RunMyProcess’ Post

Artificial intelligence can provide flood warnings earlier and over larger and more impactful events in ungauged basins! #Floodmodelling #ungaugedcatchments #AI

The frequency of devastating floods has increased worldwide. The damages and casualties can be decreased if the flood warning is issued well in advance. Globally, the flood warning is issued hardly 24 hours before. The need to forecast a few days in advance will help reduce damages and result in zero casualties. Globally, researchers are working on developing ensemble flood forecasting models with T-day lead-time forecasting. In India,very little study has been done in this area. Our research paper focuses on the use of ensemble flood forecasting for the Sabarmati River Basin of India. Lots of computation, patience, and time are required for the analysis. Congratulations, Annant Patel, Assistant Professor, Nirma University, Ahmedabad, for your tireless efforts!!

Results of a new study suggest that the artificial intelligence weather models can provide flood warnings for both small and extreme events in ungauged basins with a longer lead time than previous methods and could improve access to reliable flood forecasting for developing regions. Learn more in our latest blog post: https://vist.ly/zdpx. #Weather #AI #ArtificialIntelligence #Flood #Flooding #Forecast

LSTM is not new and we have piloted the same conecpt more than a decade ago at the flood forecast center in Manitoba. However, the discussion often omits crucial details such as the lead time and accuracy of forecasts extending beyond a few days, especially in regions lacking reliable real-time and historical gauge data. Moreover, predicting extreme events that have no historical precedent remains a challenge. Despite these gaps, this approach could be invaluable in less developed areas where dedicated flood forecast teams are unavailable. While “flood prediction” garners significant attention from venture capital and the tech industry, it’s important to also focus on “drought prediction,” given that droughts cause global losses of approximately $70 billion annually. Addressing the world’s water and sustainability challenges requires collaboration among the brightest minds in computer science and civil engineering. This is precisely why I embarked on this journey: to merge the fields of AI and civil engineering through my startup, Urbanlytics Solutions. If you're driven to make a difference, let’s join forces and innovate for a sustainable future! #FloodForecasting #DroughtManagement #Sustainability #AI #CivilEngineering #TechForGood #Innovation

A study published yesterday by Nature showed how AI is revolutionizing flood forecasting globally, while protecting 460 million people in 80+ countries! They have found that AI helped to provide more accurate information on riverine floods up to 7 days in advance! Check out the publication here: https://lnkd.in/gJxQGeyH #prevent #naturaldisasterprevention #AI #deeptech

What I can add to Lindsay Millard that you can not deal with natural process like flood with black box methdologies (AI model) or Statistics. I think the development in earth obervations is the only logical pathway. With high quality data and a more advancement in weather systems understanding, we could reach maybe a confident level in flood forecasting. #floodforecasting

Large-scale global flood forecasting has been out of reach for a long time. That has left vulnerable communities particularly in developing countries cut off from potentially life-saving information. Today we showcase how breakthrough AI technologies can close the gap and provide reliable flood predictions around the world. The stakes for accurate flood forecasting are high: some 19% of the world population is directly exposed to substantial risks from severe floods. Flooding also exacts an immense material toll, causing around $50 billion in annual global economic damages. In our study published today in Nature we demonstrate how our AI model can significantly improve the accuracy and lead time of flood forecasts compared to the global state-of-the-art flood forecasting system. Thanks to AI, flood predictions in regions of Africa and Asia are now comparable to forecasts currently available in Europe. Our model on average also extends the reliability of currently-available global nowcasts from zero to five days. Learn more at: https://goo.gle/49bRqVG

Extraordinary developments in AI for flood forecasting. Are we seeing the scales slowly tipping in favour of inductivism over deductivism?

A very active area of development these days. There are a lot of entrants into this market all looking to solve the same problem. With so many: what is the viable business model for them or will it be winner takes all?, where do public agencies fit in?, how do we stop the public being confused by various messaging platforms as confusion can lead to harm or panic. And finally, if it isn't based on physics and it is just an inexplicable black box, what will the courts think when it is eventually tested. Certainly an exciting time for the field but I wonder what the next decade is going to bring. If you can forecast floods a month out, you can probably forecast the stock market/commodity prices/bond yields and I'm not aware of anyone claiming to have solved those. 

Our latest Success Story shows how we built a flood prediction AI model for the flood-prone areas of Bangladesh. Flooding in the Ganges River Delta threatens the lives of the people of Bangladesh. Rising sea levels combined with unpredictable rainfall and the country's low topographical elevation have severely exacerbated flooding in the country, making it one of the world's most vulnerable regions to climate change. Accurate flood prediction in this scenario can mean the difference between life or death for millions of people. AI is highly suited for making these prediction models more useful and accurate. Learn how by reading our blog! #AI #AIforGood #Bangladesh #FloodPrediction #Omdena