Copy
View this email in your browser

As the world navigates the coronavirus pandemic, the importance of disaster preparedness and relief becomes increasingly apparent. The pandemic has shown us the power disasters have to alter our lives and brings to light what's broken in our emergency response systems. It's become painfully clear that effective disaster response starts with good and accessible data. 

Natural hazards don't halt because of a pandemic, and as regions around the world enter flood season, we're doing everything we can to ensure the governments and communities we work with are prepared for their next disaster. 

In solidarity,

Bessie 

Announcing a Global Partnership with Willis Towers Watson to Build and Scale Flood Parametric Insurance
Cloud to Street is thrilled to announce a new strategic partnership with Willis Towers Watson under the Willis Research Network to expand our work in flood insurance. 
 
Today, 90% of economic losses from disasters remain uninsured in the developing world, putting economically vulnerable households at greater risk and slowing recovery efforts following disasters. For the past two years, our team has collaborated with Willis Re, the reinsurance brokerage arm of Willis Towers Watson, to expand insurance coverage in Indonesia. Now, as an official member of the Willis Research Network, we’re building on this work to close the insurance gap in new markets. 
 
Through this partnership,  we will be working with Dr. Simon Young, a Senior Director in the Climate and Resilience Hub at Willis Towers Watson, to leverage Cloud to Street’s high resolution satellite and machine learning flood monitoring technology and create a reliable and scalable index-based insurance solution. This will allow us to better identify at-risk communities before a disaster strikes and rapidly detect flooded assets following disasters. Our first projects will be in Africa, and our team is keen to begin collaborating with insurers targeting this market.

For more information on our partnership, read our joint press release. You can also check out coverage of the announcement in Insurance Journal and Insurance Edge
AI Improved cropland dataset powering Congo farmer relief
Above, croplands along the Ubanji River are mapped using our improved dataset. 
When our team began working with the UN World Food Program and the Republic of Congo in 2018, many groups told us that the need to monitor croplands was a top priority. However, with no way to know where farms were located, effectively distributing flood relief to impacted croplands was almost impossible. This spring, we leveraged machine learning to fill this information gap, in the midst of record-breaking flooding.
Our scientists implemented a Random Forest classifier on satellite data in order to pinpoint where croplands actually are—creating a database with an accuracy rate of 94%, up from 20%. To determine which croplands were inundated after the floodwaters receded, we calculated mean NDVI (proxy for vegetation health) for each month in 2020 as well as for 2019. We found that decreases in NDVI value indicated crops in poor health. Using this method, the image on the right shows the change in impacted croplands between January 2019 and January 2020 in a region just north of the town of Impfondo.
With this improved dataset, the World Food Program can now better monitor how impacted farms are recovering from flooding—allowing them to better tailor their relief efforts and ensuring that aid is directed to farmers who need it most.
AI Research Accepted to EarthVision 2020's Computer Vision and Pattern Recognition Conference
We're proud to announce that Cloud to Street's research paper on the use of convolutional neural nets (CNNs) – a deep learning algorithm – in flood mapping was accepted to this year's EarthVision workshop on Computer Vision and Pattern Recognition (CVPR). The paper titled, Sen1Floods11:a georeferenced dataset to train and test deep learning flood algorithms for Sentinel-1, compares the efficacy of different types of flood detection training using CNN's. The team will be open-sourcing their new dataset, Sen1Floods11, for training CNNs to detect flooding in Sentinel-1 imagery alongside their training code and a few pre-trained models.
Dr. Venkat Laksmi Joins Water Science and Technology Board of the National Academies of Sciences, Engineering and Medicine
Congratulations to our Science Advisory Board Member, Venkat Lakshmi on being appointed to the Water Science and Technology Board of the National Academies of Sciences, Engineering and Medicine. As a board member, Venkat will help solve critical issues facing our nation’s water resources. 
Welcome our new team members
We're Hiring!
We're looking for a best-in-class Artificial Intelligence/Machine Learning Research Scientist to build tools for turning satellite data into actionable insights. They’ll work with our team of scientists and engineers with expertise in remote sensing (optical and radar), hydrology, climate, social vulnerability, UX, and machine learning to turn petabytes of satellite data into meaningful information to empower the world’s most vulnerable communities.
Apply Now
Twitter
Website
LinkedIn
Copyright © 2019 Cloud to Street, All rights reserved.


Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list.
 






This email was sent to <<Email Address>>
why did I get this?    unsubscribe from this list    update subscription preferences
Cloud to Street · 4938 McConnell Ave · Los Angeles, CA 90066-6714 · USA

Email Marketing Powered by Mailchimp