Activity Feed › Discussion Forums › Software, CAD & Mapping › Reviewer Does Not Understand Rounding
Reviewer Does Not Understand Rounding
Posted by armichael on October 8, 2019 at 11:59 amThis is probably old hat to some of you guys but I??m dealing with a certain jurisdiction in Maryland for the first time, doing a resubdivision plat submission. This reviewer keeps complaining that the Total Area shown for Buildable Lots is ??incorrect? when you add up all of the incremental areas shown in each individual lot. Maybe it??s because we label these lots to the nearest square foot, and sometimes we round them up to the next square foot when it is past the 0.49 square foot mark? Hmm.
I don??t sign plats here (yet) so I don??t have too much concern when my boss says to just go with it, but is this ridiculous? I understand that if a lot is 11,600.60 sqft it??s not the end of the world to call it 11,600 square feet. But why not show the CLOSEST area for each lot instead of fudging things so we don??t have to face the harsh realities of rounding? Somebody please explain, I am not smart enough to comprehend.
daniel-ralph replied 4 years, 6 months ago 16 Members · 19 Replies- 19 Replies
You will still be hit with the harsh realities of rounding, if you do it your way (my way, too). 21.4 + 33.3 = 54.7 rounded to 55, but 21 + 33 = 54. Of course our CAD algorithms calculate out to umpteen decimal places. Maybe send the reviewer a note with a table of the lot areas out to three decimal places, summed up to the total area, then round the total. That should shut him/her up.
I hope you realize that if you went out and surveyed the corners of your tract three times you would get three different answers for area. It might be 11,600.44 and it might be 11,600.22 and it might be 11, 600.85. We like to believe we would get exactly the same answer every time. Unfortunately, we are not as precise as we pretend to be. Let’s say you believe you have a perfectly square tract. The square root of 11, 600.60 is 107.7060815. But, the square of 107.71 is 11601.44 Or, the square of 107.70 is 11599.29. Any of us would be mighty lucky if we could get the number to come out so close when we are reporting to the nearest one-hundredth of a foot to begin with.
Any one labeling areas on land surveys to less than a square foot should be called before the board to give a statistical explanation of their methods.
I feel for you. Everyone above has made the point. But some folks just don’t get it. An organization that shall remain nameless here that I have to turn work in to that has actual surveyors on staff has rejected my plats because my distances did not add up in total to what my individual lines were, out to 3 decimal places. Same with acreage. I am, much like you, meaning 0.001 or maybe 0.002 of a unit difference, but it was enough to get my plat kicked back. By a registered surveyor.
nearest sq ft for each lot, acreage for the total, that seems to work. The nearest sq ft is a silly number anyway.
Over the years I’ve just grinned and bore it when dealing with obstructive bureaucratic simpletons. Just make sure all your numbers add up to what the reviewer is expecting to see….like training a dog to sit; it requires some patience.
I once had a plat reviewer question whether the boundary of a survey actually closed…due to the fact he was checking distances with a 1″=100′ scale.
Oh how I wish I had that problem. Many times I have told customers that their survey can’t be recorded because I won’t submit a fraudulent survey.
I have learned on here that some reviewers nit pick something every time just to reinforce the concept that they are between you and your goal of recording your survey. I prefer to believe that most reviewers are not that way. I know I’m not and I provide review services for several counties. It is my belief that the ultimate liability is on the surveyor. In this case, if they want to round to the nearest square foot or ten square feet or 100 square feet or one-thousandth of an acre, that is their choice, not mine. The primary object is to have a description that is valid, plus or minus some tiny little bit due to the inability to have total perfection because one man’s perfection is another’s imperfection.
First and foremost is to adhere to whatever the State minimum standards happen to be. Better than minimum is recommended but not mandatory. Some municipalities, however, may add stricter standards on subdivision plats. If you need their blessing, you had better toe the line on their standards.
Yes I do realize it is a microscopic difference, and that another surveyor would without a doubt measure different bearings as well.But I still label to the nearest second. When we set the corners in the field the Rodman will surely not be tapping the pipe back and forth until it’s dead on line to the nearest second. And also that area is the least important element in the priority of conflicting title elements.
Its just the premise of saying ok, let me add up the sqft of every single lot to see what this guy got, then changing labels and tables when the numbers were already correct (mathematically).
I would not put anything less than a square foot on a precise lot job in the middle of metro America.
Autocad or any Cadd does not round off numbers the way I learned them in math class.
I have not taken the time to learn why it does what it does and that is why I set my COGO to show 4 decimal places, it stores to 27 places, so when the sum of all the lengths of all the lots on a boundary do not add up to overall length, I will go thru every inverse across every lot and see how it is supposed to be rounded and then put that value in.
Sometimes the sum will still differ because a couple of them will be almost but not quite 5 and I must choose which one to manipulate to make the sum add correctly.
In a precise metro area, I would probably show it to 3 decimal places just so everything would sync.
Texas does have guidelines to follow that TSPS came up within the 1970s that I have followed and when the nitpickers want something else, I send them a copy of those guidelines and they will whimper away and leave me alone.
0.02
Years ago a wise Engineer I worked under went on a pilgramage preaching Significant figures. How I wish I still had a copy of his presentation. It was on the eve of laser distance meters. Some of us were actually reporting distances out to three decimal places because that’s what they read out to.. I’m constantly educating Flood Plain Coordinators if the Base Flood Elevation was contrived using a forty foot contour map with a planimetered drainage area and a statistical rainfall figure. How can they be interpolating the elevation down to a hundredth or even a tenth of a foot.
The best RTK GPS system on the market today claims 1 centimeter accuracy. It’s pretty tough to set a pin within three hundredths of a foot and then we record our distances to a hundredth and our angles to a second when the equipment you’re using isn’t capable of measuring that precisely.
Reminds me of the days we would tap the rebar over a hundreth to set it exactly on the point we had chained to. Acreage is even less precise. But our CAD programs will spit it out to three or more decimal places so that’s what we record on our Plats.
I realize your boundary won’t close flat, but you didn’t survey it accurately so that it would. 😕
Rounding the final answer to reflect likely accuracy is a good practice. But carrying every digit available while working up that answer is also good.
I once watched a surveyor using EDM that read out to 0.005 ft precision (before everybody had a data collector) record those readings in his field book rounded to 0.02 because that was the required precision on the final plat. Angles similarly.
That gave him a lot more error to deal with in calculating closure, etc. Every rounding adds error, which may or may not be significant in a given situation.
.- Posted by: @armichael
my boss says to just go with it, but is this ridiculous?
It is ridiculous. But your boss is right. We must choose our battles.
Had the exact same thing today. Reviewer asked why 100 and 100 added up to 201. I answered, “rounding”. Got a blank look. Latest email said to fix the problem. I will round each parcel, then add them to get the total.
-All thoughts my own, except my typos and when I am wrong.Encouraging professionals to lie. What a great idea!
?ñ
Use it and be merry.
I’m not sure how you have to present your figures on that side of the pond, but I’ve had similar situations over here, where an area has been split up but each plot has to have a minimum specified area (say 4 plots of 1 acre out of a 4.1 acre plot – less the access roads).
I list the plots in a table, one column with decimals and one column with its rounded value (using headers “Exact” and “Rounded”).
Only the decimal column gets totalled, to show the whole area – the rounded values are not totalled. I’ve had comments passed, but nobody has ever come back to dispute the overall area. Perhaps even the thickest understand it, laid out like that.
Lot of grumbling but not a lot of answers for the OP. Just complying with the reviewer without merit is not professional. Make sure that you understand the concept of significant figures and rounding before you make a decision on how to move forward. Take time, even if it’s your own time, to visit with the reviewer and understand what they are basing their comment on and try to teach them a more rational method. It what professionals do. This will help your understanding of the process and probably help us all moving forward. Finally, don’t succumb to your boss when they ask you to take a short cut with anything unless it is truly warranted. Simply not wanting to spend the time (money) to do the right thing is no excuse in my opinion for substandard, semi professional work.
Reaching out to this community was a good first step, but don’t stop here or quit too quickly.
IMHO
Log in to reply.