New Horizons in Conservation Conference Poster Submission

Python Remote Sensing Spatial Analysis Africa Agriculture MOSAIKS

Our capstone group attended the New Horizons in Conservation Conference, 2022, organized through the Yale School of the Environments Justice, Equity, Diversity, and Sustainability Initiative. The 2022 conference was held virtually, and our group contributed a poster and video explaing our capstone project including our preliminary results.

Cullen Molitor https://github.com/cullen-molitor , Juliet Cohen https://github.com/julietcohen , Steven Cognac https://github.com/cognack , Grace Lewin https://github.com/gracelewin
2022-04-23

An open-source pipleline for remote sensing of crop yields under environmental change in sub-Saharan Africa

ABSTRACT:

The environmental and social impacts of climate change are disproportionately distributed worldwide. Many highly impacted regions lack the assets to monitor and generate resource predictions, and therefore lack high-quality environmental and social data. As a result, it is difficult to make predictions about the impacts of climate change for these regions using conventional modeling. Recently, machine learning approaches applied to high-resolution satellite imagery have been successful in making predictions of a wide range of social and environmental variables. However, generating these predictions comes with significant barriers, including high computational, data storage, expertise, and financial resource costs. Reducing the financial and computational burden of machine learning approaches is essential to increasing the equity of environmental monitoring processes and outputs. Sub-Saharan Africa is one of these data-limited regions and is likely to suffer some of the largest impacts from climate change globally. To enable increased monitoring and data access across the sub-continent, we apply the novel machine learning approach, MOSAIKS, to create tabular features for sub-Saharan Africa using satellite imagery. These features, paired with ground-truth crop data from three countries, will be used to build a model that predicts crop yields over time in regions without crop data. This model will provide new insights into the historical crop yields of the region. Furthermore, these tabular features of satellite imagery, and the methodology developed to create them, will enable more people around the globe to build models and generate predictions of other social and environmental variables in this region.

Video submission

Poster submission

Please feel free to contact my team at cp-cropmosaiks.bren.ucsb.edu or inspect our code in our github organization, cropMOSAIKS. Thanks!

Citation

For attribution, please cite this work as

Molitor, et al. (2022, April 23). Cullen Molitor: New Horizons in Conservation Conference Poster Submission. Retrieved from cullen-molitor.github.io/posts/2022-04-23-new-horizons/

BibTeX citation

@misc{molitor2022new,
  author = {Molitor, Cullen and Cohen, Juliet and Cognac, Steven and Lewin, Grace},
  title = {Cullen Molitor: New Horizons in Conservation Conference Poster Submission},
  url = {cullen-molitor.github.io/posts/2022-04-23-new-horizons/},
  year = {2022}
}