Lab 5: Using ArcCollector to collect micro-climate data via crowd sourcing.

Intro 

In our last lab, students were introduced to the ins and outs of how to properly create a geodatabase in ArcMap using domains and sub-types. In this lab, students will expand upon the value of working with a well organized file structure, while also gaining experience working with the Esri field data collection package known as ArcCollector. ArcCollector, in conjunction with ArcGIS online allows for groups to have access to data published by a single member of that group, and can thus go into the field and collect data.  This ensures that all field collectors are dealing with the same attribute table, and the resulting dataset will be 100% consistent.  Since the whole group has the option to create and/or edit data (as specified by the member who created the original service) a meaningful product can be produced in a short amount of time while still providing a high amount of coverage.

The database that our class was intended to work with was designed to collect point data pertaining to the very localized climate data with the the area of interest - a micro climate. With this data, the class could then observe the variations of such things like temp and wind speed across the UWEC campus (AOI).  Building off of this lab will be a more detailed account of how ArcCollector can be applied in applications that relate to streamlining maintenance practices for companies/organizations wanting to maintain or judge the condition of what they have built.

Study Area

The study area where micro-climate data was collected was throughout the University of Wisconsin Eau Claire's lower campus.  Our class of roughly 14 students was divided up into 7 groups of 2-3 and assigned a zone.  That group would spend the time in the field collecting climate information in that specified zone.  This was a good idea because throughout the campus area, there are varying types of ground cover and topography that in all likeliness will create unique data that distinguishes itself from other areas.   Referring to figure 1, below, the diversity of ground coverage as well as the group zones can be seen within the study area.

figure 1: UWEC Campus 

throughout the zones distinguished above, one can see that there are varying ground covers, building densities, vegetation heights, and water bodies that will in their own way affect the micro-climate data collected for this lab.  Let us now look more closely at what students will be working with once they go into the field

Methods

Because ArcCollector works in conjunction with ArcGIS online (although you can cache data and work offline if working in areas with little service) the data collection processes can be conducted through a cellphone or tablet that had access to a 3G network.  To do this, all one  has to do is download the ArcCollector app to their device, log in, and select the point feature class our professor, Jo Hupy, had made for us. The attributes he made for us to record data in the field were as follows. 

1. Group number 
2. Temperature (Fahrenheit)
3. Dew point (temp)
4. Wind speed
5. Wind Direction
6. Wind Chill
7. Notes
8. Time

He must have set domain range limits for each of these attributes, because if you tried to enter a value like 150 for wind speed, it would not allow the input to be completed, which would lead me to believe he set a range for each field. To produce values for these attributes, a Kestrel climate reader was used to produce values for these fields.  This electronic measuring device can be seen below in figure 2.

Figure 2: Kestrel weather monitor. The arrows pointing left and right allow the user to scroll through the various measurements the device is able to record. 

Using this device and a cell phone, my group members and I hit the field and began taking measurements.  We began taking our measurements at roughly 3:30 PM and ended at roughly 4:45 PM.  The experience was nice considering how nice it was outside, the temperature was ranging from 50-70 degrees, with little to moderate winds.  A week prior to this date, our campus was still experiencing sub 0 degree temperatures, so this was a welcome change. My group, comprised of myself, Andrew Ferris, and Alexander Kerr, took roughly 20 points, with about 15 - 20 meters between each point taken.  At the end, we returned to the lab and exported the data we had collected in the field back into the desktop into a shared file where the other groups were also adding their data. Now it was up to us as individuals to compile the data produced by our own group and the rest of the class to create a map that shows a combination of the attributes collected. 

Results and Discussion 

In total, 164 points were taken by 7 groups.  With this data, I was interested to see the correlation between wind speed and temperature.  Figure 3 below shows a bi-variate map that shows the temperature and wind speeds of each point taken.

figure 3: Temperature and wind speed collected using Kestrel and input in to ArcCollector - UWEC Lower Campus

Taking note of the point data collected, there are several things that this map tells about the relationship between wind speed and wind temp. The highest temps were recorded in areas where ground below was either blacktop or concrete - the lowest temps were recorded in areas where trees made up a majority of the surrounding.  Wind speed showed similar results, areas that were more open yielded higher values for speed, while more enclosed areas (by either trees, hills, or buildings) yielded low values.  

In terms of a relationship, I envisioned that higher wind speeds would equate to lower temperatures, but this map would suggest just the opposite - The points with higher wind speeds also produced the highest temperature values.  This is perhaps exhibited best when you refer to the lower portion of the AOI, the points collected in the forested area.  The values there produce little to no wind and also produce the lowest temperature values.  In contrast to that,  the more central portion of the AOI produced points that had both high wind speeds and temperature. To add more variation, looking at points collected south of the river, we can see high temperatures with low wind speeds.  In conclusion, there appears to be more direct influence between temperature and wind speed at these precise locations.  A more thorough correlation study would need to be done to say weather there is or is not a relationship, but this map would suggest that they do not influence one another. 

Conclusion 

If an organization can afford a subscription to ESRI products, ArcCollector can be used to efficiently track information pertaining to any number of applications. Because it is closely involved with ArcGIS online and ArcPortal, users of the software can create features with varying levels of operability and sophistication.  Once produced, the interface is very simple to work with, and because the creator SMARTLY used domains and subtypes - the  data will be concise with low input error.  Going forward, I would like to explore the value of using this software to produce data that could help with the maintenance of the campus infrastructure. 

Parcel Mapping Presentations - Daveis Center, UWEC Campus.

Introduction 

I entered the convention at roughly 11:30 am and sat down in a chair at a table and brought out my computer and sat down to start absorbing in the thrilling enviroment. Parcel Mapping, I could smell the excitment.  But in all honesty, there seems to be roughly 60 people here and their is a buzz in the room as the group seems to be in an intermission from the presentations.  The banter, i cannot be sure, but i assume is all about maps of years past.... or not, I cant really tell. On the large conference room projector at the front of the room, the slide reads, 'How does the PLSS help improve the Accuracy'.  After not long, a lady announces over the microphone that a question session was about to begin. The chatter quickly stopped, for battle was about to begin.

Question session 

The lady then went around to the tables, and a spoke to a representative whom gave an answer for each table.  In total there were roughly 9 tables, all filled with dedicated men and women eagerly awaiting there chance to answer the question at hand.  At the back of the room was a refreshment table, stocked with tea, water, and soda.  I quietly grabbed myself a cup of tea, topped it up with some milk and sugar and made my why down to my chair.  Shortly after i sat down, the proctor of the debate (who seemed to know alot of the people there).  As I listened to the questions, I took note of the notable points people were making.  Here are some of things I heard discussed throughout the question session, which ended with a break for lunch at about 11:45.  

Question 1: Why do parcel maps need to be accurate, and what aspects of accuracy are most important?" 

• Accuracy depends on use - urban (high acc) vs rural (low acc is ok)
• No absolute accuracy - all dependent on use of map
• as long as no overlaps/errors - use is most important 
• completeness, positional accuracy, curentness = importance 
• Tax payer deserves 100 accuracy, nothing short.
• currentness should be real time - do not wait to updates (temporal accuracy)

The guy who said these last two points seemed pretty pissed off 

Question 2: How does the PLSS help improve the accuracy of parcel mapping

• Do not use PLSS - gives better visual representation
• yeah attribute info, but can put aerial imagery or lidar in it? no 
• good for attributes and index - general visual aid.
• good for small scale maps - not large 
• old subdavisions are the only none useful aspects of the PLSS
• current PLSS are mainly derived from very old parcels - system needs upgrade.
• helps taxing process immensely
• Outlines  hierarchy of how land is divided. Locks down framework to further breakdown land divisions

The heat really started turning up at this point, people were feeling it.  Alas, the tension was broken when some one made a joke about how fast the guy was typing the question answers from each table, with only two fingers.  It was actually pretty impressive. 

The overall vibe consensus i got from these questions was the the PLSS grid system is good for some stuff but is not reliable in other senses.  The use, defines its relative utility. It cannot be used in activates that require hyper accuracy, a good example I heard that it should never be used to clip aerial imagery or lidar data because it will not give an accurate clip.  However, in terms of urban planning and property disputes, it provides a huge advantage.  Another good point that i heard was how much large corporations depend on PLSS information to look for fraud in thing like insurance claims. The law has a lot to do with the systems importance, it is one of the few continuous grid system that crosses the entire country, and has long since been used for dividing land uses.

Another big thing I noticed amidst the tension was there was a clear divide between the Survey Workers and the GIS workers.  I was a bit confused by this, but the divide seems to be based on what i discussed in the paragraph above. The GIS pros seemed to be more against the PLSS and the surveyors seemed to me much more for it.  A classic battle of new school vs. old school.  In the end, the conclusion I came upon my self was that the PLSS system is good for defining property lines and subdividing larger land divisions into smaller ones.  But, for use in highly accurate geoproccessing operations, they are too archaic and inaccurate to be relied upon for much.