RE: [GRASS-user] Ability to model flash flooding

GRASS GRASS user
1

Hello Ian,

To learn how to use grass I've been trying to solve a real world
problem. The problem that I chose was the flash flooding which occurred
in south eastern Queensland. I'd like to see if grass could have
provided reasonably accurate predictions and see if it could be used as
a warning system.
Ideally what I'd like to do is use the archived radar rainfall
information from http://www.bom.gov.au/products/IDR503.loop.shtml#skip
which can be found on http://gw.barratt.com.au/radar/IDR503/2011/01/10/
I am intending extrapolate the rainfall rate using the Z value
conversions recommended for east coast lows as per
http://www.catchment.crc.org.au/pdfs/technical200207.pdf
and then convert the rainfall rate rasters to the resolution of the
aster dem.

...

Any thoughts would be appreciated.

some earlier weather-radar related work with GRASS can be found here:
http://gisws.media.osaka-cu.ac.jp/grass04/viewpaper.php?id=44

http://onlinelibrary.wiley.com/doi/10.1111/j.1467-9671.2004.00178.x/abstract

http://opus.bibliothek.uni-wuerzburg.de/frontdoor.php?source_opus=759&la=en [-> the document is in german, but the technical appendices might be of interest for you -> Z/R realtionship, etc.]

FWIW here's study concerning visual quality assessment for 3D radar-met data:
http://www.docstoc.com/docs/69767645/4D-data-visualisation-and-quality-control

Best,
Peter
--
Dr. Peter Löwe
<peter.loewe@gmx.de>

Empfehlen Sie GMX DSL Ihren Freunden und Bekannten und wir
belohnen Sie mit bis zu 50,- Euro! https://freundschaftswerbung.gmx.de

2

On Fri, 2011-08-05 at 09:32 +0200, peter.loewe@gmx.de wrote:

Hello Ian,

>To learn how to use grass I've been trying to solve a real world
>problem. The problem that I chose was the flash flooding which occurred
>in south eastern Queensland. I'd like to see if grass could have
>provided reasonably accurate predictions and see if it could be used as
>a warning system.
>Ideally what I'd like to do is use the archived radar rainfall
>information from http://www.bom.gov.au/products/IDR503.loop.shtml#skip
>which can be found on http://gw.barratt.com.au/radar/IDR503/2011/01/10/
>I am intending extrapolate the rainfall rate using the Z value
>conversions recommended for east coast lows as per
>http://www.catchment.crc.org.au/pdfs/technical200207.pdf
>and then convert the rainfall rate rasters to the resolution of the
>aster dem.
...
>Any thoughts would be appreciated.

some earlier weather-radar related work with GRASS can be found here:
http://gisws.media.osaka-cu.ac.jp/grass04/viewpaper.php?id=44

http://onlinelibrary.wiley.com/doi/10.1111/j.1467-9671.2004.00178.x/abstract

http://opus.bibliothek.uni-wuerzburg.de/frontdoor.php?source_opus=759&la=en [-> the document is in german, but the technical appendices might be of interest for you -> Z/R realtionship, etc.]

FWIW here's study concerning visual quality assessment for 3D radar-met data:
http://www.docstoc.com/docs/69767645/4D-data-visualisation-and-quality-control

Best,
Peter

Thanks for the info, I failed german when I was 13 but it might be
interesting to see what a little motivation can do. :slight_smile:

3

Ian,

You may want to look at some U.S. National Weather Service approaches with what we call Flash Flood Guidance (FFG):

http://www.weather.gov/oh/hrl/gis/workshop3.html
http://www.weather.gov/oh/rfcdev/docs/ffgitreport.pdf

http://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=1054&context=usdeptcommercepub&sei-redir=1#search=%22Application%20National%20Geographic%20Information%20System%20Database%22

Flash Flood prediction is the most problematic in hydrologic forecasting because we must make predictions for small uncalibrated catchments and at spatial rainfall scales where rainfall estimation can be fairly suspect, even with raingauge-corrected radar estimates.

With FFG we try to estimate the amount of rainfall needed within, say, a 1-hour period, to bring the streams within an individual pixel (maybe, 1x1 km2 area) to bankfull conditions. Basically, two things are needed:

(1) estimates of the flow (runoff) needed to get to bankfull at each pixel
(2) in an operational setting, since basin conditions change (wetter or dryer, for instance) calculating, through a modeling process, the amount of rainfall needed to achieve (1) — this is the FFG; so, if that amount of rainfall is either observed or is projected to occur, a flash flood warning is issued.

Doing (2) is not too difficult; the problem is how to find reasonable estimates of (1) and the references I cited describe some methods. There is also an approach that tries to identify areas with high POTENTIAL for flash flooding (http://www.meted.ucar.edu/hazwarnsys/ffewsrg/FF_EWS.App.D.pdf) due to physical characterisics: slope, soil type, forest density, percent impervious, landuse, etc. The problem is translating the potential for flash flooding to flash flood guidance, related to either observed or predicted rainfall.

Regards,
Tom

··· 
-- 
Thomas E Adams
National Weather Service
Ohio River Forecast Center
1901 South State Route 134
Wilmington, OH 45177

EMAIL:	[thomas.adams@noaa.gov](mailto:thomas.adams@noaa.gov)

VOICE:	937-383-0528
FAX:	937-383-0033