[Geors] Estimation of land potential available for agriculture using remote sensing

Estimation of land potential available for agriculture using remote sensing

Challenge:

Estimateland potential avilable for agriculture

Project idea in couple of sentences:

The project aims to create a model using Ghana as our area of interest for land cover, to map out inhabited land and reserves to determine the resulting land reserves.
We will then use remote sensing and GIS technique to estimate and categorize the potential areas for agriculture and potentially replicate the model on Cameroon.

The story behind the project: why are you specifically interested by this issue?

Available datasets to be used:

  1. Temperature: landsat data from Earth Engine
  2. Rainfall: CHIRPS data from Earth Engine
  3. Geology: shapefile from Ghana geological survey
  4. LULC: MODIS land cover data from Earth Engine
  5. Soil type: shapefile from Ghana geological survey

Methodology

  • AHP would be used to rank the factors of the map based on their influence on agriculture
  • Weighted overlay analysis would be used to form the final agricultural land potential
1 Like

1 Like

The prototype is a map showing indicators to better inform farmers in managing their plant growing, or exploring new opportunities, as well as expand agriculture to new or same plants.

The indicators (Soil quality: good/bad) are made with different data analysis that come from diverse data sources available and treatable through remote sensing like:

  • Slope
  • Soil aspect
  • Elevation
  • LULC
  • Rainfall
  • Distance from streams
  • Geology
  • Distance from roads
  • Humidity
  • Temperature
1 Like

The data engineering behind the scenes

IMG_3778 !

Defining what makes a good approach

IMG_3777 !

User journey: from the farmer to the prototype

IMG_3780

/ Pitch feedback

Vey interesting and technical project. This a innovative approach as an agro-ecological zoning in Cameroon doesn’t exist so far. The model you built could be very interesting to deploy and replicate.
Also it is relevant to estimate available land for agriculture. Very interesting for public authorities like Ministers as final users.

About the model:
Your method seems robust.
But what are the exact data that fuel your model?
Is it a simulation or a true model that you have built?

–> See Adeline to talk about the technical aspects

About the prototype model:
What are the data that require manual input into your model?
What would be your recommandations?

To improve
A lot of text on the slides
Start with “the problem in Africa”
About the list of the available datasets: say why you choosed them rather than by just giving the list, why they are interesting to your project or what you want to show. (are there crucial information?)
Give more details on your final users
Not necessary to show your algorithm in details (slide with the model) but better the result of it.

Pitch #1

The project aims to a create a model using Ghana as our area of interest for land cover, to map out inhabited land and reserves to determine the resulting land reserves for agriculture.

We will then use remote sensing and GIS technique determine to estimate and categorize the potential areas for agriculture and potentially replicate the model on Cameroon.

GEORS.pptx (663.9 KB)

Final Pitch :

An informed system that uses certain GIS data and remote-sensing technology to estimate potential areas for agriculture.

TEAM GEORS FINAL SLIDES.pdf (1.7 MB)

/ The data input:

/ The result

Hello,
I hope you are all well.
As a winner, you benefit from mentoring until the end of December to create a more complete version of your pitch. Concerning the next steps of your project, could you please indicate how you would like to move forward? This would allow Camille Pinet and I to organize ourselves.
Best regards,

Julien Chupin