Flash Flood Awareness for Central Texas

Flood Technology Workshop, Texas Flash Flood Coalition
LCRA, Austin, October 8, 2010

This is a blog was prepared by David Maidment of the comments that the workshop participants made on their perception of the overall needs as presented during the final session of the workshop.

Hatim Sharif (UT San Antonio) – focus on collaboration and bringing people together, sharing data and expertise.  The behavioral component is important.  Social science is needed.

Andy Rooke (AECOM) – More data is good but my father drives across low water crossing by his house – he’s driven across this many times and nothing happens. A virtual reality thrill ride in a flood is needed to create insight into what happens when a car is washed off a low water crossing.

David Walker (LCRA) – I remember the 1998 flood and a person drowned in the City of Austin and I felt bad about that. However, the Fire Dept said it was the first time they mobilized before a flood happened.  Focus on emergency management community. Develop a pilot project for IWRSS with GEMMS.  HydroMet information http://hydromet.lcra.org  is also accessible on a mobile app.

Cherie Rachel (UT Austin) – I am inspired but need a task force to develop a project management plan.

Luis Alday (Freese and Nichols) – Provide easy access to the data and information in schools – children tell you what to do.

John McEnery (UT Arlington) – More tools are needed for delivering hazard advisories.   NWS streams out flood information but this is a pretty big plate.  If Frisco can do it why not others?

Paul Rodman (US Army Corps of Engineers) – We don’t have all the data we need.  We’ve lost gages over the last few years.  Sharing data between agencies is critical.   We need to work with NCTCOG similarly to CAPCOG.  Get information from cities to help Corps do a better job of flood forecasting.  We need parallel communications in the Dallas- Fort Worth area to get information out to public.  There are problems with firewalls.  Security concerns are an issue.  I work with Boy Scouts and Cub Scouts councils in Texas.  Try outreach through them.  In 1978, we made a forecast that said water was going to go over a spillway and the evacuation was miscalled.  If you give a warning that is one thing but an evacuation order is quite something else.

Roy Sedwick (TFMA) – I am retired and travel, my wife has an iPhone, she does the navigating.  I’d like to pull up the radar app for iPhone that not only looks at current radar but also current watches and warnings.  Get the information where the public can get at it easily.  Why don’t car radios have NOAA weather radio broadcast frequencies on them?  What about Sirius/XM?

Bruce Barr (Tx Assn of Counties) – As you travel from one radar zone to another, the car would know which radar zone it is located.  XM radio can do this.  GEMMS from TNRIS is an important state-wide information system.  http://atlas.tnris.org/

Hector Guerrero (NWS) – We need to develop a project management group, work with Frisco and GEMMS to seek guidance.  For the sake of this group, focus on technology and deal with social science elsewhere in the coalition’s activities.

Tom Donaldson (NWS) – I’ll add to what Bruce was saying with the car radios – a civil defense mechanism.  NWS Warnings are on a polygon scale, not for the whole county, and with GPS on the cars, you should be able to warn people in small areas.

Amy Ronnfeldt (TxDOT) – The biggest immediate needs – we have to communicate with one anther – overcoming security problems.  I have a phone that runs apps but it’s not an iPhone – interest in writing apps for other phones.

Jack Kaiser (TxDOT) – TxDOT is restructuring – this is an opportune time for making comments as to priority issues.  TxDOT’s restructuring web site https://www.dot.state.tx.us/about_us/organizational_review.htm

Fernando Salas (UT Austin) – In order to organize this stuff we need to think of the perfect product.  During Tropical Storm Hermine, I surfed the web, I looked at HydroMet (http://hydromet.lcra.org), and I thought it was good for conveying lots of information effectively.  The NWS web site (http://www.srh.noaa.gov/wgrfc/)  displays observation sites they have and it puts them on a map and you can get stage – they classifying them for moderate, extreme, flood stage – we need relative height.  Twitter is really good to convey information but it is hard to censor it.  Austin Statesman twitter page processes them out for censoring purposes.  The UT shooting incident was the most tweeted thing in the world on that morning.  If you censor this through the media then you get something better.  Pull twitter information into a database.

Gordon Wells (UT Austin) – There is lots of unstructured information in international events – we need to validate and put spatial geotags on that. Outside USA, Haiti, etc.  General issues of dealing with social media – how do you combine this with instrumental data?  We have the same issue on storm surge with tidal data – it would be very useful to mobilize adhoc censor networks with data transmission to get that data to an internet gateway.  This is done successfully for coastal storm surge to densify networks on a temporary basis with disposable sensors.  We lost the primary tidal gages during the landfall of Hurricane Ike and its really the adhoc network of pressure transducers that is critical to reconstructing the event after it is over.

Tim Whiteaker (UT Austin) – My road was flooded out during Tropical Storm Hermine and Travis County published road closures on a web site – I wrote a Google Gadget that informed me when my road was open again.  Apps are a great way to reach population en masse.  Design an app that is very simple.  (Bruce Barr commented that there are issues with legal liability with rerouting traffic). 

Teresa Howard (UT Austin) – We have data overload and data gaps – let’s say you do have your modeling and can you solicit information from people there about current conditions.  Are there ways that you can filter too much information to get what you need.  Volunteer rain network called CoCoRaHS http://www.cocorahs.org/  – is there a waterwiki?  E-extension does this?  We need communities of practice that get together. 

Erika Boghici (ESRI) – what can we do to increase the time for warning?  This is a wealthy part of Texas with good data and good services but not a lot of collaboration.  Can we centralize all the information a single place?  We need an Information Cross-Roads.  Would it be difficult to create base data to which we can link all the rest?

Ilyanna Kadich (CAPCOG) – In the education realm,  Nov 17 is GIS data and we can volunteer to go out into schools and educate children on this day.  We are working with the Girl Scouts also on education issues.  You could contribute to CAPCOG’s GeoMAP program which pulls resources from many sources to create collaborative effort for data collection.

Traci Tracey (ESRI) – From a GIS perspective, we’ve talked about this with other state agencies and making your data relevant outside your community is going to make your data more critical. You data is very important for those outside the scientific community.  Why aren’t we engaging GenY through mechanisms such as a Facebook account?  Education, education, education.  When the road is underwater, it can be washed away and you don’t know the road is not there.  Just because it’s not raining here it doesn’t mean there’s no rain upstream.

Ron Anderson (LCRA) – This situation with data sharing for flash flooding reminds me a lot when the Clean Rivers data got kicked off and there was a similar need to merge data.  You need consistent data standards and standards for different data collection methods.  I get weather data from GPS in my car – it’s just a matter of getting the information to Clear Channel Communcations (http://www.clearchannel.com/) .

Bob Rose (LCRA) – To follow up on what others have said – can we establish clearinghouse for social media?  This is valuable information for NWS, and perhaps turn it around for others – start a Facebook page.

Rob Wells (LCRA) – Not all instantaneous data is good data and those who own the data know that but don’t have time to pass it on during flash flood emergency.  You may spend more time dealing with reaction to bad data than dealing with good data.

Jonathon Dobbs (Tx Dept of Emergency Management) – Get ahead of the flash part of flash flooding I would like to see this turn into something that could notify an EMC in real time not after the fact – receive data that a stream gage has hit --- and start issuing warnings and evacuations. (Karl Winters commented that the USGS has a customized alert system for notifying users of stream gage readings).

Bob Huber (LCRA) – The media area grabbing some of this – TV industry is changing – GOES is in flux – and there are various information services for the television industry.  NWS have been the ones whom people look to for authoritative information.  Streamflow is of interest to some people but not to others – everyone wants to see Lake levels but perhaps it would be good if they could become more aware of stream gage levels for particular gages.

Brian Reece (USGS) – We have approach it from two ends – we need apps and web services and collaboration technologies to reach end users.  We need to spend some time on the back end on the databases and the people who manage them – web services happen when you have better data.  A college student in San Diego may say water comes from faucet but it actually comes from Northern California.  Educate the generations through that and then you build support.

Susan Janek (City of Austin) – You need a system that is centralized in terms of flood observation data collection along with LIDAR for land elevation – this requires investment in the data backend.  Look into some public internet sites for guidance about how to present information.  The HydroNet mobileapp rocks!  Lots of data are available right now.  Perhaps develop a Super HydroMet to synthesize information from across the region.  A caution on social media – too much loses its meaning.

Karl Winters (USGS) – Paul Rodman highlighted the need for good data – how do we know that the stage data is good – we need ground truthing of flood events.  A series of flood marks along a stream reach helps to identify an indirect measurement of stream discharge.  There 7-8 indirect measurements done in the last month in which USGS back calculated the stream discharge. – wouldn’t it be great if we had going some MOAs how to do data collection and analysis as a flood response.  Rapid response is really necessary because subsequent rains and weathering change the landscape.  Rapid deployment of gages to monitor floods.

Tomislav Urban (UT Austin) – one of the functions of the Texas Advanced Computation Center is data collection and one of the purposes for that is emergency response – the data are stored in Corral, a 1.2 PB disk storage.

Paul Yura (NWS) – I am NWS warning coordination meteorologist.  I see the end result – death.  We should realise that floods affect a small % of the population that doesn’t have apps, twitter, etc.  We need to focus on increasing the time on the warning.  I was there at 2AM in the morning for Tropical Storm Hermine.  The 911 centers were too busy to answer the phone.  EMC’s people were too busy doing their jobs  to communicate with the NWS and the flow of information stops.  I was going from web site to web site searching for stuff (LCRA HydroMet http://hydromet.lcra.org, City of Austin FEWS http://www.ci.austin.tx.us/watershed/floodwarn.htm , UBCWCID dam levels http://ubcwcid.org/ ).  Concerning road closure maps – the map doesn’t tell you how bad it is – is there 3 feet of water on it or just a preventative closure?  I was looking at TV trying to figure out what was going on.  We need data that comes from the low water crossings, such as those in Comal county – we need you to give us information about the roads that are closed, and storm reports. There is a Tornado emergency.  We considered issuing a Flash flood emergency for Tropical Storm Hermine and were very close to doing that but felt we did not enough information to make the call.  We need stream spotters – the guy that lives by the stream.  Training is all about weather – we need training about floods.  Talking to emergency managers is useful but there was a lack of data flow during those 3 hours.  It was the middle of the night. There were barebone crews at TV stations, not the prime time people.  It was just after Labor Day weekend.  There is a Regional Web EOC but counties have their own WebEOC accounts so not much could be learned from regional WebEOC.  Counties are not created equal.  Emergency management is a responsibility with varying people who do it.

Larry Quick (Upper Brushy Creek Water Control and Improvement District) – as a nonprofessional I hear a lot of talk about DOD working on things.  Can we introduce these into the flash flood field?