Diurnal Heat Patterns Project Details
Description of Project:
Our project is the capture and analysis of diurnal heat patterns using UAS data and GIS software. The project will consist of first collecting thermal images of an area over several intervals throughout the day. Then we will combine the images into a composite image for ease of viewing and perform analysis on the collected data as a whole. This process will be repeated for multiple different areas of land so that we may see and compare the effect of diurnal heat patterns on different types of structures and land masses.Challenges:
There are numerous challenges with completing a project of even as relatively low scope as this one.First among them is scheduling dates and times that the entire group is able to meet up. Due to the nature of the data collection procedure and the need for the entire team to be present, it will be difficult to arrange a day where everyone is capable of being present for every flight since it will include flights throughout the entire day, preferably from sunrise to sunset.
Another challenge will be the finding of and clearance for places to collect data. We will need to do some cursory research into what areas will be diverse enough to warrant collecting data sets as well as contact any local airports or land owners that may not be too happy about finding a UAV collecting thermal images.
Training and classroom certification on the m600 and the thermal sensor will also be an important obstacle to overcome. That, as well as refreshing ourselves on the specifics of working with thermal images in either ArcPro or Pix4D will be vital to actually collecting and processing the data.
One final challenge is the research that needs to be done on the topic in order to be knowledgeable enough to make a presentation or answer questions. The topics of thermal imagery, diurnal heat patterns, and similar topics will be included in the research. We need to find out first, what we need to know.
Research Questions:
What is our end goal with our research?Why are diurnal heat patterns important?
What factors affect diurnal heat patterns?
What is the best software to use?
What local areas will we study?
How many flights will each mission be comprised of?
Does the utilization of thermal imagery require any additional flight planning?
What is the output of thermal imagery data collection?
Can animations be used to show heat changes?
Do we need Ground Control Points?
How will we display our data?
What is our end goal with our research?
The end goal of our research is to be able to create a presentation covering the topic of diurnal heat patterns (including a poster for GIS day) and be able to answer any questions that come up about it.Why are diurnal heat patterns important?
Diurnal heat patterns are the temperature variation that occur throughout the day. Diurnal means that it takes place over the course of the daylight hours of each day. They are patterns because, generally, they are the same from day to day (with several factors causing variations in the pattern). There are several reasons why the study of diurnal heat patterns are important. Some topics that are affected by these patterns are crop lifecycle (grapes are more acidic and sugary if they are exposed to a higher daily temperature variation), interior heating and cooling, expansion and retraction of materials due to heat changes, the effects that diurnal temperature changes can have on people or livestock, insect life cycles. It is also important data when looking at larger, long-term, trends of temperature in a given area and cannot be overlooked when comparing similar data.https://en.wikipedia.org/wiki/Diurnal_temperature_variation
 |
| Diurnal Temperature Variation in the US |
What factors affect diurnal heat patterns?
Several factors can affect the temperature variations.The length of the day is an obvious one, if the sun is heating the area for a longer period of time then the ambient temperature will naturally continue to increase over time. Increases temperature variation.
Along the same vein as the day length is cloud coverage. The clouds prevent the sun from heating the earth as much during the day while also preventing heat from escaping during the night. Decreases temperature variation.
Humidity and proximity to bodies of water are a major way that solar heat is absorbed. Water serves as a very effective heat sink and is able to absorb a lot of solar radiation before increasing in temperature. Along the same lines, once it has absorbed a lot of heat, it will stay hotter because of how much energy it has absorbed. If that makes sense. For this reason, places with little water (like desert regions) will experience the highest amount of diurnal temperature variations, being extremely hot during the day and extremely cold during the night.
https://www.thoughtco.com/diurnal-temperature-range-3444244
What is the best software to use?
We are planning on using Pix4D for processing our thermal imagery. Pix4D is already well equipped for working with thermal datasets. The Pix4D support site has tutorials for processing, visualizing, reducing processing time, using custom thermal sensors, and troubleshooting which should make the processing part of the project a relatively simple one where we can reference the website for the entire process.https://support.pix4d.com/hc/en-us/articles/360000173463-Processing-thermal-images
What local areas will we study?
Our general thought process for choosing places to collect data was that the area should be interesting enough to warrant a full day’s worth of data collection and accessible enough that we would all be able to access the location. An interesting location would incorporate what we’ve learned about diurnal temperature patterns, primarily that proximity to water is a factor, and include that in the data set. Someplace like the deserts of Arizona or Utah would be very interesting to see the patterns and to compare them to a more temperate location but they are outside of our reach. The locations that we settled on are the Purdue Wildlife Area, Davis Ferry Park on the Wabash River, and the Celery Bog Nature Area.1) Purdue Wildlife Area - areas of focus: the farm fields, the animal habitats, the gravel road/parking lot, the building structures, the neighbor’s small pond, the tree groves
 |
| Purdue Wildlife Area |
2) Wabash River/Davis Ferry Park - areas of focus: the river, the walking bridge, the small creeks branching off of the river, the farm field to the south, the tree groves
 |
| Davis Ferry |
 |
| Celery Bog |
How many flights will we need to do each day?
We will fly four times at each site: in the morning (8-9 a.m.), in the afternoon (12-1 p.m.), in the evening (4-5 p.m.), and at night right before sunset (8-9 p.m.). These times are contingent on flying during daylight hours, but because we are studying diurnal patterns in particular, we will need to focus on flight times around sunrise and sunset with enough flights in the middle of the day to give context to the changes that occured.Does the utilization of thermal imagery require any special flight planning?
First and foremost, flying in optimal weather conditions is paramount. Both wind (over 15 mph) and precipitation can change surface temperatures. We need to avoid flying in high humidity (over 60 percent), which will result in a haze in source imagery. To avoid thermal blur, we will keep a low flight speed (about 4 meters per second). Also of note, thermal images require no external light source, so they can fly at night.* There are several advantages to flying at night when differentiating objects with different thermal capacities (like water stored in insulation under a roof).Before we engage a flight mission, we need to allow the thermal sensor to warm up for five minutes. We have breaks between flights lasting longer than an hour, so we will need to perform the warm up again before returning to the air. Most cameras will perform a NUC (non-uniformity correction) several times throughout the mission which accounts for sensor drift.
*We are not permitted to fly the school’s equipment at night, so it’s not even being considered.
https://www.precisionhawk.com/sensors/advanced-sensors-and-data-collection/thermal
What is the output of thermal imagery data collection?
Collected thermal data is typically output in raster (image) formats, with a choice between radiometric JPEG files and TIFF files. There is a balance here: JPEG files are smaller, but information is lost. TIFF files are very large, but contain all the raw data from collection without compression.https://www.precisionhawk.com/sensors/advanced-sensors-and-data-collection/thermal
Can animations be used to show heat changes?
Pix4D contains tools that allow us to create animations within our data. A 3D flyby can be used to show the viewer around our locations and zoom in on specific details that we deem important. Our thermal data can be combined with a 3D orthomosaic to show the effects of elevation and land features on the heat patterns.Once we complete our initial processing of our data, we can then choose different waypoints within our ray cloud that determine where the camera will fly through. This animation can then rendered into a video that can easily be played on a screen.
https://support.pix4d.com/hc/en-us/articles/202560299-How-to-create-a-Video-Animation-Trajectory-in-the-rayCloud
Do we need GCPs?
GCPs make location data much more accurate. Without these, the location of our data compared to the actual location of the imagery would not line up. This would cause problems if we needed accurate measurements or if we wanted to layer different types of data without thermal data.Fortunately, the M600 Pro comes equipped with PPK (post processing kinematic). PPK systems make location data much more accurate by taking two sets of coordinates for each photo taken and then when the data is processed, the location data is corrected using a network such as the CORS network.
https://www.propelleraero.com/blog/how-it-works-gcp-vs-ppk-drone-surveying/
https://www.klauppk.com/product/dji-matrice-600-kit/
How will we display our data? (deliverables)
Our plan is to display our data using a combination of charts, videos, and pictures. We would show the average temperatures at different locations around the selected area, being sure to note important geographical features that may affect the temperature at other areas. We also plan to include composite images of both a 3D orthomosaic with thermal imagery on top. Our data points would include the temperature at a certain part of the day as well as the elevation for that point to see how elevation affects the change in temperature.Projected Schedule:
We plan to finish with research on September 8th and be certified with the equipment by the 10th. The data collection days will take place on the remaining Thursdays in September. We plan to have all of the data processed in Pix4D by the 10th of October and to continue working on the GIS day presentation throughout the month.
