More on Use of the NASA LIS for Assessing Flood Potential…

As Brian’s previous post stated, we’ve been looking more closely at the NASA 1km LIS soil moisture values over northern AL and southern middle TN while attempting to assess flooding potential for several rain events of late.  Brian highlighted an event from the 26th, but we’ve also been looking at earlier cases during the past week.  On the 20th, as one such event was approaching, promising to deliver around 1 to 3 inches (or more) of rain with two systems in quick succession, I noticed elevated 0-10 cm relative soil moisture conditions across the area (Figure 1).

Figure1. NASA LIS, Alabama 1-km Domain, 0-10cm Relative Soil Moisture (0900 UTC 01-20-2012)

So, this might beg the question…”what is elevated?”  Well, it’s all relative of course.  But, a preliminary analysis of previous events suggests that when 0-10 cm relative soil moisture values reach 60-65% or higher, the flash flooding and/or areal flooding threat increases markedly with a “typical” 1 to 3 inch rainfall event.  Now, as mentioned, this is still all preliminary and determing the potential for flooding can be complicated and involve several variables (e.g. soil moisture, rainfall rates, land use, etc), but forecasters like to be able to put a value on soil moisture saturation.  Often, this is done in a very qualitative and rather subjective way.  If it “seems” to have rained a lot recently, a forecaster may perceive soil moisture values to be elevated.  But, it’s always nice to have a number to put to the actual degree of saturation.  The NASA LIS data provide just that.  With that said, it’s still important to keep in mind that these numbers represent model output.  Fortunately though, we are blessed here in northern AL with a number of USDA Scan sites (Figure 2) that monitor soil moisture values at various depths with which to do grount-truth comparisons.

Figure 2. Map of USDA Scan sites in Alabama. Notice the relatively dense network in northern Alabama.

All of this acts to increase forecaster awareness of soil moisture values across the area, and raise confidence in a given flooding scenario.  So, with this information in mind on the 20th, forecasters elevated the threat for flash flooding and even mentioned this threat while referencing NASA LIS soil moisture values in the afternoon Area Forecast Discussion (AFD), as noted in the Hydrology section below…

“.HYDROLOGY… HEAVY RAINFALL FROM RECENT EVENTS HAVE LED TO ABOVE AVERAGE SOIL MOISTURE CONDITIONS…RATING IN THE 60-80% SATURATED ACROSS THE AREA. WITH 1-3 INCHES OF RAINFALL EXPECTED OVERNIGHT…FLASH FLOODING IS A LARGE CONCERN. IN ADDITION TO THE FLASH FLOODING THREAT… AFTER THIS RUNOFF DRAINS INTO AREA CREEKS AND RIVERS…RIVER FLOODING WILL LINGER INTO EARLY NEXT WEEK. A FEW SITES ACROSS THE AREA HAVE JUST RECENTLY CRESTED WITHIN THE PAST 24 HOURS FROM HEAVY RAIN 2 DAYS AGO…AND THESE SITES WILL LIKELY SPIKE BACK ABOVE FLOOD STAGE QUICKLY.”

So, what happened?  Well, as all forecasting events can prove to be challenging, this one was no less so.  Expected rainfall amounts did not quite materialize with this first system, as shown in Figure 3 and Figure 4.

Figure 3. Stage IV Precipitation amounts for the 24 hour period ending 12Z Jan 21, 2012.

Figure 4. Stage IV Precipitation amounts for the 24 hour period ending 12Z Jan 22, 2012.

As the figures above show (and, sorry about the fuzziness of the images), precipitation amounts totaled around 1 inch for this first event, which was towards the lower end of what was expected.  However, this was followed up by a heavier rain event late on Sunday into Monday morning (Figure 5).

Figure 5. Stage IV Precipitation amounts for the 24 hour period ending 12Z Jan 23, 2012.

Notice the swaths of heavy rainfall (~1-3 inches) that did occur with this event.  The NASA LIS 0-10 cm relative soil values by the morning of the 23rd (0900Z) show the increase in soil moisture values after most of the heavy rain had fallen (Figure 6).

Figure 6. NASA LIS, Alabama 1-km Domain, 0-10cm Relative Soil Moisture (0900 UTC 01-23-2012)

So, with the generally elevated soil moisture values and this 2nd round of 1-3 inch rainfall…did we get flooding?  Yes, quite a bit actually.  You can click here to see the many flooding reports received at our office late on the 22nd into the 23rd.

2 thoughts on “More on Use of the NASA LIS for Assessing Flood Potential…

  1. This is a great post, Kris! What started as an application to aid in summer thunderstorm forecasting at NWS BHM has expanded thanks to Brian’s efforts and yours. For those interested, let me offer some more details on what the SPoRT/LIS is, and how we’re running it.

    The SPoRT configuration of LIS currently runs the Noah land surface model (LSM) in real-time on a nested grids: a 3-km resolution domain covering must of the eastern half of the U.S., and a smaller 1-km resolution domain centered on Alabama. The Noah LSM runs apart from any prediction model with atmospheric analyses providing the information needed for solving the soil equations. Currently, the NCEP Global Data Assimilation System, North American Land Data Assimilation System, and Stage IV precipitation analyses drive the integration of LIS. We also incorporate real-time MODIS greenness vegetation fractions at 1-km resolution (updated daily) into the SPoRT LIS, thereby improving over the default climatological vegetation.

    The eastern U.S. 3-km LIS output is posted to a SPoRT ftp directory, which can be accessed by the WRF Environmental Modeling System (EMS). As a result, real-time local modeling initiatives using the WRF EMS can initialize land surface variables using the SPoRT LIS. The LIS data can improve representation of soil moisture in local model initial conditions through higher resolution and real-time vegetation information.

  2. Thanks for the additional information Jon! We do plan on ramping up our local modeling efforts as well here in the near future!

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s