Volume 33, Issue 2, May 2020
Volume 33, Issue 2, May 2020
APPLICATIONS OF DRONE TECHNOLOGY IN GEOGRAPHY
Dr Karen Joyce, Senior Lecturer, James Cook University
A Facelift for Geography
If you’ve ever played a general knowledge quiz like Trivial Pursuit, you may think geography is all about knowing the capital city, population, or flag of every country in the world. Well, that might be one aspect, but geography is so much more than that!
The global phenomenon National Geographic takes a broader view of geography as a discipline. And certainly the millions of people watching their documentaries and reading their magazines do too! Though surprisingly some people never connect the geography we learn at school with such a popular brand. It’s time to popularize geography and show people how cool it really is. We’re going to use a guest star – the drone. But before we go there, let’s start with a definition – geography is ‘the why of where’. All of a sudden, that’s a really big science! And yes, it is a science, because it asks scientists to question ‘why’.
But Geography isn’t STEM…or is it?!
Geography is actually an example of an integrated discipline that cuts across all aspects of STEM. This integration is also key in its facelift. You’ll see that in the case studies following, but just to demonstrate in brief: Science: We ask questions about why we see different features in certain locations in the environment, and also map and monitor changes across locations. Technology: We use equipment such as GPS devices, laser scanners, and drones. Engineering: We build, use, and maintain hardware and software solutions, including field survey equipment. Maths: We measure speed, distance, time, volume, area… all of which help us monitor environmental phenomena.
Figure 1: Geography as an integral part of STEM
Drones and Geography – A Mutualistic Relationship
And now for our special guest star! Most people acknowledge that drone technology is considered STEM, but did you know that almost all drones use geographic technology? We rely on their onboard GPS to make sure they are flying in the correct location, at the correct speed and altitude, and also to stamp every photo with that same information, called metadata (i.e. data about the data!).
There are many different applications of drone technology, but here we are going to focus on their ability to capture data for earth observation and environmental mapping purposes. Drones are a fabulous tool for geographical scientists to study the environment and how it changes over time.
How Do Drones Make Maps?
Because the drone stamps every photo with locational metadata (i.e. latitude, longitude, altitude), it means that we are able to place our individual pictures on a map, within a Geographical Information System (GIS). But often just one photo isn’t enough to cover the entire area in which we are interested. Therefore we need to capture many overlapping photos by flying a mission in an ‘aerial lawnmower’ pattern (Figure 2a,b), and then use a process called structure from motion to stitch the images together (Figure 2c).
Structure from motion uses a series of algorithms to identify identical features within image overlapping areas to create an orthomosaic (Figure 2d). This means that the distortions are removed from the resultant image, and we are able to make measurements from it. As part of this process, we also create a 3D model, which allows us to measure the height of buildings, trees etc, as well as ground elevations (Figure 2e).
The final stage of this process is to analyse the orthomosaics to provide information and understand the environment of interest. Figure 2 shows the regrowth of native grasses of a different colour at a lower growth height than the surrounding invasive buffel grass (Cenchrus ciliaris). In situ surveys also document the return of native fauna as well.
Figure 2: Drone imagery shows what happens when we exclude buffel grass (Cenchrus ciliaris) from an ecosystem. (A) Drone flight plan overlaid on available satellite imagery within Google Earth; (B) One of 122 individual images captured on the mission showing far greater detail than satellite data; (C) Images are located in space and structure from motion algorithms create 3D point clouds; (D) Orthomosaic created from 122 individual drone images; and (E) 3D model showing vegetation height.
Case study 1: That’s a Lot of Poop!
The Great Barrier Reef is home to many thousands of different species of organisms, all playing a vital role in maintaining the health of the ecosystem. One such creature is a holothurian, or sea cucumber. Holothurians are like the vacuum cleaners of reef sediment – they chew through small particles in the sand, facilitating availability of organic and inorganic nutrients to other reef organisms in a process called bioturbation.
Although scientists acknowledge the role holothurians play in bioturbation, we know less about the magnitude of that role. To understand that better, we need to know how many holothurians live on the reef in the first place. This is also important as unfortunately holothurians are considered a delicacy in many parts of the world, and overfishing means that they may not be providing the valuable ecosystem services that the reef needs.
We can use drones to count the number of individuals living in small areas of Heron Reef (Figure 3), and then satellites to estimate the total number across the reef. As we know that an individual holothurian (based on Holothuria atra – Figure 3d) is estimated to produce 14 kg of bioturbated sediment (AKA poop) per year, we can multiply that by the total number of holothurians counted. We estimated that their total amount of poop was about 40,000 tonnes per year! That’s about the same as four times the weight of the Eiffel Tower!
Figure 3: We can use drones and satellites to estimate the numbers of holothurians on Heron Reef. (A) Study site location; (B) Orthomosaic of 627 drone photos; (C ) From a drone, holothurians are clearly visible against the sandy background; and (D-F) Different holothurian species found on Heron Reef.
Case study 2: Measuring light
We’ve seen how easy it is to use drones to map features like holothurians on the reef when we can easily see them in the photos we take from above. But we can also use drones as a platform on which to mount sophisticated sensors to give us more information about biophysical processes in the environment.
For example, often we think of healthy trees as showing a green colour, due to the chlorophyll their cells contain to use in the biophysical process of photosynthesis. But how green is green? And can we use colour to measure the amount of chlorophyll?
Although our eyes are amazing, they can’t quantify how much light is actually getting reflected or absorbed. It’s these light interactions that we use to give us information about environmental health, and we use a spectrometer to make the measurements.
With a spectrometer flying on a drone, we make measurements from above in very specific wavelengths of light from the electromagnetic spectrum. We know that chlorophyll uses (absorbs) red light with a wavelength of around 675 nm for photosynthesis. Using a drone we see that it’s not just terrestrial plants that absorb light for this process (Figure 4)!
Figure 4: (A) Study site and sample location; (B) Small spectrometer mounted on a drone to analyse biophysical properties; and (C) Individual spectral signatures. The small ‘dip’ in each of the curves around 675 nm indicates light absorbed for photosynthesis.
This spectral signature graph shows how various features reflect and absorb light. We can see that there’s likely to be chlorophyll that’s absorbing red light in sand, shallow water, and even beach rock. That’s actually not surprising as lots of photosynthetic algae make their home on coral reefs as well.
This type of technology not only allows us to map photosynthetic potential, but also to identify differences between tree species (e.g. Pisonia and Casuarina), and even canopy densities.
Case study 3: How Cool Is Your School?
I don’t think that anyone would dispute that it’s lovely to sit under a shady tree on a hot day! Obviously the trees block the direct heat of the sun, but they also help to cool us by a process called ‘transpiration cooling’. Water passes from the soil and through the plant before eventually exiting via leaves and branches. This helps to cool the plant, just like when we sweat. But as the water leaves the plant as gas or water vapour, this helps cool the air around it as well – by up to 5oC! This transpiration cooling effect causes plant shade to be cooler than building shade.
So how cool do you think your school is? We can use drones to map trees and shade to find this out! Through the processes described above, we capture and orthomosaic drone data over school campuses.
With the #MapMySchool project, we map the total area of a school, and the total area covered by trees, and then use these figures to calculate the percentage area of shade on a school campus. Once we have this information, students and staff can also plan where they might like to plant more trees to make their school even cooler.
Top Tips for Using Drones in School
- Make sure you are aware of the legal and safety requirements for flying drones before you decide to take off!
- Figure out the minimum characteristics of the data that you need to answer your question BEFORE you select your geospatial tool. That is, don’t race out to buy the newest, shiniest drone on the market until you are sure that it will serve your purpose!
- Start with small, inexpensive drones to gain skills and confidence. It’s likely that you will crash when you first start to fly, so far better to do this with something that is less likely to cause physical or financial damage.
- Remember to include software, data storage devices, and sufficient training (for software and drone flying) into your drone start up budget.
Drones are undoubtedly one of the coolest innovations to grace the world of geography recently! But we go far beyond using them as toys, or for just taking pretty pictures. They are valuable tools for collecting data that we can use to answer real environmental challenges. Using drones as the basis for capturing data in the above examples (and many more!) is just one way to clearly demonstrate the value of geography as an integral part of STEM.
1. a) How many holothurians can you count in Figure 3c? What other different reef features can you identify?
b) If you had to train a computer (artificial intelligence) to identify holothurians, what sorts of unique characteristics could you use?
2. Visit https://arcg.is/11rGOf to learn more about the #OpenReef drone mapping run by ‘Citizen Science GIS’ at the University of Central Florida.
a) Discuss the role that drones might play in the future for citizen science projects around the world.
b) Choose one vocation that requires geographical understanding and explain how drone technology could be used in the collection of primary geographical data that would be used in such a workplace.
Watch this video (https://youtu.be/XUlQRA8-Lr4) and use it to help you create a table to document the advantages and disadvantages of using drone imagery for creating maps. Compare and contrast your findings with more traditional data capture methods such as in-situ survey, or satellite imagery.
4. a) Formulate a question about a geographical issue that you have studied, and that could be investigated using drone technology.
b) Outline the primary geographical data that could be collected using drone technology to investigate the question and inform management of the geographical issue.
c) Outline secondary data that would be useful in your investigation of the geographical issue.
5. a) Work out how cool your school is! Even if you don’t have access to a drone, you can still participate in the #MapMySchool project – check out http://mapmyschool.com.au/ Where would you plant more trees on your school grounds?
b) What other sorts of applications could you use a drone for at your school?
6. Learn about Australia’s laws for flying drones and take the quiz to test your knowledge https://www.casa.gov.au/knowyourdrone