[Bldg-sim] Wind driven rain data

Mon Jul 23 01:24:34 PDT 2018

Dear Joe,

Thanks again for taking the time to respond, it is much appreciated.

I would like to address your concerns, which I think evolve around the potential usage of the model I try to build. I am looking at the early design stages of buildings, when little is known and (long-term) local measurements are not available. In that case the dimensions/exact shape of the building are not known, but the risk of WDR might still be relevant. It is trying to find the balance between an easy to use model with limited input and higher accuracy with more input.

The reason to use (hourly) synoptic weather station data is that the quality of this data is checked and readily available. 5 or 10 minute data would indeed increase the accuracy of a model and the option of weather underground or the UK Met Office's WOW network (this was mentioned to me a couple of weeks ago by Matt Clark - Met Office) is interesting, but there is the uncertainty of the data quality and the caveat you mentioned.

And it is true that the drop size in not usually recorded, as far as I know few instruments exist that can do that accurately. I have seen some recent papers that try to address this, but apart from a few experiments I haven't seen widescale use. In the past there have been a few papers that describe the velocity and distribution of raindrops (Gunn and Kinzer, 1949; Best, 1950; Atlas et al., 1973+1977; Adirosi et al., 2016)

Kind regards,

Roel Tersteeg MSc

PhD-student
Building Energy Research Group
School of Architecture, Building and Civil Engineering
Loughborough University
T: +44 77 56 10 90 54
E: R.M.Tersteeg at lboro.ac.uk<mailto:R.M.Tersteeg at lboro.ac.uk>

Roel

From: Joe Huang <yjhuang at whiteboxtechnologies.com<mailto:yjhuang at whiteboxtechnologies.com>>
Sent: 19 July 2018 18:51
To: Roel Tersteeg <R.M.Tersteeg at lboro.ac.uk<mailto:R.M.Tersteeg at lboro.ac.uk>>; Julien Marrec <julien.marrec at gmail.com<mailto:julien.marrec at gmail.com>>
Cc: Luis Miguel Blanes Restoy <lmblanes at gmail.com<mailto:lmblanes at gmail.com>>; bldg-sim at lists.onebuilding.org<mailto:bldg-sim at lists.onebuilding.org>
Subject: Re: [Bldg-sim] Wind driven rain data

Roel,

I have the following concerns about the approach you seem to be taking to define terrain-parameters for wind-driven rain:

(1) you are still basing your investigation on a rather simplistic model to derive wind-driven rain from the reported rain fall and wind speed, i.e., do you know how reliable is that model?

(2) for these types of correlations, you might want to use wind speeds at a much finer time resolution, since what's reported at the standard weather stations are random snapshots at an hourly or half-hourly time step.  You could consider using data from Personal Weather Stations (PWS), that are typically at the 5 minute interval and many also have an additional channel for gusts.  Such weather data, i.e., 5-minute data in urban or suburban areas, are currently available for tens of thousands of PWS, especially in North America and the UK, on web sites such as Weather Underground (www.wunderground.com<http://www.wunderground.com>).  One caveat would be to know the placement of the PWS, which tend to be on rooftops (the problem is worse for official weather stations that are supposed to be at 10 m above the ground).   If you're interested in the WU weather data, let me know via e-mail.

I would not recommend using New York Central Park, or at least you should compare the results to that from some other place that's less atypical.   The Central Park  weather station  is located at the Belvedere (see attached map), which is an isolated building in the middle of a large urban park that is in turn surrounded by tall buildings and adjacent to some of the tallest buildings in the world.

Yes, I'm aware that the terminal velocity of raindrops varies with their size.  However, I ignored that factor because any  error due to using a constant value would be small compared to all the other assumptions of the model and besides, I don't think raindrop size is recorded in any weather data.

Joe
On 7/19/2018 7:30 AM, Roel Tersteeg wrote:
Dear Joe, Julien,

Thank you for the extensive replies and your suggestions.

The rainfall above an area is not very different between airfield and urban areas, but due to the buildings and trees that may shelter locations in build-up areas this can cause lower rainfall in those locations. As you mentioned the wind will be altered in a more significant way. One thing I would like to do is actually (re)define 'terrain parameters' for wind driven rain. A single number, such as in your example, hides the actual uncertainty behind that number. And for that I would like to find measured data, the location of which doesn't matter that much.

The usage of weather stations for the purpose I am looking for is usually not an option as the location of weather stations is such that buildings and trees around the site don't have an effect on the measurements (typically > 100 m around the site free of obstructions). For the model I am trying to build, the (uncertainty in the) effect of trees and buildings on rain and wind is exactly what I try to model. If you know of weather stations on such sites, I would be grateful if you could help me with that. I will also have a look at the weather stations in central park and see if they can help me.

As for the calculation presented, there are several methods (e.g. ISO15927, ASHRAE standard 55) and they mostly have the form of Rainfall*Wind_speed*cos(Wind_direction-Wall_normal), with possibly some factors. There is a bit more explanation in Climatological note 3 from Met Eireann, from which note 13 is an update, how they calculated the map.
As a site-note: the velocity of raindrops is dependent on the size of the droplet with larger drops having a higher speed (terminal velocity) that smaller ones.

Roel Tersteeg MSc

PhD-student
Building Energy Research Group
School of Architecture, Building and Civil Engineering
Loughborough University
T: +44 77 56 10 90 54
E: R.M.Tersteeg at lboro.ac.uk<mailto:R.M.Tersteeg at lboro.ac.uk>

From: Julien Marrec <julien.marrec at gmail.com><mailto:julien.marrec at gmail.com>
Sent: 19 July 2018 13:26
To: Joe Huang <yjhuang at whiteboxtechnologies.com><mailto:yjhuang at whiteboxtechnologies.com>
Cc: Roel Tersteeg <R.M.Tersteeg at lboro.ac.uk><mailto:R.M.Tersteeg at lboro.ac.uk>; Luis Miguel Blanes Restoy <lmblanes at gmail.com><mailto:lmblanes at gmail.com>; bldg-sim at lists.onebuilding.org<mailto:bldg-sim at lists.onebuilding.org>
Subject: Re: [Bldg-sim] Wind driven rain data

Hello,

FYI, I can think of at least one location where you'd find probably good quality weather data in a big urban center and not at an airport: Central Park, New York City, USA, though it's in a quite large park but there are tall buildings all around it, see map<https://www.google.com/maps/place/40%C2%B046%2744.4%22N+73%C2%B058%2708.4%22W/@40.7787947,-73.9765413,15.5z/data=%214m5%213m4%211s0x0:0x0%218m2%213d40.779%214d-73.969>.

Note that in NOAA there's a weird thing going on where you have 4 stations with different operating dates, one did 1943 to 2010, another 2010 to 2012 or so, and another started after that (I know I said there are four: one did 17 days in august 2010 but overlapped another running one). Here are the USAF-WBAN codes for the two most interesting:

725053-94728 #CENTRAL PARK (2012 to today)
725060-94728 #NEW YORK CENTRAL PARK (2010 to 2012)

This data is free and can be accessed at NOAA's ISD (see here<https://www.ncdc.noaa.gov/isd/data-access>)

I looked at France's stations in the NOAA database too, since I have a better clue of where things actually are, and it seems that there are a bunch that aren't at an airport, but there are all in seaside towns, so that is probably not what you're looking for...

Here are some stations you'll perhaps want to look at, I'm including the three airports around central park too if you want to compare the data. Don't mind the fact that the 'END" date is last year, I just haven't refreshed this file since September 2017

USAF

WBAN

STATION NAME

CTRY

STATE

ICAO

LAT

LON

ELEV(M)

BEGIN

END

KEEP

70020

99999

BOULOGNE

FR

50.733

1.6

74

19730101

20170902

YES

72000

99999

PENMARCH

FR

47.8

-4.367

19

19360102

20170902

YES

73150

99999

LA ROCHELLE

FR

46.15

-1.15

10

19290801

20170902

YES

76410

99999

SETE

FR

43.4

3.7

81

19360615

20170902

YES

76950

99999

CAP FERRAT

FR

43.683

7.333

144

19730101

20100412

YES

725030

14732

LA GUARDIA AIRPORT

US

NY

KLGA

40.779

-73.88

3.4

19730101

20170904

YES

725033

94728

NYC CENTRAL PARK

US

NY

KNYC

40.779

-73.969

47.5

19431201

20100731

YES

725033

99999

NEW YORK CITY  CENTRAL PARK

US

NY

40.767

-73.983

27

20100801

20100817

YES

725053

94728

CENTRAL PARK

US

NY

KNYC

40.779

-73.969

42.7

20120801

20170904

YES

725060

94728

NEW YORK CENTRAL PARK

US

NY

KNYC

40.779

-73.969

47.5

20100801

20120801

YES

744860

94789

JOHN F KENNEDY INTERNATIONAL AIRPORT

US

NY

KJFK

40.639

-73.762

3.4

19730101

20170904

YES

Best,
Julien

--
Julien Marrec, EBCP, BPI MFBA
Owner at EffiBEM<http://www.effibem.com>
T: +33 6 95 14 42 13

LinkedIn (en<https://www.linkedin.com/in/julienmarrec>) | (fr<https://fr.linkedin.com/in/julienmarrec/fr>) :

2018-07-19 3:55 GMT+02:00 Joe Huang via Bldg-sim <bldg-sim at lists.onebuilding.org<mailto:bldg-sim at lists.onebuilding.org>>:

Roel,

I'm not an expert in this field of liquid precipitation, but here are my thoughts in response to your comments:

1) I'm not aware of drastic differences in rainfall between the airport and more urban areas of a city.  However, there are noticeable differences in wind speeds for which, in the absence of measured data, are accounted for in building energy modeling by TERRAIN-PARAMETERs dependent on the general site description, e.g., 1.00 for flat terrain such as an airport, 0.85 for rural,  0.67 for urban or forest, and 0.47 for urban centers with tall building, and the height difference between the measured data and the site.  if you're having difficulty in finding either measured wind speed or driving rain data (good luck!), I would suggest you just use the airport data and adjust that with a Terrain Parameter.

2) If you insist on getting rain and wind speed data for a suburban location, I'm sure those can be found among the 8,000+ weather stations for which I have hourly weather data, but the locations would be anywhere in the world.  Do you care where is the location?

3) The Met Eirerann paper describes a driving rain index, but does not provide the equation.  In thinking about it again (note that I've only thought about this after reading your original query), I think a plausible approach would be:

   a. find the approximate downward velocity of falling rain (a quick web search for "How fast do raindrops fall" gave the answer of 10 m/sec)

   b. divide that by the (adjusted) wind speed to get the fraction that would fall on a vertical surface, i.e., driving rain.

   c. multiply the "driving rain fraction" by the reported rain fall to get the amount of driving rain.

   c. use the wind direction to apportion the driving rain by the wall orientation.

It's still not clear to me what is the primary motivation for your investigation.  Is it to determine the amount of driving rain at a particular location or weather file?  to validate the method to calculate driving rain?  Or to just get a handle on how much driving rain is there ?

Joe

Joe Huang

White Box Technologies, Inc.

346 Rheem Blvd., Suite 205A

Moraga CA 94556

yjhuang at whiteboxtechnologies.com<mailto:yjhuang at whiteboxtechnologies.com>

http://weather.whiteboxtechnologies.com for simulation-ready weather data

(o) (925)388-0265

(c) (510)928-2683

"building energy simulations at your fingertips"
On 7/18/2018 5:43 AM, Roel Tersteeg wrote:
Dear Joe and Luis,

Thank you for your replies to my question and apologies for my very late reply to you both. Unfortunately the data from official, synoptic weather stations is airfield data and thus not applicable to my situation as there is no difference in microclimate (no changes in wind or rainfall due to surrounding building). The same goes for the Met Eireann information (if of interest, the Climatological notes 13 and 3 about wind driven rain can also be found and downloaded here as well: https://www.met.ie/education/publications/climatological-notes)

I am looking specific for wind driven rain data, or indeed combined wind and precipitation data that can be converted to wind driven rain, for different (sub)urban settings. The suggestion for looking at 'personal weather stations' is very interesting, although this has the issue of uncertainty about the quality of the measurements.

Kind regards,

Roel

From: Luis Miguel Blanes Restoy <lmblanes at gmail.com><mailto:lmblanes at gmail.com>
Sent: 06 July 2018 14:34
To: Joe Huang <yjhuang at whiteboxtechnologies.com><mailto:yjhuang at whiteboxtechnologies.com>
Cc: bldg-sim at lists.onebuilding.org<mailto:bldg-sim at lists.onebuilding.org>
Subject: Re: [Bldg-sim] Wind driven rain data

Joe
Try this link
https://es.scribd.com/document/346851908/MET-Driving-Rain-Index-IRELAND#close_user_settings_menu
Regards
Luis

On Thu 5 Jul 2018, 04:28 Joe Huang, <yjhuang at whiteboxtechnologies.com<mailto:yjhuang at whiteboxtechnologies.com>> wrote:

Luis,

Thanks for the tip.  I went to the link, which by the way is for the Irish Meteorological Service,  and read that it said, "The amount of rain received by a wall correlates fairly well with the product of rainfall and the component of the wind speed normal to the wall. This led Lacy and Shellard to propose the product of the mean annual rainfall and the mean annual wind speed as a driving rain index which is proportional to the total rainfall driven in one year on to a vertical surface always facing the wind",  which is functionally quite close to what I was envisioning, i.e., WDR = LP * f(wspd, wdir),  where LP is Liquid Precipitation or Rainfall. However, when I clicked on the link to the actual paper cited,  the Meteireann web site responded that  "404: Page not found".  Therefore, I still don't know the function that converts LP or rainfall to WDR.

Joe

Joe Huang

White Box Technologies, Inc.

346 Rheem Blvd., Suite 205A

Moraga CA 94556

yjhuang at whiteboxtechnologies.com<mailto:yjhuang at whiteboxtechnologies.com>

http://weather.whiteboxtechnologies.com for simulation-ready weather data

(o) (925)388-0265

(c) (510)928-2683

"building energy simulations at your fingertips"
On 7/4/2018 2:30 PM, Luis Miguel Blanes Restoy wrote:
Hi Roel and Joe

Did you look at available data from Met Eireann? There is a paper regarding WDR index calculation method that may be of your interest. See below link.

https://www.met.ie/climate/what-we-measure/rainfall#<https://www.met.ie/climate/what-we-measure/rainfall>

Regards
Luis

On Wed 4 Jul 2018, 22:12 Joe Huang via Bldg-sim, <bldg-sim at lists.onebuilding.org<mailto:bldg-sim at lists.onebuilding.org>> wrote:

What is your definition for Wind-driven Rain?  Practically all weather station reports, including official stations available in NOAA's ISD as well as unofficial "Personal Weather Stations" hosted on WeatherUnderground, contain information on liquid precipitation (LP), wind speed, and wind direction.  Can the WDR be calculated from these standard parameters, maybe as the LP * f(wspd, wdir) ?

Joe
On 7/4/2018 3:52 AM, Roel Tersteeg via Bldg-sim wrote:
Dear all,

For the verification of a model to estimate wind driven rain (WDR) I would like to use WDR data from different microclimates (rural, (sub)urban). I have access to several data sets of free field or airfield WDR measurements from different locations and the results from a measurement campaign in Vancouver. To better evaluate the model I would like to use measured data from one or more locations with different microclimates, not being free field or airfield sites.

My question is if someone knows if measurements are available of wind driven rain in one or more microclimates?

Kind regards,

Roel Tersteeg MSc

PhD-student
Building Energy Research Group
School of Architecture, Building and Civil Engineering
Loughborough University
T: +44 77 56 10 90 54
E: R.M.Tersteeg at lboro.ac.uk<mailto:R.M.Tersteeg at lboro.ac.uk>

_______________________________________________
Bldg-sim mailing list
http://lists.onebuilding.org/listinfo.cgi/bldg-sim-onebuilding.org
To unsubscribe from this mailing list send  a blank message to BLDG-SIM-UNSUBSCRIBE at ONEBUILDING.ORG<mailto:BLDG-SIM-UNSUBSCRIBE at ONEBUILDING.ORG>

_______________________________________________
Bldg-sim mailing list
http://lists.onebuilding.org/listinfo.cgi/bldg-sim-onebuilding.org
To unsubscribe from this mailing list send  a blank message to BLDG-SIM-UNSUBSCRIBE at ONEBUILDING.ORG<mailto:BLDG-SIM-UNSUBSCRIBE at ONEBUILDING.ORG>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.onebuilding.org/pipermail/bldg-sim-onebuilding.org/attachments/20180723/8ffec332/attachment-0002.html>