NSF Org: |
TI Translational Impacts |
Recipient: |
|
Initial Amendment Date: | August 26, 2020 |
Latest Amendment Date: | October 13, 2020 |
Award Number: | 2026025 |
Award Instrument: | Standard Grant |
Program Manager: |
Peter Atherton
patherto@nsf.gov (703)292-8772 TI Translational Impacts TIP Dir for Tech, Innovation, & Partnerships |
Start Date: | September 1, 2020 |
End Date: | July 31, 2022 (Estimated) |
Total Intended Award Amount: | $241,820.00 |
Total Awarded Amount to Date: | $241,820.00 |
Funds Obligated to Date: |
|
History of Investigator: |
|
Recipient Sponsored Research Office: |
388 BEALE ST APT 1514 SAN FRANCISCO CA US 94105-4410 (650)656-5718 |
Sponsor Congressional District: |
|
Primary Place of Performance: |
353 Sacramento Street San Francisco CA US 94111-3620 |
Primary Place of Performance Congressional District: |
|
Unique Entity Identifier (UEI): |
|
Parent UEI: |
|
NSF Program(s): | SBIR Phase I |
Primary Program Source: |
|
Program Reference Code(s): |
|
Program Element Code(s): |
|
Award Agency Code: | 4900 |
Fund Agency Code: | 4900 |
Assistance Listing Number(s): | 47.041 |
ABSTRACT
The broader impact of this Small Business Innovation Research (SBIR) Phase I project will be to provide timely and highly localized climate forecasts, plus information such as extreme heat and frost risk, to insurance, energy, and agricultural stakeholders. Climate forecasting at sub-seasonal to seasonal (S2S) timescales is challenging, yet essential for proactive risk management of extreme natural hazards. This project will leverage artificial intelligence and cloud computing to implement a data-intensive approach for revolutionizing global climate forecasting. The project will provide efficient and accurate seasonal forecasts at relatively low computational cost in a user-friendly web environment.
This Small Business Innovation Research (SBIR) Phase I project aims to utilize advanced artificial intelligence techniques in order to develop a localized, timely, and reliable climate forecasting system that is industry-focused and crop-specific. In this project, state-of-the-art artificial intelligence techniques will be deployed to advance operational climate forecasting skill at a global scale. While conventional forecasts are trained exclusively on observational data, this project will train models on historical simulations and reanalysis, then evaluate them with observations. In this approach, the training dataset is substantially larger, consequently improving accuracy. This processing at scale is enabled with cloud resources.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
PROJECT OUTCOMES REPORT
Disclaimer
This Project Outcomes Report for the General Public is displayed verbatim as submitted by the Principal Investigator (PI) for this award. Any opinions, findings, and conclusions or recommendations expressed in this Report are those of the PI and do not necessarily reflect the views of the National Science Foundation; NSF has not approved or endorsed its content.
With the support of the National Science Foundation’s SBIR Phase I funding of award 2026025, ClimateAI was able to build an Artificial Intelligence Based Global Seasonal Forecasting System. The primary outcome of this project was the data and forecasting system required to power an agricultural web app for climate-related decision-making up to six months ahead. We achieved product market fit with 20+ customers using our Climate Enterprise Platform and achieved a significant improvement in decision-making skill.
Over the last 50 years, on average, one disaster related to climate or water risks occurred every day, causing 115 deaths and US$202 million in losses daily. These disasters will worsen in the next decades due to human-caused global warming, which is predicted to shave 11-14% off global economic output by 2050, accounting for a reduction of US$23 trillion to the global economic output.
Agriculture is the world's largest industry which generates over US$1.3 trillion worth and employs more than one billion people. Nevertheless, it is also one of the most vulnerable industries to changes in the climate. Higher temperatures eventually reduce crop yields and contribute to weed and pest proliferation. Unforeseen changes in climate patterns increase the likelihood of short-run crop failures and long-run production declines. In 2022 around 193 million people will have experienced acute food insecurity and ClimateAI is working to blunt this humanitarian need.
ClimateAi has developed a platform for providing timely and place-based operational climate forecasts (along with useful hyperlocal agriculture-specific information such as growing degree days, extreme heat risk, etc.) in a user-friendly web environment with practical visualizations, which provide a seamless experience for helping customers with their operations and climate-informed decision making. The technical backbone of our platform is an end-to-end extreme weather event forecasting system that is evaluated by our customers on the ability of making decisions using seasonal information. The system includes statistical and dynamical models, methods of achieving higher spatial and temporal cover, methods of optimally merging multiple forecasts simultaneously, and impact functions to connect weather to yield and quality metrics. The impact of our customer-centric S2S forecasting is better decisions, more resilience to a changing climate which leads to food security. With the National Science Foundation grant we have proved that AI based forecasts can deliver value to farmers and are looking forward to scaling our technology and approach to the entire agricultural market.
Last Modified: 08/29/2022
Modified by: Maximilian Evans
Please report errors in award information by writing to: awardsearch@nsf.gov.