I'm creating a geodatabase in ArcMap 10.1 to manage a GPS quality control assessment. We have a Trimble R10 GPS and we're trying to assess variations in the data over time and under different conditions. Every few weeks I collect GPS data at the same landmarks on campus. I've been compiling this data in a geodatabase for both future statistical analysis and mapmaking endeavors.
My question is, how do I best display / compile data from the same landmark on different days? Right now I have individual points for each data collection that are essentially stacked on top of each other. Should I create multipoints for each landmark? Any better ideas? I'm relatively new to GIS and I've never created multipoint features; any instruction on how to do so would be much appreciated.
I'm creating a geodatabase in ArcMap 10.1 to manage a GPS quality control assessment. We have a Trimble R10 GPS and we're trying to assess variations in the data over time and under different conditions. Every few weeks I collect GPS data at the same landmarks on campus. I've been compiling this data in a geodatabase for both future statistical analysis and mapmaking endeavors.
My question is, how do I best display / compile data from the same landmark on different days? Right now I have individual points for each data collection that are essentially stacked on top of each other. Should I create multipoints for each landmark? Any better ideas? I'm relatively new to GIS and I've never created multipoint features; any instruction on how to do so would be much appreciated.
Thanks! Lindsay
While you can create a multipoint feature from stacked or attribute related features, that may not be your best option for compiling time based data. That is because if you merge all of your overlapping points together into a multipoint you will only have one feature and therefore only the attributes for one feature. That single feature could be useful to store information about the location that does not change over time, such as a unique location ID and location names, but not useful for storing the actual time based data, such as dates and sample readings like temperature, rainfall, etc. The relationship between the control multi-point and the time samples would be a one-to-many relationship. So if you have actual date based samples you should retain the individual points and investigate the help on time based data. A link to the first topic for the time based data help is here.
The Dissolve tool will collect information from single points and based on a common attribute or attributes combine the geometry into multipoint features. That would be easiest way to generate your control positions feature class. If you have multiple samples of a location on a given date, the Dissolve tool can create a single feature for that location and date and perform summaries such as averages, sums, and min and max values from the multiple observations, which may be more relevant to your analysis than the individual samples on that date.
The main difference between creating or modifying multipoint features over single point features when editing is that when you start creating point geometry each new mouse click continues to create a point within the current multipart feature you are editing, not a new point feature, and the feature is only a sketch that can still disappear from memory until you explicitly finish the sketch. You can finish a sketch by pressing the F2 key or right clicking over the points and getting the edit context menu and selecting the finish sketch menu item. Only after you finish the sketch will you see the feature appear in the Attribute editing window or in the tableview for the feature class. If you stop editing in the middle of editing a multipoint feature sketch before explicitly finishing the sketch, the sketched feature or edits will disappear and not be stored in your feature class. Other help on multipart features is here.
A few other observations about data. Sample data is the most refined level you have and generally the only way to resolve and understand data oddities that may arise when you group or aggregate them, so always retain the original sample data unaltered. The original samples will usually contain the best information for deciding if methodologies changed over time or if any data bias was present in your data collectors.
For field information that must be validated and corrected to match an authoritative source, do not overwrite original field data values to correct them. Create new fields that initially copy the field data and modify that copy only when you believe the original data contains errors. Then you can compare original field data to your corrections and determine if some training or quality control procedures needs to be established and where best to implement them. You may also find that your correction was in error due to an incorrect assumption on your part about your data, and such false assumptions are critical to discover and correct early in your process in addition to having the ability to restore the original data.
It is likley best to keep your date stamped points as separate features. Accessing and displaying stacked points is awkward, though. You could use Collect Events to create an index layer, that, for example, could afford accsess to the dates available for a particular place.
