Tracking raindrops, one molecule at a time

Science Daily  October 25, 2022 A team of researchers in the US (U Mass. Amherst, University of Alaska) studied the mechanisms of precipitation deuterium excess (d-excess) seasonality in low-latitudes and mid-latitudes through a new analysis of precipitation isotope databases along with climate reanalysis products and moisture tracking models. The ultimate d-excess signals are produced after complex modulations by several reinforcing or competing processes. They developed a simple seasonal water storage model to show that contributions of previously evaporated residual water storage and higher transpiration fractions may lead to relatively low d-excess in evapotranspiration fluxes during periods of enhanced continental moisture […]

Researchers explore using light to levitate discs in the mesosphere

Phys.org  February 15, 2021 To improve weather prediction sensors need to be sent to mesosphere. The satellites and rockets currently used have problems as the air is too thick and friction and heat would make long-duration flights impractical. Researchers at the University of Pennsylvania constructed and demonstrated light-driven levitation of macroscopic polymer films with nanostructured surface as candidates for long-duration near-space flight. The disks were made of 0.5-micron-thick mylar film coated with carbon nanotubes on one side. When illuminated with light intensity comparable to natural sunlight, the polymer disk heats up and interacts with incident gas molecules differently on the […]

Developing models to predict storm surges

Science Daily  September 8, 2020 To develop the models researchers at the University of Central Florida linked large-scale climate variability events, such as El Niño, to variability in storm surge activity. They tested the models by having them predict past storm surge variability and compared their predictions with what actually occurred. The results indicated that the models matched the overall trends and variability of storm surge indicators for almost all coastal regions of the U.S during both the tropical and extra-tropical storm seasons. According to the researchers there is some capability in predicting storm surge variability over inter-annual to decadal […]

Computer scientists predict lightning and thunder with the help of artificial intelligence

Science Daily  June 26, 2019 Current satellite-based approaches to predict thunderstorms are usually based on the analysis of the observed brightness temperatures in different spectral channels and emit a warning if a critical threshold is reached. Researchers in Germany have developed a method using the error of two-dimensional optical flow algorithms applied to images of meteorological satellites as a feature for machine learning models. They trained different tree classifier models as well as a neural network to predict lightning. The results show a high accuracy of 96% for predictions over the next 15 minutes which slightly decreases with increasing forecast […]