A successful field campaign collecting meteorological and hydrological data almost always involves a forecasting element. A detailed high quality forecast briefing on the potential evolution of weather events allows the scientists behind the scenes and out in the field to make crucial decisions that ultimately may determine the success of a project. While many joke that weather forecasters are allowed to be wrong half of the time, making a forecast that can affect the outcome of data collection is a serious task. Whether it be forecasting for strong winds (which affects the manual installation of multi-million dollar high resolution radar systems) or large hail (which can damage these expensive tools), providing the most accurate information possible was the primary goal for us at Iowa State University who provided a weather briefing every morning during IFloodS.
Each day, our forecast team of Dr. William Gallus, Ben Moser, and myself (Brian Squitieri) woke up bright and early to prepare the 9 AM weather briefing. We began each day by evaluating precipitation and stream data from the recent past to determine the extent of ground saturation present or the current height of the streams. This information is vital for making a flood forecast, particularly during a progressive weather pattern, where a series of atmospheric disturbances moving overhead may bring multiple days of heavy rainfall.
We as forecasters may determine a specific region or location that may be at greatest risk for experiencing flooding if we know the precipitation and river/stream gauge data for specific areas most affected by previous rainfall events and the current observations and model output –mathematical representations of the laws of physics used to simulate atmospheric features into the future. This is especially the case if model trends place the same area at risk for heavy rainfall that had experienced significant accumulations in the recent past.
While one of the main goals of the IFloodS project was to evaluate precipitation events, observing and understanding severe thunderstorm events proved to be an equally vital component of the mission. Not only do many heavy rainfall events originate from severe thunderstorms, but the storms themselves can prove to be very damaging and pose a threat to the researchers and their equipment. As such, we had the additional duty of providing the most up to date forecast information from observations and model runs regarding the potential development of severe weather. Details included what the primary hazards were (strong straight line winds, large hail, tornadoes, or some combination of the three), where the most likely regions of impact were and what time frame was most favorable for the activity to occur.
One challenging event requiring accurate forecasting of both heavy rainfall and severe weather occurred June 12, 2013, where eastern Iowa was under the gun for significant severe thunderstorms. Observations and model outputs showed that the atmosphere was primed for a potential outbreak of thunderstorms. Supercells were expected with an initial threat of large hail (perhaps larger than golfballs) and strong tornadoes.
Within a few hours of development, a line of powerful storms was expected to further develop as the supercells would merge with each other and produce a widespread swath of very damaging straight line winds. Our Iowa State University Forecast team had briefed the IFloodS project at 9 AM of the potential for a dangerous weather event, particularly for the eastern most Iowa counties in the IFloodS project domain. The Waterloo area was of specific concern because the N-POL radar was located there and was susceptible to hail and wind damage.
This forecast proved to be rather accurate, as a northwest to southeast oriented line of supercell thunderstorms developed and moved just a few miles north of the city of Waterloo. While the damaging tornadoes were mainly confined immediately to the west and east of I-35 during a 45 minute period, large hail and winds threatened the rural areas immediately north of Waterloo.
Overall, the IFloodS project forecasting tasks proved to be an enlightening, challenging and exciting experience for the three of us. We learned a lot, and were especially happy with the use of the NASA Unified Weather Research and Forecasting (NU-WRF) mode, which proved to greatly enhance the accuracy of the forecasts themselves. While this tool was experimental, its first 24 hours of output handled the placement, timing and intensity of rainfall events very well. We quickly noticed the skill of this product and openly praised its usefulness in operations. Lastly, we thoroughly enjoyed this experience, and though the work was difficult at times, we look forward to possible similar future collaborations.
From May 1 to June 15, NASA and Iowa Flood Center scientists from the University of Iowa measured rainfall in eastern Iowa with ground instruments and satellites as part of a field campaign called Iowa Flood Studies (IFloodS). They will evaluate the accuracy of flood forecasting models and precipitation measurements from space with data they collect.
As an undergraduate (from February 2008 to May 2009), I worked for Dan Ceynar, the engineer who coordinates the instrumentation networks for the Iowa Flood Center, and a large part of my job then and since returning to the Iowa Flood Center last October has been taking care of our rain gauges in the field.
For the IFloodS campaign itself, I have also been working with rain gauges. I have been a part of the process of developing these instruments all the way through their construction, deployment, and now maintenance.
We use what are called, tipping bucket rain gauges that are a couple steps up from what many people have in their yards. The most important part is a central swinging shaft: on the top are two small buckets, and on the bottom is a magnet. Each bucket holds exactly 1/100th of an inch of rain, and when one fills up it makes the shaft swing to the other side. As it swings, the magnet on the bottom passes over two wires and causes them to touch and complete an electrical circuit– and that’s how we record that rain has fallen.
The engineers I work with designed the electronics that sit under the rain gauges and record their tips. I built the electronics boxes, which sit inside the rain gauge platform and have a cell phone modem, computer, and battery, which the rain gauges use to communicate data back to us every 15 minutes. A network of these gauges like what we have now in the Turkey River Basin can give us very detailed information about where and when rain has fallen.
I also helped build the platforms they sit on — starting by picking up 50-gallon trough/tubs from a farm supply store in Cedar Rapids. All of this is our own design — Jim Niemeier’s and Dan Ceynar’s — and we just order the tipping bucket rain gauges themselves.
When I started with the Iowa Flood Center last year, I also built the latest batch of stream-stage sensors. The ones I built that have been deployed thus far are in the Ames area. So, they are not directly in the IFloodS area of interest, but they are identical to the ones we have around the state. So I still feel connected to them all!
The rain gauge and soil moisture sensor installations involve a fair amount of digging — and with the lingering winter, we did some of our digging with a pickaxe because the ground was still frozen solid! We had quite a few cold days, which made the few warm days feel like a gift. It was surprising a couple of times how much faster things went when our fingers weren’t cold and stiff!
I have loved getting to know this side of science research. Seeing all the different people and concerns that are a part of a project like this — getting to meet the landowners who let us install on their property — has been really cool.
Kara Prior is a research assistant at the Iowa Flood Center, where she helps oversee a network of rain gauges. She recently finished three years of study and teaching English in China and South Korea, where she earned a graduate certificate in Chinese studies from the Johns Hopkins University-Nanjing University Center for Chinese and American Studies. Kara also holds a BA/BS in international studies and environmental science from the University of Iowa. She has a strong interest in hydroscience and water resource management.
Vijay Mishra and I went to do a maintenance check yesterday on one of the 4 X-band radars that the University of Iowa is contributing to the IFloodS field campaign. Below is a photo of the radar, located at a topographical high point near Elkader, Iowa.
The X-band radar has a panoramic view of part of the Turkey River basin. Another University of Iowa X-band radar is located about 20 kilometers away and has a similar view of the watershed, although from a different perspective, of course.
A significant amount of rain has fallen during the IFloodS campaign, some areas within the campaign area received more than 11 inches of rain since May 1. The Turkey River is now and at or above flood stage.
Matt Schwaller is the GPM Ground Validation project manager with responsibilities for coordinating the development and operations of GPM GV in the pre- and post-launch phases of the GPM mission. He also occasionally takes the role of a field campaign mission scientist.