Your browser has javascript turned off or blocked. This will lead to some parts of our website to not work properly or at all. Turn on javascript for best performance.

The browser you are using is not supported by this website. All versions of Internet Explorer are no longer supported, either by us or Microsoft (read more here:

Please use a modern browser to fully experience our website, such as the newest versions of Edge, Chrome, Firefox or Safari etc.

Climate adaptive decision support system

AI assisted radar rainfall algorithms for advanced engineering decision support system

Rainfall is vital for life on earth, yet capable of becoming a threat to humans by causing flash floods. Increasingly frequent droughts and extreme events call for developing knowledge on areal rainfall at resolutions as fine as a transient local storm to promote efficient, climate resilient solutions to the socio-environmental challenges concerning stormwater.


Compact high-tech X-band weather radars that are recently installed in Sweden (read more here) are unique instruments of providing such details for every grid of 100-500 meters on the ground by scanning weather normally at 4-6 vertical levels within a minute. However, currently employed methods for radar rainfall estimation are not optimized for the new technology and do not take into account the sporadic major error sources.

A new algorithm and interdisciplinary applications

The project validates a novel, deep learning algorithm development for high-resolution, spatiotemporally complete rainfall estimation that keeps the initial product’s resolution. Also, innovative, interdisciplinary applications are introduced for the new data. X-band weather radars were basically installed to prevent urban flooding in developed countries. So, new applications, e.g., in agriculture, forest, traffic management and building design areas, will increase the added value of an instrument with rather a high initial price, making it worthwhile for developing countries as well. 

A broad reference groups form Lund University, SMHI, Sweden Water Research, and water utilities of VA SYD and NSVA support the project.

This includes CMES researchers Prof Ronny Berndtsson and Hossein Hashemi who are leading related radar research at LU since 2018.

Research outcomes

Based on preliminary discussions with interdisciplinary stakeholders, the innovative framework by this project has high potentials to upgrade usual engineering judgments fundamentally because such data were never available for many applications. It is expected that by the end of the project, a complete list of applications is identified that could be a valuable benchmark for developing interdisciplinary decision-support systems, of course, with interrelated and synergetic aspects related to hydroclimatic variability at local scale. 

Photo by Noah Silliman on Unsplash

Exploratory Pre-Seed Programme of LU-innovation

LU-innovation Pre-Seed funding targets the innovative ideas that have active plans to continue after successful completion of the project.

The project in this stage will have deliverables including a developed framework, and a pack of algorithms, and descriptive, visualized outputs to easily illustrate its superiority over existing frameworks and algorithms. The plan for the next step is to seek an opportunity, e.g., by LU-innovation, to focus on a specific prototype of a decision-support system that will be upgraded to take full advantage of the innovation.