LS adj of record mapping

.MAPMODE is the coding function you are looking for. Page 94 in your StarNet v8.1.1 Reference Manual. I don't think I've ever adjusted a map to found positions, but I have done just the reverse - adjusted control to hold record positions.

LS should be unnecessary with recent maps, however with ancient maps it may be necessary.

Many ancient surveys never had any adjustments done at all, simply mapping the field observations. Without the original notes the record map is the only record available and could be adjusted. I did one once first correcting a blunder, distance transposition that was shown correctly on a record map of an adjacent parcel. Then you can adjust, especially if metes and bounds descriptions give you a good hint at which were the closing lines that may never have been surveyed in the field. After that one could also adjust for scale. I did the first two, the third was not required.

When adjusting a record document one must make assumptions on the field work, such as angles to nearest 1/4 minute or 1/4 degree and distance to nearest, link, foot or tenth. I gave the highest priority to the record surveyors ability to prolong a straight line, holding a bearing true through three different subparcels and a second line through two.

Paul in PA

LS analysis of an old plat may suggest some things to you, but I wouldn't hope for it to resolve everything.

I've spent a lot of time trying to fit the 1950's plat of the neighborhood our church is in, with very limited success. Least squares analysis detects conflicts, but doesn't show how to resolve them. You have to make a lot of guesses at what might have happened.

I found some places I mis-read the numbers the first try, but I'm pretty certain that there are instances where the lots along each side of a line don't add up to the same total distance. Then there are the curves that have redundant parameters that don't match each other. I strongly suspect an initial plan was taken to the field, slope-staked, and some of the numbers changed according to what was left over at the end of the staking.

Then there is the plat where a friend's son bought a house. The exterior has what appear to be five straight lines with interior angles at the vertices. But they don't add up to the expected value by 26 minutes. Trying least squares has not pointed to any single assumption that would make things fit decently. I'd hope for 1 minute and 0.1 ft, but that's not happening by a long shot.

The project is an urban block with many curves I wanted to check some of assumptions in a different fashion.

Many curves? Older deeds and maps often do not have the perfect curve geometry we see today. Most often curve tables were used, so without an ancient curve table you may not get what the ancient surveyor used. Many curves are not perfectly tangent when one assumes they should be. Curves with less than ideal elements disclosed often may have multiple solutions. In PA the priority is direction over distance, so it is reasonable to fit in a somewhat different curve between two called bearings. Holding distance over direction may never get you to an ideal fitting curve. Old urban maps quite often have little or no calls. Hundreds of sidelines have to be presumed to be perpendicular to front and/or rear lines.

I recently completed an urban survey. The filed maps include many pages and blocks. The only bearings are on the sheet 1 project outbounds. Additional maps (block or blocks) have some angles at some intersections, centerline to centerline distances and lot distances hear and there. I was fortunate to have one called for monument at one street intersection near my parcel. The only other monument found within my block was in a front garden and had been displaced per neighbors statements. I found 1 monument at an intersection to the West and 2 monuments at a tangent to 2 curves to the East. My block/map called for those adjacent block monuments so I was in like Flynn; 0.02', 0.04' & 0.02' off of perfect geometry. BTW, the monuments were granite, no iron, 3 of 4 were above grade and I dug at a few other locations to no avail.

Paul in PA

Paul, thanks very much for chiming in and providing insight. I do not start any job with any prior assumptions In other words I take each project on its own merits. I do not assume that because a map is old it will be riddled with imperfect geometry. I analyze the evidence and from that analysis determine what it is that I have to do to properly execute the survey.

Dane Mince, post: 323915, member: 296 wrote: Anyone ever do this? Use least squares(Star*net) to adjust a record map to found monuments? Just wonder about coding the dat file?

Frequently - using TRANSFORM by Primacode. This software allows you to compare the record data against found monument locations to determine transformation, rotation and scale factors between the two. The documentaion alone is worth the price, and will allow you to see the "whole picture" instead of bits and pieces here and there. I can't recommend it highly enough!