So, this afternoon was spent establishing new control points in the concrete surfaces of public sidewalks in the vicinity of a particular block slated for reconstruction. There are plenty of control points in place now from which boundaries can be located, but the redevelopment work may well destroy most of the marks now in the old sidewalks. So the task was to extend control to newly reconstructed sidewalks on the theory that it would be more than ten years before they got replaced with something fancier yet.
The new control points were 253 through 256.
The control points 2 and 6 were installed about 25 years ago in the course of a survey. Control points 678 and 680 were set 16 years ago. Control points 102, 103 and 91 were set about 11 years ago in the course of a survey in the next block.
So, the task was to integrate the new control points 253 - 256 into the existing network to be able to both calculate their coordinates in the same coordinate system and to be able to have reliable estimates of the uncertainties of their positions with respect to the various things tied by the earlier surveys.
The solution was just to add today's measurements to the existing network adjustment in Star*Net. These are the residuals from the adjustment of this afternoon's measurements:
[pre]
Adjusted Observations and Residuals
===================================
Adjusted Measured Geodetic Angle Observations (DMS)
From At To Angle Residual StdErr StdRes
678 680 254 97-31-02.71 0-00-01.71 5.60 0.3
678 680 253 269-13-21.77 0-00-03.02 3.00 1.0
678 680 253 269-13-21.77 -0-00-01.23 3.00 0.4
680 253 256 271-22-46.00 -0-00-01.25 3.02 0.4
680 253 256 271-22-46.00 0-00-01.25 3.02 0.4
680 253 102 271-21-28.53 -0-00-00.97 2.53 0.4
680 253 102 271-21-28.53 -0-00-02.14 2.53 0.8
253 102 103 179-24-20.08 0-00-02.58 2.64 1.0
253 102 255 270-08-51.20 -0-00-00.80 2.53 0.3
253 102 255 270-08-51.20 -0-00-02.63 2.53 1.0
102 255 678 270-30-38.31 -0-00-03.19 3.17 1.0
102 255 678 270-30-38.31 -0-00-01.94 3.17 0.6
102 255 2 227-25-38.17 0-00-02.92 4.14 0.7
102 255 91 89-27-10.16 0-00-00.41 2.63 0.2
102 255 91 89-27-10.16 -0-00-00.09 2.63 0.0
255 678 680 178-45-40.19 -0-00-08.06 3.73 2.2
255 678 254 158-38-12.66 -0-00-03.34 3.59 0.9
254 2 255 138-00-25.38 -0-00-09.12 4.20 2.2
Adjusted Measured Distance Observations (FeetUS)
From To Distance Residual StdErr StdRes
680 678 186.9525 -0.0005 0.0066 0.1
680 254 72.5956 0.0056 0.0064 0.9
680 678 186.9525 0.0022 0.0066 0.3
680 253 290.6710 -0.0025 0.0068 0.4
680 678 186.9525 0.0017 0.0066 0.3
680 253 290.6710 -0.0035 0.0068 0.5
253 680 290.6694 0.0036 0.0068 0.5
253 256 182.9889 -0.0009 0.0066 0.1
253 680 290.6694 0.0029 0.0068 0.4
253 256 182.9889 0.0009 0.0066 0.1
253 680 290.6694 0.0041 0.0068 0.6
253 102 358.2366 0.0079 0.0069 1.1
253 680 290.6694 0.0036 0.0068 0.5
253 102 358.2366 0.0019 0.0069 0.3
102 253 358.1764 0.0034 0.0069 0.5
102 103 288.2468 0.0093 0.0068 1.4
102 253 358.1764 0.0059 0.0069 0.8
102 255 290.7165 0.0045 0.0068 0.7
102 253 358.1764 0.0037 0.0069 0.5
102 255 290.7165 0.0023 0.0068 0.3
255 102 290.7674 0.0074 0.0068 1.1
255 678 163.4204 -0.0029 0.0066 0.4
255 102 290.7674 0.0009 0.0068 0.1
255 678 163.4204 0.0004 0.0066 0.1
255 102 290.7674 0.0021 0.0068 0.3
255 2 103.6822 0.0102 0.0064 1.6
255 102 290.7674 0.0029 0.0068 0.4
255 91 289.8738 -0.0042 0.0068 0.6
255 102 290.7674 0.0016 0.0068 0.2
255 91 289.8738 -0.0055 0.0068 0.8
678 255 163.4269 -0.0028 0.0066 0.4
678 680 186.9252 -0.0013 0.0066 0.2
678 255 163.4269 -0.0071 0.0066 1.1
678 254 209.1935 0.0027 0.0066 0.4
2 254 282.5615 -0.0065 0.0068 1.0
2 255 103.6945 0.0110 0.0064 1.7
[/pre]
I should have mentioned that adding this afternoon's measurements to the network did produce small changes in the coordinates of the older control points from their positions as previously computed. As may be seen, most of the horizontal changes were on not much more than the millimeter level, which is pretty much as good as it gets in land surveying practice.
[pre]
1999 & 2004
Pt.No. N (ft) E (ft.) Elev.
2 10071305.791 3112574.430 476.371 SPIKE.WASHER
6 10071003.471 3112840.795 481.948 SPIKE.WASHER
91 10071345.705 3112203.191 474.942 SPIKE.WASHER
102 10070984.417 3112393.722 465.467 SPIKE.WASHER
103 10071067.032 3112117.592 468.135 SPIKE.WASHER
678 10071211.433 3112636.092 476.166 SPIKE.WASHER
680 10071157.831 3112815.143 479.352 SPIKE.WASHER
2015
Pt.No. N (ft) E (ft.) Elev.
2 10071305.789 3112574.431 476.367 SPIKE.WASHER
6 10071003.467 3112840.795 481.954 SPIKE.WASHER
91 10071345.705 3112203.190 474.939 SPIKE.WASHER
102 10070984.422 3112393.731 465.458 SPIKE.WASHER
103 10071067.029 3112117.598 468.129 SPIKE.WASHER
678 10071211.439 3112636.098 476.155 SPIKE.WASHER
680 10071157.830 3112815.142 479.375 SPIKE.WASHER
[/pre]
One thing that a least squares user will discover about networks is that as you add more and more redundant measurements to the network, at some point you reach the point of diminishing returns. That is, the uncertainties in the coordinates change less and less as additional measurements are added. The trick is knowing at what point you've gotten to the point of best results for least effort.
The indicators are the estimates of the uncertainties of the coordinates, which can be expressed as either standard errors or 95%-confidence error ellipses.
Here's where the network points are at the moment. These are the standard errors of their coordinates with respect to one particular point in the network:
[pre]
Standard Errors of Coordinates
Pt. No. N (ft.) E (ft.) Elev (ft.)
2 0.002129 0.002151 0.005896
6 0.003067 0.002685 0.006838
91 0.003197 0.002732 0.006041
102 0.002616 0.002766 0.006030
103 0.004445 0.003147 0.006932
253 0.002996 0.003510 0.007726
254 0.003913 0.002694 0.007667
255 0.002257 0.002464 0.006024
256 0.003752 0.005656 0.011940
678 0.002171 0.002461 0.006066
679 0.002422 0.002800 0.006612
680 0.002763 0.002178 0.005945
[/pre]
What the uncertainties of, say, Pt. No. 254 mean is that there is about a 68% chance that the absolutely best possible estimate of the actual coordinates of Pt. 254 is within 0.004 ft. in N and 0.003 ft. in E (rounding to the nearest 0.001 ft.) of the values calculated from the adjustment of the set of survey measurements that presently exists.
In my experience, +/- 0.003 ft. in N and E is pretty much as good as things ever are likely to get in the context of a land survey and are also good enough for just about any usual purposes.
