Table of Contents:
1. Overview/Setup
2. A brief model analysis
3. Methodology of Images
4. MODIS images
5. References
1. Overview/Setup
Many areas across central Iowa received over 6.00" of rain 1-11 August 2010 (*1, Fig. 1), with Ames, Iowa recording 10.99" of rain (*2, Fig. 2).
Fig. 1: Rainfall 1-11 August 2010 from the Iowa Daily Erosion Project (*1)
Fig. 2: Ames, Iowa preliminary monthly climate data from the National Weather Service (*2, NWS)
While the rainfall in the first third of August was a major factor in the flooding that occurred in Ames, Iowa, previous months were also unusually wet. The period of June and July 2010 was the 3rd wettest since records began in 1895 (*3, Fig. 3).
Fig. 3: NWS public information statement regarding temperature and precipitation (*3)June 2010 in Iowa was the 2nd wettest on record (*4, Fig. 4) and July 2010 in Iowa was the 4th wettest (*5, Fig. 5). All climate divisions in Iowa in June (*6, Fig. 6) and seven of the nine in July (*7, Fig. 7) reported "much above normal" rainfall.
Fig. 4: June 2010 precipitation rankings (*4)
Fig. 5: Same as fig. 4, but for July 2010 (*5)
Fig. 7: Same as fig. 6, except for July 2010 (*7)2. A brief model analysis
From a forecasting perspective, a quick glance at the NAM and GFS models would suggest that the general idea of high rainfall amounts (heavy rainfall) was likely across some part of Iowa. Forecasted precipitable water (PWAT) standard deviations from the 20100808 1200Z runs of the NAM and GFS models were +2 over at least half of Iowa from 08/2100z to 11/0300z. (*8) As the NAM model simulated precipitation over Iowa, PWAT standard deviations were +3 to perhaps as high as +4 from 09/0300z – 09/1200z (Fig. 8). While the GFS model (Fig. 9) did have an area of PWAT standard deviations of +3, it was not nearly as large as the NAM's. The concept of standard deviations greater than +2 associated with modeled precipitation over Iowa continued in the 20100809 and 20100810 12z runs of the NAM model (Figs. 10, 12 respectively), but not in the GFS runs (Figs. 11, 13 respectively). Finally, the SPC Mesoscale Analysis Archive (*9) indicated modeled PWAT values (not standard deviations) generally between 2.00” and 2.50” when thunderstorms were occurring over central Iowa (not shown).
Fig. 8: 20100808 1200z NAM forecasted PWAT (top panel) at 20100809 at 0600Z (*8)
Fig. 9: Same as fig. 8, except for the GFS model (*8)
Fig. 10: 20100809 1200z NAM forecasted PWAT (top panel) at 20100810 at 1500Z (*8)
Fig. 11: Same as fig. 10, except for the GFS model (*8)
Fig. 12: 20100810 1200z NAM forecasted PWAT (top panel) at 20100811 at 0000Z (*8)
Fig. 13: Same as fig. 12, except for the GFS model3. Methodology of Images
I obtained all images from the NASA/GSFC, MODIS Rapid Response website, which can be found at http://rapidfire.sci.gsfc.nasa.gov (*10). I selected the AERONET_Ames subset of images at a resolution of 250m. After downloading the images, I used Adobe Photoshop 7 to crop the image to central Iowa. Text, labels and pointers were also added. This was repeated for all images.
To the best of my ability, I tried to keep the changes of where labels were placed from image to image to a minimum for readability and comparison purposes. There are exceptions though, I know. Since I was primarily focusing in on Ames, I tried to make certain that labels would not block any point of interest in this area.
The images in this album are from three days: 10 August (pre-flood), 11 August (when rivers/creeks crested in Ames) and 12 August (waters beginning to recede). Each day has three images and will be displayed in this order: true color (satellite), NDVI and 721. What are those last two? Well, it's time for the acronym list. (*11)
MODIS: MOderate Resolution Imaging Spectroradiometer, instrument aboard the satellites Terra and Aqua
Terra is the satellite which orbits Earth from north to south across the equator in the morning
Aqua is the satellite which orbits Earth from south to north across the equator in the afternoon
NDVI: Normalized difference vegetation index (a color scheme)
721: 721 band to RGB mapping (another color scheme if I understand it correctly)
Also, note that on 10 and 12 August, the Terra MODIS was used because Aqua MODIS images were not as sharp. 11 August I used Aqua MODIS as Terra MODIS images were not available.
4. MODIS images
Fig. 14: One day prior to the 11 August 2010 floods from MODIS true color. The South Skunk River can be seen southeast of Ames, but not directly south of Ames.
Fig. 15. Same as fig. 14, except NDVI.
Fig. 16: Same as fig. 14, except 721.
Fig. 17: The South Skunk River is now visible just south of Ames. If you look closely, the Squaw Creek and South Skunk River in Ames are more readily identifiable than on 2010 August 10. Note that the blue circle west of Ames is part of the original image.
Fig. 18: Same as fig. 17, except NDVI
Fig. 19: Same as fig. 17, except 721.
Fig. 20: The South Skunk River south of Ames is still clearly seen the day after the rivers crested in Ames. A patchy cumulus field is over the right half of the image.
Fig. 21: Same as fig. 20, except NDVI.
Fig. 22: Same as fig. 20, except 721.
All images are from NASA/GSFC, MODIS Rapid Response, which can be found at http://rapidfire.sci.gsfc.nasa.gov (*10).
5. References (images are in order after the MODIS images)
*1. "Rainfall on [1-11] Aug 2010." Iowa Daily Erosion Project. Web. 12 Aug 2010.*2. "Ames Preliminary Monthly Climate Data (CF6)." NWS Des Moines, IA. National Weather Service, 12 Aug 2010. Web. 12 Aug 2010.
*4 "June 2010 Statewide Ranks." NOAA NCDC. Web. 12 Aug 2010.
*5 "July 2010 Statewide Ranks." NOAA NCDC. Web. 12 Aug 2010.
*7 "Jul 2010 Divisional Rankings." NOAA NCDC. Web. 12 Aug 2010.
*8 Model and Ensemble Forecast & archive viewer. NWS State College-Penn State Meteorology, n.d. Web. 12 Aug 2010.
*9 SPC Hourly Mesoscale Analysis - Archive. Web. 12 Aug 2010.
*10 MODIS Rapid Response System - Subsets. Web. 12 Aug 2010.
*11. MODIS Website. NASA, 12 Aug 2010. Web. 12 Aug 2010.
