January 02, 2014 -- Over the last year, we here at US Land Grid have learned a lot about producing more accurate and easier to use data. We've also learned a lot about our users and future customers. Surprisingly, what we've found is that most people don't know bad data until it's too late.
All too often we get calls from new customers who are in a time crunch. They have purchased bad data, and that bad or incomplete data has only surfaced as they come to finality in building their project. For the end user, this has, and can, mean that the basis of their project is incomplete, or even worse, just plain wrong.
Below is a recap from 2013 of what we have seen and heard from customers:
- data had a plethora of duplicate polygons (which has a huge effect on land systems when auto generating polygons and for well spotting from footage calls)
- polygons were not complete or closed (which has a huge effect on land systems when auto generating polygons and for well spotting from footage calls)
- data didn't match the topos (this is a very bad sign)
- data didn't match the imagery (in Texas, this is probably the best source you can use to test the accuracy of your grid)
- all the available layers were not included (e.g. lots and quarters)
- charged for extra formats and projections
- leased the data with yearly fees (and auto-renews that they didn't expect)
- long time to get the data, and when they did, it wasn't very good or complete
We think the message here, and the most important message when it comes to any data, is to know and check the sources. It really isn't about the cost of the data (although that is obviously important), it's about being able to use useful data. For example, the PLSS land grid uses the USGS 1:24 topographic maps as the source. Using a land grid in those states without comparing it to that source before building your project can mean disaster. In Texas, a good and easy check is the imagery. Does your Texas land grid seem to match the imagery? If not, what is the source your data vendor is using and how do they validate it? Simply saying the BLM or the RRC is the source is not an accurate statement when determining true source and accuracy.
The following link shows an example from Arkansas. The White River and Arkansas River can cause all sorts of problems for land grid buffs and the systems they use. The polygons in red show data from other vendors and/or the GCDB/BLM. The red lines that show through are mapped incorrectly and the data bust can easily be seen when we zoom in with the topos.
http://b7c6e.s47.it/images/1/AR_Grid.jpg
The image below is the zoomed in version with a transparent topo in the background. Although good for the vintage, the red lines are clearly incorrect. As part of our 2013 data updates, US Land Grid made major improvements to the Oklahoma, Louisiana and Arkansas land grid datasets.
http://b7c6e.s47.it/images/1/AR_Grid_Zoomed.jpg
The bottom line is this - data that looks right does not mean it is right. Visualization is one thing... and visualization is useful when building a map. Using the data, however, for more than visualization, is where those pictures turn into true spatial analysis and understanding of that data. What this means to your business is that accurate and clean data results in accurate lease polygons and well spots, seamless integration into land systems without the pain or annoyance of duplicate polygons and the confidence that you are using data that improves upon the original and true source. Other things that should be important include consistency in the attribute values across layers and datasets, while also keeping in mind the data formats and data projections that your company needs and uses.
The other catch-alls we have seen:
The LOTS Layer
All too often we've seen our customers bamboozled in the past by data providers claiming evolution and uniqueness in datasets. The lots and/or quarters layers in the PLSS states are a good example of this type of magic smoke. Land grid that only includes sections and townships is not doing you a good service and not providing you the full land grid picture. If you buy land grid from a data provider, you should expect all the layers that come as land grid and are represented in the original source. In other words, if it's on the topo sheet, it should be included in the land grid. Nothing is worse than finding out that you didn't get all the layers you expected, and then being told by the data provider it will cost another ransom just to complete the land grid picture within the AOI.
Cultural Layers
Cultural datasets (roads, rivers, streams, etc) have received a bad name over the past few years. The data has been devalued. In the past, our customers found that the culture data they purchased from other data vendors showed outdated street information and incomplete data layers. In all of these cases, we have found that the data vendor is using outdated Census data as their source. In some cases, their cutlure data has not been updated since 2007. This has a huge impact on the accuracy of your map in AOI's like North Dakota and Oklahoma - where the culture data changes have been huge. Without knowing the date and source of these cultural layers you run the risk of using incorrect and incomplete layers - with the street layers being a prime example. Making matters worse, these same data vendors are leasing that outdated data on a yearly basis - with little to no improvement in the quality of quantity of that data.
The last year has been a learning experience for us here at US Land Grid, as well... In quite a few cases, the USGS topos needed a lot of work. In one case, our source file was projected incorrectly. Offering the new and multiple file formats was a real challenge. We have worked hard over the past year to improve upon the original work done by the USGS and our first release datasets by cleaning up duplicates, re-digitizing badly digitized areas and including more lots and quarters in all of our land grid datasets. On the cultural data side, we've updated our culture to 2013 - including the latest changes to roads, rivers, etc. And lastly, we've also updated over 60+ counties of our tax parcels - with a big focus on Texas and Oklahoma.
In 2013, we also added our now famously popular free datasets. This included geology by the state, county boundaries and railroad districts in Texas.
All of this was only possible through the help and feedback of our friends and customers.
Throughout 2014 we will continue to update and improve our data while adding some new paid and free datasets. Our update schedule and offerings will continue to be driven by requests and feedback.
We have also made some improvements to our cloud operations (US Land Grid started its service in the cloud the day we opened our doors!) allocating more resources for our larger online datasets and heavier visit loads on the website. The same as before, you will still have instant access to your data with free backups waiting in our cloud for future use. WMS services are also available.
We will be continuing our core business model (and we believe a big reason for our success) of "buy to own". Paying for data year after year is not a fair practice. We feel a lot better about charging for improvements and updates based on your needs and AOI's rather than handcuffing you to potentially old data that is chained to a yearly renewal. If there is a big update, we'll let you know about it and it will be up to you whether to pay maintenance for that update. If you're happy with the data and the vintage of that data, then it's yours to keep - no strings attached - and no future charges!
Lastly, we will be reinvigorating our data and GIS tutorial series. Over the next few weeks we will be sending out our first instructional video of 2014 showing the best ways to determine good quality data using the original sources. If there is anything specific you would like us to cover please don't hesitate to contact us directly and we will make sure we have it covered. We will also be offering a free data audit service for both customers and non-customers - where we help you formulate and build accurate and easy to deploy datasets across your organization. At the very least it has helped some of our customers become more aware of data busts and data holes that, in most cases, can be easily fixed before they come a bigger problem.