EIIP Virtual Forum Presentation — June 11, 2008

The Science of Flood Safety
U.S. Geological Survey and National Weather Service Perspectives

Brian E. McCallum
Assistant Director of Hydrologic Monitoring
Georgia Water Science Center, USGS

Kent Frantz
Senior Service Hydrologist, Atlanta (Peachtree City) WFO
National Weather Service

Amy Sebring
EIIP Moderator

The following version of the transcript has been edited for easier reading and comprehension. A text transcript is available from our archives. See our home page at http://www.emforum.org

[Note: Kent Frantz was unable to participate as scheduled. His prepared comments were included for the presentation part of the program.]

[Welcome / Introduction]

Amy Sebring: Good morning/afternoon everyone. On behalf of Avagene and myself, welcome to the Virtual Forum! We are pleased you could join us today, including any first-timers. We want you to feel at home.

Our topic today is The Science of Flood Safety: USGS and NWS Perspectives. Please note there is an unscientific poll with a related question on our homepage: "What precision should be used in flood stage forecasts?" As of this morning we had 50 votes, so if you have not voted yet, please take a moment to do so after our session.

Now to introduce our guests. Brian McCallum has been Assistant Director of Hydrological Monitoring for the USGS Georgia Water Science Center located in Atlanta since January 2000. In 1993, he started full-time with the USGS in Baton Rouge, Louisiana, where he helped establish the Louisiana HydroWatch, a statewide hydrologic monitoring network. Brian continues to be active in the National Hydrologic Warning Council, a group for flood warning system operators.

Kent Frantz has served as Senior Service Hydrologist for the National Weather Service's (NWS) Weather Forecast Office (WFO) at Peachtree City, GA since 2005. Previously he served as Hydrologic Analysis and Support (HAS) forecaster for the NWS Southeast River Forecast Center (SERFC) from 1999-2005. For approximately 20 years before assuming the SERFC position, he served with NWS as meteorologist and meteorologist technician at WFO's in the southeast and midwest. He served four years in the U.S. Navy as Aerographer's Mate for weather.

Please see the Background Page for further biographical details, and you may want to check out the podcast interview with Brian for additional brief remarks, which is linked from both the Background Page and home page.

Now, we are currently missing Kent, and hope he has not had an emergency. I will assist with inputting his prepared remarks during the overview and hopefully he will be able to join us in time for the Q&A.

Welcome Brian and I turn the floor over to you to start us off please.


Brian McCallum: I would like to thank Avagene and Amy for allowing Kent and I to chat with you about a resource that many emergency managers may not be aware of, but certainly depend heavily upon—the U.S. Geological Survey (USGS) streamgaging program. I am glad that Kent Frantz, an avid supporter of streamgages and a close colleague in fighting floods in Georgia, has been able to join me for this discussion.

Kent Frantz: Thanks for asking me and I am glad to participate. The USGS streamgaging program is vital to the NWS Flood Warning program. It is the ground truth information we need for warning decision-making. Simply put, our service hydrology program would not be nearly as accurate or specific without the USGS data.

Brian McCallum: A little background on how we got to this point first. I met Avagene at the IAEM Region IV Conference this past April where I presented this topic to the emergency managers in attendance. It was well received there, and Avagene felt this was a suitable topic for this forum.

So, let’s get started. The USGS is one of the most respected science agencies in the Federal government and the world. As part of the Department of Interior, you probably know of us either because of topographic maps or earthquakes and volcanoes. In fact, one of our first tasks as an agency was to begin monitoring water in streams across the United States.

Why is it important to monitor water? Quite simply—water is a strategic resource for the United States.

  • Life depends upon it,
  • Economic growth relies on it,
  • Our quality of life revolves around it.

We, as a society, place significant value on clean and abundant water.

However, when excessive rains come, lives and property are at risk and YOU need to know when a river flood will occur and how high. That is where the partnership between USGS, NWS, and the local emergency manager is critical. On the other hand, when the rains don’t come, USGS streamgages tell you how much water is left during droughts.

Kent Frantz: The USGS real-time monitoring network provides continuous data for ALL users from local officials and water authorities to public recreational activities. I use the data for monitoring high water events and potential flood warnings to low streamflows and drought information statements.

Brian McCallum: As you can see from this first slide, the average annual impact of hydrologic hazards to the United States is quite significant, in excess of $15 billion and more than 250 lives lost. And this doesn’t include Katrina or any other significant event since 1995! Amy, slide 1 please.

[Slide 1]

We monitor water because we value water...for work, for play, and for life. Informed decisions are required to manage our water resources and the USGS streamflow data, combined with NWS forecasts, provides the information for you to make those informed decisions.

All USGS real-time streamflow data are available at http://water.usgs.gov. We call it "from the stream to the screen." From this URL, you can get to many of the links listed in this transcript.

Starting with the map of the United States that shows the nearly 7,400 streamgages in the USGS network, you can quickly see the hydrologic conditions across the country. The blue and black dots represent gages that are measuring high river flow conditions, like the recent Midwest floods this past April. The red dots represent stations that are measuring extremely dry conditions, like the drought in the southeast U.S. Amy, slide 2 please.

[Slide 2]

From the Website, you can click on a state, and see a map of all the gages in that state. Then, you can click on a dot, and get to a graph of river flow at that location. Important to note are the triangles that represent "normal" flow conditions, and the green line designates the flow at the NWS flood stage.

Flood stage is determined to be the height of the creek or river where the "Minor Flood" category begins. Normally in Georgia, this level is where the natural flood plain begins to flood with minimal impacts to the public. It is usually considered to be nuisance type flooding of lowlands, trees, pastures and other agricultural areas. Some river access roads and boat ramps may be affected. The level is established by your local NWS Service Hydrologist through a survey, and in coordination with the county EMA director.

Streamgages are used for:

  • NWS flood forecasts and warnings
  • Creating FIRM maps for FEMA
  • Planning, designing, and operating flood management infrastructure (dams, levees, etc.)
  • Drought monitoring
  • Climate change trends
  • Hydropower generation
  • Agriculture and public water supply monitoring
  • Wastewater effluent discharges
  • Hurricane surge monitoring
  • Habitat protection
  • Navigation and recreation

So what is a streamgage? The next picture shows a coastal streamgage that continuously collects water level, streamflow, rainfall, wind speed/direction, and even water quality data. Most transmit via NOAA satellite, normally every hour, and every 15 minutes during severe storms. Amy, slide 3 please.

[Slide 3]

The USGS receives it directly from the satellite system and posts it to our database and then automatically to our webpage.

Kent Frantz: The USGS data is transmitted to the NWS through our HADS. HADS is an acronym for the Hydrometeorological Automated Data System, a real-time data acquisition and data distribution system operated by the Office of Hydrologic Development of the National Weather Service. HADS web pages are structured to provide necessary system information and site meta-data to National Weather Service Weather Forecast Offices and River Forecast Centers.

Brian McCallum: Many folks ask "Why do you need streamflow and not just water level?" The answer is that streamflow, or discharge, which is the quantity of water passing by the gage, is crucial for river modeling and flood forecasting. We relate streamflow to the water level collected by the gage, which we call a "rating".

Kent Frantz: Streamflow or discharge is in cubic feet per second (CFS). CFS is more of a constant physical element that can be used at any location while stage height is relative and variable to each location. For example a stage height of 3 feet on a river can mean something totally different from a stage height of 3 feet on a creek. CFS is used in the river model for more accurate volume data and when routing water downstream.

Brian McCallum: We must visit and inspect all of these gages usually every couple months to ensure the sensors are calibrated and to make a discrete streamflow measurement to verify our stage-streamflow rating. This is because the river channel changes over time due to sediment buildup or scour, vegetation growth, beaver dams, and many other reasons. These all affect the amount of water flowing in a river relative to the water level.

So we have explained what a streamgage is and how the data are used, but how do we quantify the benefits to the public? The National Hydrologic Warning Council (NHWC) recently released the results of a study investigating the "Flood Management Benefits of the USGS Streamgaging Program," which is available at http://water.usgs.gov/osw/pubs/nhwc_report.pdf.

In it, they focused on only three of the many uses I previously listed that were specific to floods:

  • FIRM mapping by FEMA
  • Designing flood management infrastructure (dams, levees)
  • Flood warning and reservoir operations

They reported that long-term streamgage data was critical for FEMA flood maps, which cost $56 million annually. Long-term streamflow data allowed for better levee design, saving more than $70 billion in the construction of the 10,000 miles of levees in the United States.

Finally, it was estimated that the value of accurate flood forecasts by NWS represent more than $1 billion annually. If you attribute 5% of that value to the USGS streamgaging program, that is more than $50 million annually. It was documented that in 2004, the cost to operate the entire USGS streamgaging program was $114 million.

"In the aggregate, nationwide, the benefits of gages in context of reducing flood damages greatly exceed the costs of collecting the data used for decision making." —from the executive summary of the NHWC report.

Kent Frantz: And that is only for those three uses. They have many other benefits not always documented. Yet, despite these benefits, gages lose their funding and are under threat of shutting down all the time.

Brian McCallum: The next slide depicts the tenuous nature of keeping streamgages funded. The graph on the left shows that over that past 25 years, the USGS program has lost more than 2000 gages with long-term records. These are when gages become most valuable. They are like a fine wine—streamgages get better with time.

When you have long-term data records from these gages, you are able to more accurately determine the flood statistics, like the 100-year flood level, and put the current data into better historical perspective.

The graph on the right shows the funding distribution for the gaging program, and only a small amount (blue area) is dedicated strictly to funding streamgages, which is called the National Streamflow Information Program (http://water.usgs.gov/nsip). Amy slide 4 please.

[Slide 4]

The yellow represents the matching dollars that USGS use in cooperation with other state and local agencies to gages (http://water.usgs.gov/coop). These dollars are dependent upon the cooperators needs, which can vary over time—making long-term gages vulnerable. The red are other Federal agencies contributions, like the Corps of Engineers. Thus, the USGS is extremely dependent upon its cooperators to keep gages running, even as Federal dollars to USGS are reduced.

Kent Frantz: The NWS budget does not receive any federal money for streamgages. It was determined years ago to reduce redundancy between federal agencies that money designated for this should go directly to the USGS. Through USGS and NWS collaboration, potential streamgages are installed where there is the greatest need for public safety and benefits to all.

So, to bring the presentation to a close, I will end with the question "What does all this mean to you as a local EMA director?" Your job is to protect life and property from flood hazards. You depend upon the National Weather Service to provide timely and accurate flood forecasts and warnings. And we at the NWS depend greatly on the USGS streamgaging program and the data it provides.

Brian McCallum: Hopefully we showed that the USGS streamgaging program is an extremely valuable resource to you as a local EMA director. However, this resource is eroding due to funding cuts and cost-of-living increases.

Kent Frantz: It is extremely important the EMA community speak loudly in support of the USGS streamgaging program to local officials and others involved in the budgetary process.

Brian McCallum: We encourage you to work closely with your local USGS and NWS offices. They are there ready to assist you. Another way to get involved is to join the National Hydrologic Warning Council (NHWC). This is a group of professionals involved with the operation of real-time hydrologic monitoring networks for the purpose of protection of life, property and the environment. They work closely with the emergency management community.

The NHWC is having their 2009 conference in Vail, Colorado May 18-21. The USGS and NWS are major participants in this conference. ASFPM members get continuing education credits for attending. With that, I will stop and turn it back over to Amy for questions. Thank you again for your time today.

Amy Sebring: Thank you very much Brian. Unfortunately, Kent still seems to be missing in action and we apologize.

[Audience Questions & Answers]

Steve McGee: Could you expand on HADS / streamflow data please?

Brian McCallum: I will give it my best shot since Kent isn't here. The NWS gets the data directly from the streamgages using satellite telemetry through the GOES satellite. It is received at Wallops Island, VA. Where it is then piped into the HADS system and distributed throughout the NWS network for forecasting use.

Amy Sebring: HADS is a system internal to NWS, correct?

Brian McCallum: Yes, that is correct. We receive the satellite data right at our office in Atlanta after it is re-transmitted via a commercial satellite. It is automatically posted to our database.

Steve Bays: A link for more info on HADS: http://www.weather.gov/oh/hads/

Isabel McCurdy: What is the cost of a gage?

Brian McCallum: Isabel, the cost varies from state to state. In Georgia, it costs $23,000 to install (one-time cost) and $13,100 in FY 2008 to operate. That translates to a dinner for 4 to keep a gage running each day.

Ric Skinner: In my "former life" as an aquatic biologist we used the Strahler Stream Order concept to estimate stream flow effects when we didn't have stream gauge data. Is that concept still relevant?

Brian McCallum: I am not familiar with that name. We do a lot of drainage basin ratios to estimate between gaged and ungaged locations.

Andrew Rooke: As a FEMA mapping contractor, I can attest to the critical value of streamgages in the calibration of streamflow models, which are used for floodplain mapping and flood forecasting. And, as Brian mentioned, long-term gage data is the most valuable.

Brian McCallum: Unfortunately we have lost over 2000 in the past 20 years.

Tom Bean: I represent a local government that is a USGS cooperator. We fund several river gages that we use for flood warning purposes. We are working to improve our automated use of this data, for instance by establishing a voice mail system that will allow citizen access to synthesized voice flow data. Might USGS have standardized tools to help with this type of application? If not, might USGS develop such tools?

Brian McCallum: We are continually looking for ways to get the data out. We here in GA have a thing called Streamail, which delivers gage data to cellphones and PDAs. If you know the USGS gage number, you can type "[email protected]" and put the gage number in the topic line. It should work nationwide. It takes a minute or two to return the values. Give it a try. It was developed here locally, but can be used across the country.

Tom Bean: Thanks, that's a new one to me.

Matt Raschke: Have more gages been lost because of lack of operating funds or lack of capital replacement funds?

Brian McCallum: Mostly lack of operating funds. The local cooperator has funding cutbacks or changing priorities which make it difficult to keep long-term gages running. To counteract this loss of local funding, the USGS worked with Congress to start the National Streamflow Information Program (NSIP). NSIP is designed to build a Federally-funded backbone of gages. Unfortunately, it has only been funded a fraction of the full implementation costs.

Bill Newman: Where in Atlanta do we have a gage? I'm in Gwinnett.

Brian McCallum: Bill, if you go to http://ga.water.usgs.gov and click on real-time streamflow, you'll get a list of 250+ stations in GA. We have over 15 in Gwinnett County, and some along the Hooch (Chattahoochee non-Georgians).

Ric Skinner: Are there any advances in real time flood mapping?

Brian McCallum: Ric, yes. The USGS, NWS and State of North Carolina are doing an intense real-time mapping initiative, I believe on the Tar or Neuse River. We also are working with the City of Albany, GA to create flood maps before the storm hits, so they can see what houses are affected at various river levels.

Linda Hutchins: We would love to have those in MA, too. We have been begging to get started on inundation mapping, but there doesn't seem to be funding available for it. Unless anyone knows something I haven't heard about

Brian McCallum: My best advice would be working with FEMA or your state EMA.

Andrew Rooke: The NWS is also expanding the real-time initiative into other states (e.g. Texas). Most of these are based upon pre-generated static maps that can be pulled up very quickly, based upon stage forecasts.

Brian McCallum: That is very similar to what we are doing in Albany.

Steve McGee: Does the Hydrology XML Consortium (HydroXC) work closely with the CAP Common Alert Protocol folks?

Brian McCallum: I wish Kent were here to answer that. I know there was a CAP talk at the IAEM Region IV meeting I was at, and they seemed to be communicating.

Brian McCallum: For all you ASFPM members, please go to http://www.hydrologicwarning.org for the NHWC. With your membership and attendance at the 2009 Vail conference, you get 12 continuing credits in ASFPM.

Avagene Moore: Brian, to bring this home and make timely application in light of the serious flooding going on right now in Indiana, if there were no streamgages, how much more seriously would the people in the proximity of the flooding and downstream be impacted?

Brian McCallum: Basically, the forecasters would be flying blind, not knowing what the river is doing and not being able to warn people about what is heading towards them. A NWS forecaster in KY talked about a flood that claimed lives due to the loss of streamgages on the Licking River. He basically said "He was in the dark." The nation is being impacted more and more by hydrologic events such as floods, droughts, hurricanes, etc., and the forecasters have got to have the gages to let people know what is going on.

Amy Sebring: Speaking of funding, I have a question. Brian, USGS HQ at least works with the Natural Hazards Caucus in the Congress. Is the funding need for NSIP being communicated to that group?

Brian McCallum: The message seems to be finally getting through because while other agencies are getting cuts, NSIP is getting small increases. Unfortunately, our cooperative funding program (coop program) which funds even more gages than NSIP, took a large cut in FY 2008 and is projected to again in FY 2009. This is a program where we can match 50/50 with a local or state cooperator for gages.

Chris Trice: Once a gauge has been lost, can it be started again with the necessary funding?

Brian McCallum: Chris, yes, usually. It depends on whether the gaging house structure has been removed many times. Sometimes it’s a matter of just putting the instrumentation back in.

Steve Bays: Has there been discussions on O&M increases due to fuel cost?

Brian McCallum: We have to have these discussions with our local cooperators each year as costs increase. The added fuel costs have put everyone in a bind, including our programs. It will come down to funding priorities for the cooperator again.

Steve McGee: Do the Chinese have a similar system to ours? Would be useful with those earthquake lakes. High percentage of world's population along those rivers.

Brian McCallum: Many countries use systems similar to ours. USGS has an international outreach program that assists other countries in transferring the science such as China, Iraq, etc. Definitely the gages would help for the quake lakes. The problem comes down to how to telemeter the data. The GOES system covers North America and South America primarily. But the instrumentation vendors have other transmission options available internationally.

Andrew Rooke: There are a number of systems similar to the USGS gaging network, which are in use internationally. ALERT is a protocol that has been used for several decades, and the group at NHWC is well versed in these systems.

Brian McCallum: Andy is correct, and there are a lot of two-way radio and other line-of-sight types too.

Amy Sebring: Brian, does the USGS work with NWS for public awareness and outreach?

Brian McCallum: Yes, quite a bit. We do segments and presentations like this all the time with EMA directors. We work together to create Flood Tracking Charts for major river basins, and try to find any other useful means to easily communicate to the general public. The key is putting the data into context for the general user. For example, we relate a gage reading to elevation, and then compare it to the past five major floods in the basin so they can remember how high it got "back in '98" for example, and take precautions.

Ric Skinner: From the time a request for a gauging station is received by USGS, how long does it take to put in place, calibrate, and add to the data system?

Brian McCallum: For us here in Georgia, we can usually get a gage installed in a day or two. If the equipment is on hand, the turnaround time is pretty quick. The thing that takes time is developing the streamflow-water level relation (rating) because it requires discrete flow measurements over a range of water levels.

Once the rating is established, continuous real-time streamflow can be computed and used for forecasts. It is critical to maintain these ratings over time because the river channels change, which affect the water level-streamflow rating, such as due to sediment buildup or scour, debris buildup, vegetation growth, etc. All these affect streamflow and its accuracy.

Josh McSwain: Is it possible to send a request for data (maybe through NWIS web) asking for data values for a group (not just a single gauge) of gages and have the data be available in a table (txt. csv. etc.) for download?

Brian McCallum: Yes, if you email me Josh, I can show you how. [email protected] Phone: (770) 903-9172.

Amy Sebring: Can you tell us just a bit more about NHWC Brian, who is in it, purpose, etc?

Brian McCallum: The NHWC is a group of hydrologic warning systems operators and users. We meet biennially to transfer knowledge and experiences for the support of folks wanting to start new local systems of their own. The USGS and NWS are major players in NHWC. We get all the different vendors to show their wares and the latest technologies for collecting, processing, and disseminating real-time data. And we have a little fun too!


Amy Sebring: Let's wrap it up for today. Thank you very much Brian for an excellent job and we appreciate your time and effort. So sorry Kent must have gotten waylaid. Please stand by just a moment while we make a couple of quick announcements.

Again, the formatted transcript will be available later today. If you are not on our mailing list and would like to get notices of future sessions and availability of transcripts, just go to our home page to Subscribe.

Before we adjourn, please take a moment now, or after you review the transcript to Rate today's session and/or write a review or post your comments. You can access the form either from today's Background Page or from our home page. If you do not have time to write a short review or comment, then please just take a moment to do the rating. It should take less than a minute, and will assist future visitors to our site to find useful information.

Please join us next time, Wed. June 25th, when our topic will be the Church Disaster Mental Health (CDMH) Project. See http://www.churchdisasterhelp.org/index.html.

Note: We are now posting our upcoming schedule under "Coming Soon" on our homepage. Currently scheduled are:

  • July 9 - New Madrid Fault Catastrophic Planning
  • July 23 - Target Capabilities Implementation Project
  • Aug 13 - Public Assistance Cost Recovery
  • Aug 27 - First Responder Guide to the Media

Becoming an EIIP Partner is a way to show your support for what we do, and possibly help us to keep the services we provide available to you. It is easy to do; see the link to Partnership for You from our home page, and complete the simple form provided.

Thanks to everyone for participating today. We stand adjourned but before you go, please help me show our appreciation to Brian for a fine job.