Ground Snow Loads

Started by Medeek, January 19, 2014, 02:22:06 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

Medeek

I've recently spent some time on looking at the snow load requirements of the ASCE 7-10 and the various state modifications to the IBC and IRC.  Obviously this is a gargantuan task, to summarize all this data and present it in a usable fashion.  I have managed to chip away at a few states that have standardized snow load data or equations.  The latest state I've tabulated is New York State:

http://design.medeek.com/resources/snow/newyorkgroundsnowloads.html

My map of the ASCE 7-10 ground snow loads is still unfinished, not because I cannot finish it but because I am currently waiting on a response back from the ASCE licensing division about the reproduction of the data presented in Fig. 7-1 (ground snow load map).  Reproducing or displaying a scanned version of the map seems to be less of a concern that creating an accurate electronic version of the map that is much more useful to the general public and engineering community.

I am summarizing each States snow load requirements on this page:

http://design.medeek.com/resources/snow/statesnowloads.html

As you can see I've only just started.  Some states such as Colorado let the local jurisdictions (City and County) set their own snow load requirements so creating a map for the entire state is more difficult but not necessarily impossible.  Some states such as Oregon have developed much more sophisticated online systems, I applaud their efforts.
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer

UK4X4

Colorado is not flat, snow fall depends on elevation, close mountain tops, storm direction etc etc

One number for the state would not work.

I'm at 90# a few miles away it's 70 and in the Vally it's 40

Not even elevation only works, it all depends on how you are situated in regards to the mountains


Medeek

Quote from: UK4X4 on January 19, 2014, 03:44:59 PM
Colorado is not flat, snow fall depends on elevation, close mountain tops, storm direction etc etc

One number for the state would not work.

I'm at 90# a few miles away it's 70 and in the Vally it's 40

Not even elevation only works, it all depends on how you are situated in regards to the mountains

I'm definitely not advocating one number for the entire state or even a single equation like what is used in Utah.  However, a statewide study and then corresponding map seems like the best course of action for any state jurisdiction that has variable geography and where snow is a serious factor. 

Oregon has done a very good job:

http://snowload.seao.org/lookup.html
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer

Medeek

Updated the Snow Load Calculator so that it now gives the reactions for a given truss or rafter based on the O/C spacing.
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer

Medeek

Oregon Snow Load map is up. This one, like the Montana Map, connects to the snow load database hosted by the SEAO. It also compares the retrieved value against the 20 psf snow load minimum as well as checks the modeled elevation against the actual site elevation and flags the user based on these checks.

The advantage to using this tool is that you don't need to know the lat. and long. off hand, just click on the map and it does the rest.

http://design.medeek.com/resources/snow/oregongroundsnowloads.html
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer


MountainDon

Maybe it was my connection or whatever but "Acquiring data from SNOWLOAD.SEAO.ORG..." went on and on and on and.......... . Never did complete.  :(
Just because something has been done and has not failed, doesn't mean it is good design.

Medeek

Its appears to be working, try it again?
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer

MountainDon

Anybody else try?  It still doesn't like me.  :(   I have a fast dsl line, but it just sits there acquiring.......  I've tried around Portland, Bend and Coos Bay, not that it should make any difference.
Just because something has been done and has not failed, doesn't mean it is good design.

Medeek

Still plugging away at the ASCE Fig 7-1 Ground Snow Load Map here:

http://design.medeek.com/resources/snow/groundsnowloads.html

I've just secured permission from the ASCE to digitize and display this map with an associated licensing fee.

15 of 48 states done so far.  It takes about an hour per state depending on how complex the snow loads are for that state. 

The biggest problem I'm having is with the map projection (mercator?).  Anyhow, when I try to do an image overlay into Google Earth the image is distorted so that the county lines don't line up correctly.  This forces me to only trace out small sections at a time and then re-adjust the overlay repeatedly.    Honestly I think the ASCE should be paying me for going to the trouble to digitized their map.  In the end I'll have a single .KML file that contains all of the data points and can be utilized in other mapping apps etc...
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer


Medeek

Just completed the Ground Snow Load Map tonight.

http://design.medeek.com/resources/snow/groundsnowloads.html

Digitizing it so that it was accurate took far longer than I had planned on but once I was underway I wasn't about to stop.

Note how most of the western US requires case studies, hence many of these states have their own snow load maps and research to back them up.
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer

MountainDon

That is cool!  Thanks so much for posting this. I've made the topic sticky so it won't get buried.

We are very close to the CS borderline with our home.
Just because something has been done and has not failed, doesn't mean it is good design.

Medeek

Interestingly I had a company that manufacturers and installs solar panels contact me regarding the ASCE ground snow load map.  They were wondering if there was a way so that their website could send an http request to the map with a latitude and longitude and have it kick back the snow load.  Since the map is already in digital format you would not think this would be hard to do.  So I got thinking about how I could build a program that given a certain lat and long would automatically click on the map and then fire back the elevation, snow load, etc... 

The only problem is google maps implementation of the KML layer does not allow this transmittal of information.  So I basically banged my head against the wall for almost 2 weeks trying to do the impossible.  However, I was not ready to throw in the towel just yet.  Since the KML data is just plain text, (coordinates) I thought there must be some way to analyze this data and determine if a given location is within a certain polygon and if it is then be able to assign that polygon description (ASCE snow load details) to a variable which is then delivered to the client.  Turns out this is called the point in the polygon problem and it is well documented:

http://en.wikipedia.org/wiki/Point_in_polygon

So I was about to write an algorithm for ray casting which would have taken some time.  Then I thought maybe someone else has done this before and sure enough I found a couple of good perl modules that handled this nicely and I narrowly avoided reinventing the wheel.  Just a few lines of Perl later and I had managed to create a nice little API for the ASCE ground snow load data:

http://design.medeek.com/resources/medeekapi.html
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer

Medeek

Added seismic and wind data to the API.
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer

Medeek

#13
Just put up an interactive map for Basic ASCE and IRC wind speeds:

http://design.medeek.com/resources/wind/basicwindspeeds.html

and the IRC Seismic Design Categories:

http://design.medeek.com/resources/seismic/sdc.html
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer


Medeek

Wisconsin map is now up:

http://design.medeek.com/resources/snow/wisconsinroofsnowloads.html

Note that this is not the ground snow load but the minimum roof design snow load.  Wisconsin does things a little differently when it comes to residential design and has their own code (UDC) versus the more widely adopted IRC.
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer

Medeek

Minnesota Ground Snow Loads is now up.  This state replaced the IBC and IRC snow load maps with their own zone map (two zones).

http://design.medeek.com/resources/snow/minnesotagroundsnowloads.html
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer

Medeek

Quote from: MountainDon on January 23, 2014, 06:12:51 PM
Anybody else try?  It still doesn't like me.  :(   I have a fast dsl line, but it just sits there acquiring.......  I've tried around Portland, Bend and Coos Bay, not that it should make any difference.

I just now realized that certain of these maps were not working with Firefox, apparently IE and Chrome did not have this problem.  After some research I determined it was the way the javascript was configured in one of my functions.  Try the Oregon and Montana map now with Firefox, I think we have addressed the problem.  I sincerely apologize, I should have probably asked what browser you were using.  I initially was only testing with IE and Chrome so this one slipped by me.
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer

MountainDon

Quote from: Medeek on March 24, 2014, 08:29:45 PM
should have probably asked what browser you were using. 

I should have thought of that as well. I am aware there are various things / pages that FF has trouble processing. For them I switch to Chrome or IE. It never occurred to me to try another browser or to ask you what you used.   d* d*   But I still have a "thing" for FF.    ;D

And yes, I tried OR and it popped right up with the answer.

I love this!!!
Just because something has been done and has not failed, doesn't mean it is good design.

Medeek

Michigan Ground Snow Loads map is now up.

http://design.medeek.com/resources/snow/michigangroundsnowloads.html

This zone map is essentially an approximation of the ASCE snow load map with isolines fitted to county and township boundaries. 

I'm not entirely satisfied with the appearance of this map since the county lines make the map appear too busy.
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer

Medeek

Vermont Ground Snow Loads Map is now complete:

http://design.medeek.com/resources/snow/vermontgroundsnowloads.html

This zone map defines ground snow loads by township and city boundaries and is derived from the zone map on page 48 of the 2012 Vermont Fire & Building Safety Code.

Vermont is a small state so the KML for this map was relatively easy to create.
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer


Medeek

I've just started working on New Hampshire Snow Loads.  Luckily the New Hampshire NH GRANIT has the township boundaries available on their website in KML format, that has already saved me hours of work.

However, there are 295 townships, each with a unique ground snow load and target elevation as given by Ground Snow Loads for New Hampshire, US Army Corps of Engineers, ERDC/CRREL TR-02-6.  So this will take me a while to complete.

You can view the progress here:

http://design.medeek.com/resources/snow/newhampshiregroundsnowloads.html
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer

Medeek

Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer

Medeek

I began working on the idaho snow load map based on the the snow load study by Dr. Ronald L. Sack done in 1986 and published by the University of Idaho.  It has taken nearly three days to digitize the isolines to sufficient enough accuracy so that the normalized ground snow loads can be calculated.  Other than the national snow load map this is probably my most ambitious project to date.  I am now converting the KML polygons shown here:

http://design.medeek.com/resources/snow/idahogroundsnowloads.html

into KML paths (isolines).  The tricky part will be programming the algorithm that can interpolate between isolines.  The basis for this algorithm is my previous study shown below.



The interpolation is further complicated with peaks, valleys, ridges and saddle points.  In order to deal with these degenerate cases will require the addition of some virtual isolines to provide additional data points to the algorithm. 

When I am finished the user should be able to click on the site location on the map and get an accurate NGSL which then automatically multiplied by the site elevation will give the ground snow load for that location.
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer

Medeek

Idaho Ground Snow Loads Map is now in Beta testing:

http://design.medeek.com/resources/snow/idahogroundsnowloads.html

This map is quite complex compared too all of the previous ones and requires a much more complicated alogrithm to interpolate between NGSL isolines.  Additionally, there are some specific locations on the map that do not conform to the isolines and those are in the process of being added into the final calculations.

This one was definitely a challenge but that is what makes it fun.  The detailed report is also a work in progress but it should provide some insight into how the interpolation is achieved and what limitations, if any, exist.
Nathaniel P. Wilkerson, P.E.
Designer, Programmer and Engineer

MountainDon

Yikes!!   There are some mighty significant snow loads in ID!
Just because something has been done and has not failed, doesn't mean it is good design.