For Texas Surveyor's

Well, creating a link does not work like I always did it on the other board.

Someone else knows how to do it, I'm sure.

Randy

Randy's Link

That could come in handy. Thanks Randy and Carl.

> Stumbled upon this link for Google Earth.
>
> Always wished you had the patent lines in Google Earth?
>
> Well here they are.

Hmm. That's the Texas Railroad Commission's GIS data. I started out with one original survey that I'm dealing with at the moment. Distance between NW corner of survey per GIS and as actually located on the ground? 860 ft.

I'm going to check some other corners to see how the GIS did with them.

Checked another one, the NW corner of a league on Onion Creek in Hays County West of Buda. The survey was made in 1835, but the corner is proven by original evidence. The Railroad Commission's GIS coordinates for the corner miss the actual position of the corner by 163 ft..

Kent

I use the ones from tnris. Most of the time, it's under 100', but sometimes its way wrong. Still, its better than nothing for recon from the office et cetera.

Checked one more. This one is the East corner of the T.B. Westbrook Sur. on the Old San Antonio Road in Hays County. The Railroad Commission GIS position of the corner is 212 ft. distant from the actual position of the corner on the ground and on the completely wrong side of the highway!

Kent

> I use the ones from tnris. Most of the time, it's under 100' ...

Well the average miss of the first three I checked was 412 ft. That's in an area where there actually are fences and roads that are close to some survey lines.

I'm really not clear on what value knowing where a survey corner might be within 861 ft. really is. That's a very large search radius if you're wanting to actually find the corner. If the GIS data were any good, that would be a fantastic tool, but I'd have to say it's a long way to the top if you want to rock and roll.

Kent

Like I said, its better in my area.

Kent

Those first corners that I checked were in Hays County in Central Texas. I thought I'd jump out to West Texas to see how things looked. I checked two corners of a survey known as Survey 339 on King Spring in Terrell County.

The coordinates of the NE corner of Survey 339 were good (in error by about 14 ft.), which one would expect considering that the corner was originally tied by course and distance to a spring a short distance away that shows up well in aerial imagery. However, the coordinates of the SW corner of Survey 339, nominally a mile South and West of the NE corner, that the Railroad Commission GIS gave missed the actual location of the original corner on the ground by 385 ft..

Kent

DISCLAIMER:

This dataset is NOT intended to be used as an authoritative public record for any geographic location or as a legal document and has no legal force or effect. Users are responsible for checking the accuracy, completeness, currency and/or suitability of this data. The Railroad Commission makes no representation, guarantee or warranty as to the accuracy, completeness, currency, or suitability of this data, which is provided "AS IS".

SOP for GIS.

The current GIS method is to input the parcels however. Then they will by tuned up in an iterative process over time. You have to start somewhere.

The GIS folks know there is a problem. Some sort of survey along with an adjudicative land court will be required to ever get it all right (legally establish all corner positions). It's probably the greatest opportunity for surveyors ever. Surveyors need to get on board and work with the GIS folks to get it done. If surveyors don't someone else will be tasked to do it. The public wants all this stuff they just need to be convinced to pay for it and have it done right.

I suppose in the long term the downside is that once it's all done the professional opinion survey will be rarely needed. Technicians will be able to simply replace destroyed corners at that point.

Kent

> Some sort of survey along with an adjudicative land court will be required to ever get it all right (legally establish all corner positions).

You're probably thinking of Utah, Leon. This is Texas that Randy posted about.

Kent

So there are no pincushions in Texas? If there are different opinions about a corner and you want the final GIS coordinate to represent the true corner, who decides?

I've actually wondered how GIS is going to go in Texas. Since the state and counties won't allow the filing of surveys isn't it going to be a giant leap over a big canyon to get proper survey data into a GIS.

We are considering requiring ties to geodetic coordinates in Utah be required so that the import into GIS would at least be possible for survey plats. How are you going to do that in a state that won't even allow filing of survey plats?

I suppose the GIS folks could start going through the deed records looking for coordinates and I'm sure they would find some. But still, how do they get official and something the public can rely on. Who decides whether the coordinate represents an established legal corner? Do Texas surveyors have that adjudicative authority? Do land owners have any control over their boundaries.

GIS is probably the greatest opportunity for surveyors ever and some don't even want to talk about it!

Kent

Shifting over to Kendall County, which is in the Hill Country and probably closer to conditions in Blanco County where Randy is based, I began by checking the coordinates that the Railroad Commission GIS gives for the corners of a survey known as G.C.& S.F. R.R. Co. Survey 331. The GIS position of the SW corner of Survey 331 is 572 ft. away from where the patent corner is really located on the ground.

I checked another corner of Survey 331 and find that the GOS position of it is 743 ft. away from where the corner actually is on the ground.

It looks to me as if the GIS has a working quality that is remarkably poor.

Kent

So, to be specific, I compared the actual positions of seven original patent corners that I've found in various parts of Texas to the positions of the same corners carried in the Texas Railroad Commission's GIS and found the following misses:

860 ft.
163
212
14
385
572
743

As I mentioned, that 14 ft. miss was as small as it was because the corner was tied by the original surveyor to a spring very close by that is readily visible in the aerial imagery.

The standard deviation of the GIS-derived coordinates that my sample reflects?

+/-512 ft. = standard error

Kent

There you have it. The only way that real coordinates are going to be populated into a GIS is to go out and measure them. A survey hopefully done by a land surveyor.

You can either be part of the steamroller are part of the road!

The GIS people are reaching out for help. Get on board. Let's get it done and done right.

Kent

> There you have it. The only way that real coordinates are going to be populated into a GIS is to go out and measure them.

Of course. Who would think differently? Because of the metes and bounds nature of Texas land surveying, that task is exponentially more difficult than in the relatively simple PLSS. There's a very good reason why that Texas Railroad Commission GIS is unlikely to ever get tremendously better in the areas I've sampled.

For Texas Surveyor's-Kent

What are the distances on the lines between the GIS points versus the distance recorded in the deeds? That would be more useful to know instead of just how much absolute positional error is on the ground at a select point.

If the geometry is good on their parcel information and you have their GIS files then you can move the files to your found, verified and positionally correct points. If the geometry is good then this gives you a 1000% increase in accuracy and usability.

It's how we used the product from the BLM years ago and the DLG files. We knew they were not accurate but they were something to start with for many attempting to place their work in a system for easy retrieval.

All information is useful in one way or another. Some have skills to get the most out of it and others don't wish to make the effort.

For Texas Surveyor's-Kent

> What are the distances on the lines between the GIS points versus the distance recorded in the deeds? That would be more useful to know instead of just how much absolute positional error is on the ground at a select point.
>
> If the geometry is good on their parcel information and you have their GIS files then you can move the files to your found, verified and positionally correct points. If the geometry is good then this gives you a 1000% increase in accuracy and usability.

That's the whole point. In retracing 19th century metes and bounds land grants, you typically aren't dealing with nice, neat geometric things. You're usually dealing with land grants that adjoin previously surveyed grants, often run at different variations, and with chaining errors that can be significant.

The only way to get fairly good search coordinates in the office is to :

a) do considerable research of the later subdivisions of the original grants by abstracting chains of title to identify reported or reputed locations of lines and corners at the time when the first wire fences were built,

b) sequentially plot the metes and bounds grants in the order in which they were surveyed, paying particular attention to calls for topography as an aid in plotting,

c) estimate the bearings at which a particular grant was run using the stated variation and historical estimates of the actual variation,

d) compare landmark fences and old roads (date of establishment determined from the Commissioner's Court records) to both the plot of the patent field notes and to the early subdivisions of the grants.

All of the above is very time consuming, but this is how you actually get a useful picture of the land grants to then go out on the ground and ... find them.