Activity #8: Campus Microclimates

Introduction

In Activity #6, we created the geodatabase, domains, and feature class that we would be using for Activity #8. In this activity, groups of two were assigned different parts of campus and were tasked with collecting various weather related data. After the data was collected, we were to combine the data from all the groups, analyse it, and make several maps showing the different microclimates all over campus. The original goal was for each group to collect 50 points, each containing the predetermined information we decided on during Activity #6. Due to time constraints, equipment problems, and cold weather, most groups were not able to gather the full 50 points. We utilized what would could.

Study Area

Figure 1 - Study area for the different groups. My group's
study area is in the red box and our classroom is in the
purple box.

The first thing we had to do was divide up the parts of campus that each group was responsible for. We did this by simply allowing groups to pick, so first-come-first-serve. My partner and I did not speak up very quickly, and so ended up with a study area relatively far away from the building we had class in. Figure 1 shows the different groups and where their study area was located. Zach Howard and I were group 4, and our study area is withing the red rectangle. The purple rectangle is Phillips Hall, where our class meets every Monday.

Figure 2 - Our group's study area - field and parking lot
west of Haas Fine Arts Center

Figure 2 shows a close-up of our study area. This area lies to the west of the Haas Fine Arts Center. Our area included a grass (mostly snow covered at the time) field and a parking lot that had Water Street to the north and the bike trail and Chippewa River to the south.







Below are some pictures from our study area:

Figure 3 - Open field west of Haas Fine Arts Center, looking north.

Figure 4 - West side of Haas Fine Arts Center, looking northeast
from approximately same position as in figure 3.

Figure 5 - Parking lot to the west of field, from approximately
same position as figures 3 and 4.

Figure 6 - Parking lot south of the field, camera facing
southeast.

Figure 7 - Parking lot west of field, camera facing northwest.

Methods

Exporting Geodatabase

Figure 8 - Feature class and background raster in ArcMap

Before we could start gathering data, we needed to get the geodatabase, complete with domains and feature classes, prepped and onto the GPS device. Before exporting to the device, it's important to get the map ready so that it can easily be used in the field. We already had domains and a feature class created from Activity #6, but we also wanted a background image. Using an aerial image of campus as a base map, we brought everything into an ArcMap file (Figure 8).

Figure 9 - Image extent used when exporting to GPS device

Since our group would only be using our study area from this image, I zoomed to the field by Haas Fine Arts and made sure that when I exported the data to the GPS device, it would only be our study area, as opposed to the full image (Figure 9). At this point, it would be a good idea to double check that the domains and feature classes are properly set up before exporting (especially in this case since it has been a few weeks since I created them). We forgot this step, and thus had to manually fix the attribute table after we finished collecting data (more on this later). ALWAYS DOUBLE CHECK!

Figure 10 - Extensions menu where we select
ArcPad Data Manager

In order to export the map, a couple of things must be done within ArcMap. Under the Customize menu on the main tool bar, click on Extensions. Extensions are "add-ons" to ArcMap that usually cost more to have access to. The ArcPad Data Manager is one such extension that we will need access to. In this menu, we can check the boxes next  to any extensions we may need (Figure 10). Luckily, UWEC has access to all of them. Once we have access to ArcPad Data Manager options, we can add the toolbar. Once again under the Customize menu, select Toolbars and then ArcPad Data Manager (Figure 11) and the toolbar shown in figure 12 will pop up.

Figure 11 - How to add the ArcPad Data Manager toolbar

Figure 12 - The ArcPad Data Manager toolbar

Clicking the first icon on the left (with the arrow point to the right) opens the menu for choosing what data we want to export. The list automatically brings up the data that is on the current map. Where it says Action, click and choose Checkout all Geodatabase layers. The window should look like it does in figure 13. The next window is where we specify output options. I set the folder to export the data to (a previously created folder within my class folder on the network) and named the folder that would store the data micro_haasta_howardza (micro + both of our usernames) (Figure 14). Lastly, in deployment options (Figure 15), I made sure the box next to Create the ArcPad data on this computer now was checked. 

Figure 13 - Get data for ArcPad window

Figure 14 - Output options for data

Figure 15 - Last step in ArcPad data creation 

Figure 16 - Two copies of ArcPad data, in case a backup is needed

It may take a few minutes to create the data. Once it is complete, the data will be in the folder specified. Before moving ahead, copy-and-past this folder so that there is backup data in the event of any problems (Figure 16).

Figure 17 - Timble Juno
3 series

For this exercise, we are using the Trimble Juno GPS device (Figure 17). The device connects using a simple USB chord. Once it is connected, all that we need to do is cut-and-past one of the ArcPad data folders (in my case, micro_haasta_howardza) into the GPS device's storage using Windows Explorer.





Gathering Field Data

Once the geodatabase was on the Trimble, we were ready to head outside and collect the data. We brought several devices and tools (Figure 18) to help with this process. Here's the list:

Figure 18 - Equipment Zach and I used to collect data

1.) Trimble Juno GPS
2.) Kestrel 3000 weather meter
3.) Folding measuring stick
4.) Bandanna
5.) Orienteering compass









Once outside, it did actually take us a while before we were able to start collecting data. Neither of us were very familiar with using the Trimble Juno, and it took some time to get it working properly. After getting help from both Professor Hupy and Martin Goettl (the geography department's GeoSpatial Technology Facilitator), we found out that TerraSync (Trimble's GPS software that is also on the device) was also open. Only one program can be acquiring satellites on the device at a time, which is why ArcPad wasn't working.

When we got to our study area, we starting collecting points. For each point that was collected, we entered the following data:

Time
Temperature (degrees F)
Wind direction (azimuth)
Wind direction (cardinal)
Wind speed (mph)
Dewpoint (degrees F)
Relative humidity
Snow depth (inches)

When creating the domains, I had set the Group domain as being coded values. I either forgot to enter the codes, or they didn't save. Either way, I wasn't able to enter our group number until after we exported the collected data back to the computer. This wasn't that difficult, since '4' was the entry for every point, but had it been any other field, it would have been another setback. This is why you should always double check your domains before sending the geodatabase to your GPS device. This will avoid problems in the field.

Temperature, wind speed, dewpoint, and relative humidity could all be found using the Kestrel. All that was necessary was to scroll through the different features using the arrow buttons. The measuring stick was used to measure snow depth. Many of our points didn't have snow, as about half of our study area was parking lot, and much of the winter snow had melted by this point. The bandanna was used to find the wind direction azimuth. We could hold up the bandanna by a corner and hold the compass under it to get a general idea of wind direction. For the time, I simply used the time shown on my cell phone.

One of us would take measurements using the trimble, measuring stick, Kestrel, and compass/bandanna, while the other person plotted GPS points and recorded the information called out by the other. Zach and I switched off roles about every 10 points. This was partially so we could each get practice doing both, but also because our hands would get very cold doing the same thing for too long. In the end, we were only able to collect 27 points since class ended at 6:00 P.M.

Checking-in Data from Trimble Device

When we got back to the class room, we had to export our newly collected data from the GPS device to the computer. This was pretty simple. The Trimble was plugged back into the computer using the USB chord. Using Windows Explorer again, I navigated to the folder I created before going out into the field (micro_haasta_howardza, now with our fresh data) and cut-and-pasted it back into my personal class folder.

In ArcMap, we simply did the opposite of what we did when sending data to the Trimble. On the ArcPad Data Manager toolbar (again, make sure the ArcPad Data Manager extension is turned on), click the Get data from ArcPad button (Figure 19). Make sure you have the correct folder selected and click Check In.

Figure 19 - Check in data from ArcPad

In order to create maps and analyse the data, the shapefiles created by each group needed to be combined into one shapefile. One of my classmates was nice enough to do this by merging all seven of the separate shapefiles into one, and providing the result to the rest of the class to use.

Results

Figure 20 shows temperature data for my group's study area only. Warmer temperatures are on the north end of the parking lot, and are cooler to the south. This could be because of more shade closer to the bike path. There was more shaded areas from trees in that area, where as there was more exposure to sunlight closer to Water St. The black asphault could have been radiating more heat, as compared to the field that was also in the open. The ground was snow covered in those areas.

Figure 20 - Map showing distribution of temperatures for Group 4 study area
(using IDW interpolation method).

Figure 21 shows the temperature data for the entire class. There is one very high temperature in the middle of campus, which is why I used Natural Breaks classification for this map instead of a equal interval like I did in figure 20. This way, the outlier did effect the data.

Figure 21 - Map showing distribution of temperatures for campus using
data from all groups. Created using Nearest Neighbor interpolation method
and Natural Breaks (Jenks) classification.

Figure 22 shows the wind data. Each arrow points in the direction that the wind is blowing to and is color-coded to represent wind speed. The day that we collected data was not very windy in general, though there were some gusts.

Figure 22 - Map showing wind speed and direction for class data.

Figure 23 is a TIN (Triangulated Irregular Network) that shows snow depth around campus.

Figure 23 - Map of TIN showing snow depth around campus.

Discussion

The snow had started to melt before we collected the data, so that is why areas of higher snow depth seem concentrated. The areas where snow depth is highest is in the middle of campus. I would guess that this is simply from snow being plowed off of the sidewalks and pilling up, as opposed to actual snow accumulation depth. The other area where snow depth is deepest is down by the river, which makes sense.

The wind map in figure 22 and the temperature map in figure 21 seem to show that areas with higher wind tended to have lower temperatures. On upper campus where there was generally stronger winds (probably because it is out of the valley) had lower temperatures than much of campus. The area behind and around the residence halls on lower campus had very low winds (significant cover from buildings), and had higher temperatures (in general). Our study area (by Haas Fine Arts) had higher winds AND higher temperatures. This could be because our location was exposed more to the sun (and wind) than other areas on campus, and had parking lot asphalt reflecting the energy.

The one very high temperature in the middle of campus (81.8) was puzzling. There could have been ventilation from the nearby building, or simply operator error. Also, several points from one of the groups displayed on the equator. This could have been because of proximity to buildings and subsequent lose of satellite connection. But those errors did have to be removed so that data analysis could be done properly.

Conclusion

It was interesting to see just how different some of these variables are from one place on campus to another. It would have been better (though colder) to do it on a day when there was more wind. Since most wind speeds were below 4 mph, the differences shown on the map are not very conclusive. We also had to work together for this project; not just in our groups, but as a class to consolidate the data in a usable manner. I thought this went well, with people sharing the information with the rest of the class when they could. I general, I liked this assignment. I just wish spring would come already so I didn't have to be distracted by frozen fingers during every field exercise.