Linear Interpolation of Wind Speed Maps (ASCE 7-16)

[medeek]

I don't know about the rest of you but when I look at the new wind speed maps in the ASCE 7-16, in particular the eastern United States I find the contour lines a bit confusing. With that in mind I spent a couple minutes this morning trying to fill in some of blanks so to speak, see below:

Its obvious that there a number of saddle points in the isolines and that is what complicates things at first glance. I am curious how well these hand sketched isolines would correlate with the ASCE hazard tool.

The way the map is shown in the printed manual doesn't define the isolines in enough detail in my opinion, too much is left to interpretation and user error. The one spot value at the end of Lake Michigan is a case in point.

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[medeek]

I suppose one could argue that the discrepancy caused by poor interpolation will probably result in a wind speed difference of only 2-3 mph which will not drastically alter the wind speed analysis and loads calculated for a structure. However, I would argue that we are doing engineering here not just throwing around warm fuzzy numbers that are somewhere in the ballpark of where they need to be. If this is the case then we can probably dispense with most of our current, overly complicated wind load chapter and all its factors and other accoutrements. Why go to all the trouble to devise such a complicated wind load analysis when don't really even know the wind speed for a particular site?

This brings to mind the numerous discussions regarding significant figures that I had from my university professors back in the day. We are misrepresenting our data and falsifying its accuracy when we use such ambiguous maps but then turn around and try to provide such in depth analysis with our complicated algorithms.

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[medeek]

I've compared my hand sketched isolines with the ASCE Hazard Tool and so far I'm getting pretty good agreement.

I'm wondering if anyone could recommend a surface program that could extrapolate the surface given a number of spot values and isolines?

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[EngineerEIT]

You can use the below tool from ASCE and find the ASCE 7-16 wind speed using the address or longitude/latitude of your project (wind/tsunami values are free to use)

https://asce7hazardtool.online/

 

[medeek]

Yes, I know but that is not my point. Why can't the ASCE provide a higher quality map in their print edition, something with a few more contour lines like the sketch I've given above.

The funny thing is that they have very detailed maps now for the State of Hawaii, yet the map for the continental US is quite crude in comparison. I know we can do better than this.

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[Jerehmy]

The quality in a lot of the images in ASCE 7-16 is clearly lacking. Many images were clearly copy and pasted from 7-10, but by someone who doesn't know what they're doing because they are very pixelated and grainy compared to the 7-10 version. Not very good QC on the images I must say.

 

[JStephen]

In some places, the local authorities will simply specify a wind speed in their local amendments to the building code, so check there first.

I'm not afraid to calculate seismic or wind loads to 8 decimal points in a spreadsheet, but keep in mind that the uncertainty associated with reading the maps is probably less than the uncertainty in the actual wind speed data.

 

[medeek]

I agree with all of your comments but at the same time why add more uncertainty into the process with the interpolation of sub-standard maps.

A confused student is a good student.
Nathaniel P. Wilkerson, PE

http://www.medeek.com

 

[GregLocock]

medeek, fitting 3d surfaces to a cloud of points is non trivial. However many CAD programs will do this. Matlab will do it, although the learning curve is a tad steep.

And... here's an add-on for Excel that will do it.

http://www.xlxtrfun.com/XlXtrFun/ReadMeXlXtrFunAndSurfGen.htm

The latter is so good that it is part of my standard build for a PC.

Good luck with the latter it seems to be dead now.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?