Can I analyze data without creating a square or rectangle shape? There are gaps in the GPS data points.

[RobertReynolds2]

I have data points that are GPS cartesian coordinates with values that I analyze using IDW or kriging.  Is there a way to remove the area where there is no data?

The software fills in the area with no data with colors that are projected out from nearby data so it is not accurate and I would like to remove it.  

In the attached image, the bottom right has no data so it should only show the map below, with no color scheme.

In some cases I've created several squares and rectangles that touch each other but I don't like the way it looks.

Thanks in advance!

 

[clt_cabq]

So the point of the IDW or Kriging analysis you are doing is to do pretty much what you got - an interpolation of data from locations that have data to others that do not. What are you trying to achieve? This is probably where you need to start, identifying the goals of your analysis given the nature of your dataset, so you can use the correct analysis tools and methods to visualize those results.

[RobertReynolds2]

I'm trying to achieve data analysis where there is data, and no analysis where there is none.  Again, the bottom right corner is analyzing an fairly large area with no data, thus it is inaccurate.

I've seen analysis with rounded edges where there is no data from conductivity and resistivity surveys.  They must have been using different software. 

In the image below, I've had to separate areas so that the IDW or Kriging will not fill in areas that were not surveyed.  The surface water and dense forested areas shouldn't be analyzed if there is no data present. Analyzing all of the data without separating each area into smaller geographical areas would create a large rectangle that covers surface water area and wooded area.

This may be the best result I can achieve using the limitations of ArcGIS, as it always creates squares and rectangles with no rounded edges.

[clt_cabq]

Methods that interpolate a surface like IDW and such aren't appropriate for what you are trying to do I think. I suppose you could apply a mask in the form of a buffer around each point so your analysis only returns values that fall within each, but I'm not sure how useful that is going to be. I'm not sure what 'analysis' you are trying to achieve - what are your data points, and what are you trying to determine about each location? 

[RobertReynolds2]

The raster is made up of X,Y cartesian coordinates, and a conductivity value at that specific GPS location. The data is imported as a .csv file from Excel. 

The IDW analysis is done only on the conductivity value, but placed on the map accordingly.  The goal is to show conductive hot spots at the specific GPS locations that it exists at.  

Example raster below:

 This image below is what I would like to be able to do in ArcGIS.  I'm not sure what software they used. The areas where no data was gathered is not obscured with the color scheme that analyzes the data.  For example, the area to the bottom right where there is dense vegetation has no data, so it is not analyzed.