Embedded Files
This educational project is maintained by students and a work in progress. Final version is expected september 2025.
6th grade
We are team "Terrain Explorers," conducting research for the ESA Climate Detectives competition.
Stay updated on our project and follow our progress here!
Phase 1: preparatory research AND dossier preparation
Phase 1: preparatory research AND dossier preparation
12 september '24
Teachable Machine
Teachable Machine
On our Chromebook, we experimented with Teachable Machine—a web-based tool by Google that lets anyone create machine learning models without writing a single line of code. We trained our model to recognize different images.
26 september '24
Webcam hack IR
Webcam hack IR
We modified a standard webcam by removing the IR filtering glass and adding polarizing filters, which enhance contrast in our photos. This makes it easier to distinguish real plants from artificial ones.
Submitting final file phase 1
Submitting final file phase 1
We formulated our research question and action plan and submitted them to ESERO. Next, we reached out to the weather station in Flobecq, but we are still awaiting a response regarding potential data collaboration.
Phase 2: development of research
Phase 2: development of research
17 october '24
Power supply for solar panel, soldering, wiring schedule
Power supply for solar panel, soldering, wiring schedule
We soldered the power supply to the solar panel, allowing it to run on solar energy. This eliminates the hassle of constantly replacing batteries.
17 october '24
Research + choice ASA plastic 3D-design
Research + choice ASA plastic 3D-design
The Low Warp ASA filament from REAL is the perfect material for outdoor prints. Its water-, UV-, and heat-resistant properties make it highly durable against moisture and weather conditions. Additionally, it offers excellent impact resistance, minimal shrinkage, and greater flexibility and hardness than ABS, making it a popular choice among users. It prints easily without a heated chamber and supports relatively high speeds. With its versatility, Low Warp ASA allows us to create strong, functional outdoor prints with ease.
17 october '24
3D design automatic pluviometer / reed contact
We designed a pluviometer in Tinkercad, which we later 3D-printed to measure rainfall. The reed switch was provided by our teacher, while the mounting component was also designed in Tinkercad by the same student.
17 october '24
Schedule, test, coding the soil sensor (soil moisture)
Schedule, test, coding the soil sensor (soil moisture)
Testing the soil sensor by dipping it in a small bucket of water, while simultaneously coding it, so the sensor doesn't give us the wrong results.
24 october '24
24 october '24
Difference between real and fake plants
A new student used our modified webcam and Teachable Machine to tell real plants from fake ones. The polarizing filters in the webcam improve contrast by allowing specific light directions, helping the machine learning model to learn and make accurate distinctions between the real and fake plants.
24 october '24
24 october '24
Coding the pluviometer.
Coding the pluviometer.
Coding the pluviometer to count how many times the bucket tips, this can let us know how much rainfall there was.
26 november '24
26 november '24
Advice from experts from KMI & VITO!
We arranged a meeting with an expert from KMI where we introduced our project and asked questions that could possibly help us further.
Later we had another meeting with an expert from VITO who also gave us helpful advice for our project.
3D-printing
3D-printing
9 January '25
During the Christmas holidays, there were hours of 3D-printing to have all the pieces ready in time. Now we can start assembling the final sensor.
last changes of the code
last changes of the code
9 January '25
9 January '25
The code has been carefully tested, and where necessary changes have been made to ensure optimal performance and accuracy. In addition, the pluviometer is fully wired, with all connections properly made and checked to ensure reliable operation of the device.
installation of the raspberry pi with PInoir camera
installation of the raspberry pi with PInoir camera
9 January '25
9 January '25
We connected the keyboard, mouse, and camera. We made the code to take pictures with the Pi noir camera.
first test with the pinoir camera
first test with the pinoir camera
9 January '25
9 January '25
We tested the Pi noir camera to see what the photos looked like. With a Pi noir camera, you can also see in the dark with infrared lighting. The infrared can make the photos look strange during the day.
Perfecting the pinoir camera
Perfecting the pinoir camera
16 January '25
16 January '25
Schematize wiring
Schematize wiring
16 January '25
16 January '25
We made a sketch for the complete wiring of different sensors, we will connect these on a printed circuit board with each other and with the raspberry pi pico. Furthermore we also start with soldering the first wires.
Continued work on the website
Continued work on the website
16 January '25
16 January '25
We continued working on the website. They collaborated to implement new features, resolve existing issues, and ensure that the website met the desired standards of quality and usability.
Soldering wiring diagram
30 January '25
30 January '25
We continued to solder the wires between the OLED-display, rain gauge, soil moisture sensor and the Raspberry Pi Pico. This was all connected to each other on a printed circuit board.
Contrast change photo's
Contrast change photo's
30 January '25
30 January '25
We tried converting photos using a code that should process images directly with nIR. At first, there were some small issues, but after a few adjustments, it worked as expected. We used a Raspberry Pi and programmed in Thonny. Now, we are going to test it with our own photos.
We tried converting photos using a code that should process images directly with nIR. At first, there were some small issues, but after a few adjustments, it worked as expected. We used a Raspberry Pi and programmed in Thonny. Now, we are going to test it with our own photos.
Working on a video explaining the project
Working on a video explaining the project
6 february '25
We explained the different steps of the project in a video, so people can understand what we did.
Trying to automatically take photos
Trying to automatically take photos
6 February '25
We worked together on how to let the raspberry pi take photos on it's own, without any human control.
Testing the Pluviometer
Testing the Pluviometer
6 February '25
We tried to measure how many milliliters of water it took for the pluviometer to tilt. We ended figuring out that the quantity differed because of friction.
Per measurement, we poured 25ml of water using a stand and buret. We measured a total of 5 times.
1st time: it tilted 14 times
2nd time: 14 times
3rd time: 11 times → a lot more water was necessary on the right side to tilt.
4th time: 8 times → after the first tilt it would create friction which caused the next one to take way more water and overflow, it became very inconsistent.
5th time: 9 times → same as the 4th, friction that caused the measurements to be off.
For next time, we have an idea which could help make the measurement more consistent and better.
online research
online research
13 February '25
Today there was a strike. The teacher wasn't there, so we did some research with the entire group.
PUTTING THE PROJECT TOGETHER
13 march '25
The other part of the class put together the project and hung the solar panels on the stand
Phase 3: processing data/ comparison/ conclusions
Phase 3: processing data/ comparison/ conclusions
SATELLITE DATA VIA TERRASCOPE
SATELLITE DATA VIA TERRASCOPE
20 march '25
We found satellite data on Terrascope on the location where a student’s family runs a farm in Everbeek-Beneden (Brakel). This was the perfect opportunity to use their plots for our project.
Processing of external Data
Processing of external Data
We received offcial weather data at the KMI for the nearest station in Chièvres for our test location. This is aprox 20 km from the our test field. We turned this data to graphs.
ACTIONS on short-term: Use of adapted crops
ACTIONS on short-term: Use of adapted crops
At our testsite, maize and potatoes are grown. These crops are sensitive to both drought and excessive rainfall. Potatoes, in particular, require a regular water supply, but too much water can lead to root rot. Grain crops, on the other hand, vary in their tolerance depending on the type of grain. Some types are more resistant to water, while others can better withstand drought.
Conclusion comparison satellite data
Conclusion comparison satellite data
27 march '25
We compared the satellite data from early October with that from December. it is was noticeable that there are more yellow plots in December, including some of our plots.
actions on long term:
Soil cultivation and soil permeability
actions on long term:
Soil cultivation and soil permeability
Farmers can use non-inversion tillage to improve soil structure and water retention, though its benefits appear after several years and require new equipment.
Adjusting soil permeability with materials like sand helps prevent waterlogging.
Management agreements encourage nature-friendly measures like grass strips to manage water and support biodiversity.
These actions help make farming more sustainable and resilient to extreme weather.
Page updated
Report abuse