. And this is sort of my question that came away from it was can impersonal graphics convey personal risk and you can think of this on lots of different ways in terms of scale. Can national scale graphics represent county scale or street scale risk? Or can graphics that like the cone that talk about where the track the center is going to be can that really convey where those impacts are. The second part is really sort of a summary across the first two parts because we discussed some of what happened in the first panel. And also I should say that we should have asked Jen to do this because she had all the notes there on a slide. Yeah, we just need to grab that slide. So there was a lot of emphasis throughout the first two panels on the speed and location of impacts and we heard some of this also in the panel that Marshall just just moderated. There was discussion a lot of discussion of intensification of wind and rates of rainfall in particular as examples of this, but also that there are different hazards in different locations for the same named event, and often very localized was very difficult to track that kind of dynamicism and important to track it because that's what people are experiencing. Among the researchers we heard the need for more research on how information moves across platforms and how to leverage a tools for good in these dynamic and evolving situations. So both the hurricanes and the tropical cyclones and, and all the associated hazards are evolving and dynamic but the communication systems are too. And then there was a discussion, Roxy mentioned particularly a need for research infrastructure and support that can support pre disaster studies and be agile for studying these kinds of rapidly intensifying or sudden real time and dynamic events. And for social observational data, as Julie emphasized we need monitoring and observational data analogous to what exists for the physical sciences to be able to track the dynamics and specifics. These localized experiences that people have and that are accumulating to to determine their future responses. So one of the takeaways was this lovely quote about thinking beyond the zone that we really need to be able to understand the specifics and fluidity of these kinds of events to really understand what's going on cumulatively so I'm and we, we would like to leave a little time for people to get organized for the breakout so he should we turn it over to you. So I'm just, I'm going to stand in for Marshall really fast and talk about session two. And unfortunately I won't be able to do nearly as good a job as Marshall would have done but I do have a couple of notes with his assistance here. So thinking from session two. There were a number of really strong takeaways I think that there are multiple ways to define a compound or cascading disaster and how we do has consequences for people. And as a result the ethics around who's responsible for messaging and the classification of these events is complex and evolving. Hurricanes are multi impact events, there are open questions around how we convey that complexity, there are specific decisions that are made to focus on one hazard or another and what are the consequences of that. Thinking about benchmark storm benchmark storms, pardon me, and thinking about how they impact how people perceive other storms in their area. There's an opportunity to take advantage of that tendency that people have to help with relative messaging moving forward. Thinking about our traditional messaging tools, graphics and language they may send mixed messages particularly in some particular cases, and a strong coordinated community is really key to help maintain that message consistency. And that consistent and nimble messaging is really needed because every storm is different. And it can be different than people's previous experiences particularly in the context of a new normal of increasing extremes. And with that I'd like to transition really quick to Jeanette Sutton who's going to introduce our next session going into our breakout groups, and then I'll have some instructions for people in the room. We'd like to thank all of the speakers in the first two sessions, they were fantastic.