Field Exercise 7 - ArcPad Data Collection Part 2

INTRODUCTION

          In the previous blogpost (Field Exercise 6 - ArcPad Data Collection Part 1) I explained how to transfer a geodatabase from a desktop platform onto a portable GPS device. Unlike the previous exercise, where each group created their own specific geodatabase to be deployed, every group in this exercise will use an identical master geodatabase agreed upon by the class. In the previous blogpost I also explained how to collect and record climate data for the purpose of creating a microclimate map of our area of interest (AOI). In this field exercise seven groups of two will be measuring and recording climate within their individual AOI or zone. This climate data will then be stitched together to create one giant microclimate map which will encompass the entirety of The University of Wisconsin's Eau Claire campus. Finally seven different maps will be constructed to show off the individual climate data fields including wind speed, dew point, surface temperature, temperature at two meters, wind chill, wind direction, and humidity.

STUDY AREA

          The study area or area of interest for this field exercise is the entirety of UW-Eau Claire's campus as well as seven zones within the campus. The campus has been broken up into seven different zones so that each group of two can focus their measuring and recording efforts on a designated zone. In my partners and my case we were tasked with collecting forty climate points within zone seven. Below in figure one the seven different zones of the UW-Eau Claire campus can be viewed.

Figure 1: shows the seven different zones that the UW-Eau Claire campus was broken down into. The highlighted zone on the right hand side of the map (zone 7) was the zone my partner and I were tasked with measuring and recording climate data.

           Even though seven different groups are recording climate data in seven separate zones on seven different GPS units, this data will eventually be combined so that one comprehensive climate map can be created for the UW-Eau Claire campus. Breaking the campus down into individual zones and designating one zone to each group was a smart move because more data points could be measured and recorded within each zone; Otherwise, each group would have to cover too much ground and fewer data points would be recorded. This method of data collection also emphasizes collaboration between the individual groups. Only after communicating with each other and putting forth a group effort can the data recorded within each individual zone be added to the comprehensive campus climate map.

METHODS 

          The first task to be performed for this exercise was deploying the uniform geodatabase agreed upon by the class onto a portable GPS device. Again the device used was the Trimble Juno 3 handheld GPS unit and I explained the deployment process in the previous blogpost. The next step was measuring the 40 different microclimate points within our designated zone. Each microclimate point had to include the fields agreed upon for the master geodatabase which were as follows: wind speed, dew point, surface temperature, temperature at two meters, wind chill, wind direction, and humidity. These  seven different climate fields were measured using the Kestrel 3000 weather meter and recorded using the handheld Juno GPS unit. Figure two below shows the dispersal of the points my partner and I recorded within zone seven.

Figure 2: shows zone seven and the forty microclimate points. It can be seen that not all of the recorded points fall completely within the outlined area. The dispersal pattern of the points in general follows the walk ways of the UW-Eau Claire campus, however their are two distinctive points which appear on the roof of the building in the bottom of zone seven. 

          After my partner and I collected our forty points we had to transfer the recorded data from the Juno GPS unit into ArcMap. This process is also explained in the previous blogpost. Once our data was transferred my partner and I had to collaborate with our peers and create the comprehensive UW-Eau Claire campus microclimate map. In figure three below a comprehensive map of the UW-Eau Claire campus showing all the data points from the seven groups can be viewed. In table 1 below an attribute table of the merged data with the microclimate fields can be viewed.

Figure 3: shows the seven zones of the UW-Eau Claire campus as well as all of the data points recorded within. Each Individual group had to collaborate with each other in order to create this one comprehensive map.

Table 1: shows a snapshot of the data point merger of the seven individual groups. Points 142 - 169 are shown with their corresponding microclimate field attributes. From left to right the columns in the attribute data show wind speed, dew point, surface temperature, temperature at two meters, wind chill, wind direction, and humidity

          Now that all of the data points along with their corresponding microclimate attributes have been merged into one comprehensive map, the construction of seven different microclimate maps can take place. The seven subsequent maps were created using ArcMap, designed using Adobe Illustrator, and include a wind speed, dew point, surface temperature, temperature at two meters, wind chill, wind direction, and humidity map (figures 4, 5, 6, 7, 8, 9, 10 respectively).



Figure 4: shows a map of the wind speed on UW-Eau Claire's campus. I have used both a point feature class to show wind speed at the recorded data points and a continuous surface feature to show an interpolated average wind speed in between the recorded points. I also used the red boxes to delineate the seven zones. I used the Kriging method of interpolation to create this continuous surface.

Figure 5: shows a dew point map measured in Fahrenheit of the UW-Eau Claire campus. I have used a continuous surface feature to show the interpolated average dew point in between the recorded points. I also used the red boxes to delineate the seven zones. I used the Kriging method of interpolation to create the continuous surface.

Figure 6: shows a map of the surface temperature on UW-Eau Claire's campus. I have used both a point feature class to show the surface temperature at the recorded data points and a continuous surface feature to show an interpolated average surface temperature in between the recorded points. I used the Kriging method of interpolation to create this continuous surface.

Figure 7: shows a map of the surface temperature at two meters on UW-Eau Claire's campus. I have used both a point feature class to show the temperature at the recorded data points and a continuous surface feature to show an interpolated average temperature in between the recorded points. I used the Kriging method of interpolation to create this continuous surface.

Figure 8: shows a wind chill map measured in Fahrenheit of the UW-Eau Claire campus. I have used a continuous surface feature to show the interpolated average wind chill in between the recorded points. I used the Kriging method of interpolation to create this continuous surface.

Figure 9: shows a map of both wind speed and direction on UW-Eau Claire's campus. I have used both a point feature class to show wind direction in azimuth at the recorded data points and a continuous surface feature to show an interpolated average wind speed in between the recorded points. My partner and I were the only group to actually record wind direction, therefor, zone seven is the only zone shown on the map. I used the Kriging method of interpolation to create the continuous surface.

Figure 10: shows a percent humidity map of the UW-Eau Claire campus. I have used a continuous surface feature to show the interpolated average percent humidity in between the recorded points. I also added the red boxes to delineate the seven campus zones. I used the Kriging method of interpolation to create the continuous surface.



DISCUSSION

          The maps created above may look similar to one another, but there are subtle changes in each one to enhance the climate feature being displayed. Different color schemes, different symbology the addition or omission of the zone boxes, and different classification methods are but a few of the various ways in which the maps were tweaked. Creating the perfect map is always a tricky process and normally takes twice as long as I originally intended. This exercise was no exception. Finding the perfect transparency for the interpolated continuous surface was very tricky as well as simply fitting the map image into a aesthetic shape. I had a lot of difficulty cropping the first two maps until I realized that there was a ghost background keeping them from fitting onto a cropped background.
          As far as what the maps actually show regarding UW-Eau Claire's microclimate foot print, it's about as normal as it can possibly be. Two distinctive patterns that I did notice were that it is much windier on top of the hill than down in the river valley, and It is coldest right at the base of the hill where the cold air sinks and chills out. No pun intended.
          My partner and I encountered no problems with any part of our data set and I attribute some of this luck to the geodatabase domains established in the previous lab. Planning ahead of time by creating safety nets in the form of domains really helped our group once we were out measuring and recording microclimate data in the field.
          As discussed above, every individual group was held accountable for measuring climate data in their campus zone and merging it into the master geodatabase. Every group but one did not record the wind direction, ergo the wind direction map shows only zone seven. There was also a group that didn't measure the temperature at two meters. Although these oversights effected the group as a whole, they did not ruin the master data set.

CONCLUSION

          Overall this field exercise allowed me to go back over the process of transferring data to and from a handheld portable GPS unit, which was important because it can be a tricky process. This exercise also allowed me to brush up on my ArcMap and Adobe Illustrator map making skills. I spent roughly three and a half to four hours creating these microclimate maps and I could have spent double that If I really wanted to make them standout. This goes to show that aesthetics and creativity take a lot of effort and time especially with map making. After finishing this lab I feel extremely comfortable not only using a portable GPS unit but transferring data to and from it as well.

Field Exercise 6 - ArcPad Data Collection Part 1

INTRODUCTION

          In the previous blogpost (Field Exercise 5 - Microclimate Geodatabase: Working with Domains) I created a geodatabase as well as a feature class that contained relevant microclimate fields. For this exercise I will be deploying the microclimate geodatabase onto a Trimble Juno 3 Handheld global positioning system in order to collect microclimate data in the field. Although I will be deploying the geodatabase and collecting microclimate data  using the handheld Juno unit, This exercise is only a trial run to work out all the kinks in the deployment and collection processes.
          The program utilized on the handheld Juno unit was ArcPad because it is one of the programs that is compatible with the many Arc platforms. Although the microclimate data was recorded using the handheld Juno unit, the actual measurements were taken using a small gizmo called a Kestrel weather meter. For this reason seven groups of two were created so that while one partner was taking the actual measurements, the other was recording the data. The methods for deploying the geodatabase onto the Juno unit as well as microclimate data collection methods using the Kestrel unit will be discussed in detail below.

Study Area

          As always every field exercise has a study area or area of interest (AOI). The AOI for this particular exercise is the small area on which my partner and I collected our five microclimate test points which was just off the corner of UW-Eau Claire's Schneider Hall. The AOI as well as the five test points can be viewed in figure one below.

Figure 1: shows the broader study area, which is the UW-Eau Claire campus, the AOI, which is outlined in red, and the five test points, which are depicted by green dots. 

METHODS

Deploying the Microclimate Geodatabase

          The first step in the deployment process was to add the geodatabase to ArcMap and add the point feature class that contained the microclimate fields. Next a base map of the area of study should be added so it can be viewed while collecting data. This is not necessary but makes it easier to see where data is being recorded on the fly. The point feature class and base map added to ArcMap can be seen in figure two below.

Figure 2: shows both the microclimate point feature class (Data_Deploy) and the base map (ortho) 

- The next step was to add the ArcPad data manager tool to ArcMap which is shown in figure three   below.

 Figure 3: Adding the ArcPad Data Manager Toolbar  is done by clicking the customize bar on the top of the screen, extensions, and checking the ArcPad data manager toolbox.

- Once the ArcPad Data Manager Toolbar is displayed click on the Get Data for ArcPad button to display the window in figure four below.

Figure 4:  First click the action button at the top of the window. Next select checkout all geodatabase layers in order to export the microclimate point feature class onto the handheld Trimble Juno unit. Lastly set the raster data set to Background TIFF.

 - The next step is to save the file to an appropriate folder. For our classes purposes we made a copy of this folder in case the original one became corrupted during data collection. The last step is to copy the created folder over to the Trimble Juno's SD card. The handheld Juno unit pictured in figure five below is now ready to record and collect microclimate data in the field.

Figure 5: shows a Trimble Juno 3 handheld global positioning system (GPS) unit

 - Once the data has been collected out in the field connect the Juno into a USB port and navigate back to it's storage card. Once here copy the folder with the desired data and place it into the correct folder. Now using the ArcPad Data Manager Toolbar in ArcMap click on the Get Data from ArcPad tool to import the data from the Juno unit. 

The Kestrel Weather Meter

          As stated above the Kestrel weather meter was the piece of equipment used to measure the microclimate data. The Kestrel 3000, which is the model that I used, has the ability to measure a plethora of climate conditions including wind speed, air water and snow Temperature, and relative humidity. For a full read-up on the Kestrel 3000 click here. For my purposes the Kestrel had to be able to measure every field I added into my microclimate point feature class which were as follows: wind speed, wind direction, humidity, dew point, surface temperature, temperature at two meters, and wind chill. In order to take accurate measurements with the Kestrel make sure it has been acclimated to the elements (let it sit out in the weather it will be measuring), and don't allow the readings to be affected by external errors such as body heat. Reading the Kestrels measurements is quite easy, simply understand the weather symbols that tell the user what they are currently measuring. A Kestrel 3000 is pictured in figure six below. 

Figure 6: shows a Kestrel 3000 weather meter.

DISCUSSION

          After deploying the microclimate geodatabase onto the Juno and becoming familiar with the Kestrel, my partner and I actually collected and recorded some climate data from our area of interest. We only collected five points, but five was all that was necessary to see if there were any problems with our geodatabase and collection methods. Figure one above shows the AOI as well as the five test points we recorded. At this point my partner and I could have created some very small scale microclimate maps with the five data points we collected. Below in figure seven is one such map which shows the temperature range of the five data points.



Figure 7: shows a map of the five surface temperature points as well as an interpolated range of temperatures in between the recorded points. This was done using the IDW interpolation method on ArcMap. The IDW interpolation method is described in my field exercise two blog. Considering this map only contains five data points it is not very useful or accurate. 

CONCLUSION

          My partner and I had no problems deploying our geodatabase and microclimate point feature class onto the Trimble Juno 3 handheld GPS unit, however, we did encounter issues when transferring data from the Juno to the desktop. In my opinion there was really no rhyme or reason for the difficulty in transferring the data other than technology can act up sometimes. After rebooting the Juno and re-plugging it in to the desktop it successfully transferred the data.

Field Exercise 5 - Microclimate Geodatabase: Working with Domains

INTRODUCTION

             The objective of field exercise five is to create a microclimate geodatabase with appropriate fields, attribute types, and domains for the purpose of deploying it onto a portable GPS device. The portable GPS device equipped with the geodatabase would then be used in the field to collect various climate measurements. Creating a deployable geodatabase with the proper specifications prior to data collection is done to speed up the collection process as well as to increase data integrity.

Geodatabases

            Understanding how a geodatabase operates and performs is essential when working with large sets of geospatial data. In a nutshell, a geodatabase is a repository for files that contain the same spatial data, and it has the unique ability to perform interoperable tasks. Interoperability is the ability of feature classes to work/communicate with one another and is one of the main strengths of a geodatabase. The old system of geospatial technology relied on shapefiles that contained separate spatial data and could not perform interoperable tasks. With a geodatabase, however, all files stored within create seamless feature classes that can be fit together like puzzle pieces.

Domains

            Geodatabases also contain fail safes in the form of attribute domains. An attributes domain can be set by the geodatabase creator, and is simply the legal set of values for that attribute. Using geodatabase domains is encouraged because it increases data integrity. There are two basic domain types: range domains and coded domains. A range domain specifies a valid range of values for a numeric attribute. When creating a range domain, you enter a minimum and maximum valid value. A range domain can be applied to short-integer, long-integer, float, double, and date attribute types. For a coded value domain the attribute being measured receives a coded value that corresponds to the actual attribute being measured (this can be for any attribute type). For instance the actual attribute might be a classification scheme for pipes; so water and sewage might be two two possibilities. A coded value domain assigns numbers to these classifications so water might equal a coded value of one and sewage might equal a coded value of two.

METHODOLOGY

Microclimate Geodatabase

This section covers the creation of the microclimate geodatabase.
Step 1: The first thing I had to was create the geodatabase and name it. The name I used was generic: Micro_Climate_myusername.gdb. Figure one below shows how to execute step one.


Figure 1: right click on the desired folder, click new, click file geodatabase. It will then ask the user to name the geodatabase

Step 2: The next thing I had to do before completing the microclimate geodatabase was set the parameters for the field domains. As stated above domains create a legal set of values that will be accepted for an attribute and can either be a range or coded value domain. The geodatabase will not allow a number to be entered that is outside of the legal range. The legal set of values I created were different for each domain created. I will go into further detail below in the fields, data types, and domains section. By right clicking on the geodatabase and scrolling down to properties, the domain properties window can be accessed.  Figure two below shows a list of my microclimate geodatabase domains.


Figure 2: shows where domain names and parameters were created. A detailed list of the domains will be provided in The fields, data types, and domains section below.  

Step 3: the next step was to create a point feature class that contained all the microclimate fields I felt were necessary to creating an accurate climate map of the UW Eau Claire campus. The fields I deemed necessary were as follows: Wind_Speed, Wind_Direction, Humidity, Dew_Point, Temp_Surface, Temp_2Meters, Wind_Chill, Ground_Cover, and Notes. All of these different fields were stored under different data types depending on what I would be measuring them with. In general, if the measurement being taken is a simple number a short integer data type will suffice. If the measurement being taken is a very long number that includes decimals, a long integer or float data type may be necessary. In order to record visual observations a text data type may be necessary. In figure one below a list of my field types and corresponding data types can be seen.

Figure 3: shows the list of microclimate fields and the corresponding data types they are stored in. 

 

DISCUSSION

The fields, data types, and domains

Wind_Speed - I chose wind speed as a micro climate field because it will show where there are wind tunnels as well as calm areas on campus. I will be measuring wind speed using the short integer data type. The domain I set up was a range domain to be measured in miles per hour with the minimum value set at zero and the maximum value set at 50. I chose these numbers because wind speed can never be less than zero mph and it is very unlikely that it would be higher than 50 mph.

Wind_Direction - I chose to measure wind direction because it will show how the campus buildings affect the air's behavior around campus. I chose to measure wind direction using the short integer data type. I used a range domain measured in azimuth with a minimum value of zero and a maximum value of 360. I chose these numbers because these are the degree parameters on a compass.

Humidity - I'm measuring humidity because I want to see how both elevation and proximity to the Chippewa River affect it. I will be measuring humidity using the short integer data type. I used a range domain measured in percent with a minimum value of zero and a maximum value of 100. I chose these numbers because there can't be negative humidity and there can't be humidity over 100 percent.

Dew_Point - I'm measuring dew point to see how saturated the air is with moisture. I expect it to be quite low because it is cold and dry outside at this time of year (early March). I will be measuring the dew point using the short integer data type. I used a range domain to be measured in degrees Fahrenheit with a minimum value of -30 and a maximum value of 60. I used these temperatures because it is unlikely that the temperature would be below 30 degrees Fahrenheit or above 60 degrees Fahrenheit in early March.

Temp_Surface - I'm measuring the surface temperature in order to see how it varies throughout campus. I'll be measuring the surface temperature using the short integer data type. I used the same range domain for the surface temperature as I did for the dew point. This is because the temperature minimum and maximum will be the same.

Temp_2Meters - I chose to measure the temperature of the air two meters above the ground just to see if it varies from the surface temperature. I chose to measure the temperature two meters above the ground using the short integer data type. I again used the same range domain for the temperature at two meters as I did for the dew_point and temp_surface fields because the temperature range stays the same.

Wind_Chill - I'm also measuring wind chill because it will show the effect the wind has on the temperature. I will be measuring the wind chill using the short integer data type. I used the same range domain for wind_chill, temp_2meters, temp_surface, and dew_point because the minimum and maximum temperature range for those measurements were all the same.

Ground_Cover - I will be measuring the ground cover to see if it has an effect on any of the other measurements. I will be measuring the ground cover using the text data type. I used as coded value domain for the ground cover field because I simply needed to input the type of ground I was standing on. The coded values were as follows: 1 = snow, 2 = concrete, 3 = blacktop, 4 = grass, 5 = gravel, 6 = sand, 7 = water, 8 = other.

Notes - the notes section is for me to record anything unusual or noteworthy I suppose. The notes will be recorded using the text data type. The notes don't necessarily need a domain because they will be different every time, but a coded value domain can be used to input often used observations.

 CONCLUSION          

            Attribute domains are supposed to increase the efficiency of data collection as well as data integrity, and the domains I created for my microclimate geodatabase did just that. When I was out in the field physically collecting the data I knew that the data I was entering could only be relevant data due to the domain fail safes I had set up prior to collection. I was also able to enter the data faster because all of the microclimate fields were present with readily available scroll down lists. Creating a deployable geodatabase prior to the data collection also allowed me to become familiar and comfortable with the different climate fields I would be working with. In other words I had a set plan I was able to follow which can be important as sometimes obstacles in the field can confuse or sidetrack a data collector. Creating a deployable geodatabase with set domains may have seemed like an arduous process, but in actuality it only took about twenty minutes to complete so it's well worth it to create a game plan before heading into the field.