UNIVERSITY PARK, Pennsylvania — Machine-learning techniques that can recognize human faces could also help improve weather forecasts, according to a team of scientists.
“The idea behind this study comes from Google’s FaceNet, but we don’t compare your photo to images of faces in our database, we compare weather to historical forecasts.” A machine learning scientist at the University of San Diego and a former PhD student at Penn State University.
Scientists have applied deep learning algorithms to analog weather forecasts that use past weather conditions to make future predictions. In this case study, using machine learning to analyze predictions of surface wind speed and solar irradiance in Pennsylvania from 2017 to 2019, he found that the accuracy of analog predictions improved. .
“We want to know how much energy we can expect for the day,” said Hu, who has a doctorate in geography from Pennsylvania State University. “We need to understand the risks. Whether we overestimate or underestimate, we end up with penalties such as power shortages and overproduction. It shows that we can improve.”
Analog forecasting is an alternative to numerical weather forecasting (NWP) that uses computer models to simulate how initial weather conditions will change in the days or weeks ahead. Although the NWP has made great advances in forecasting over the last few decades, uncertainties remain.
These uncertainties are partially addressed by running a number of simulations called ensembles. The simulations show a range of possible future atmospheric conditions, but they are also computationally intensive and expensive to create, the scientists said.
“But with analog prediction, you can generate ensembles without running expensive model iterations,” says Hu. “It works by finding the past predictions that are most similar to the target prediction, and the past observations associated with the most similar past predictions make up the members of the ensemble.”
Analog ensembles combine deterministic forecasts (one very detailed run of the NWP model) with past weather observations (temperature, pressure, humidity, etc.) from past forecasts that are similar to the current forecast. is generated.
The best similar forecast is selected based on a similarity metric that weighs individual weather predictor variables, but the process uses a constrained exhaustive search that limits the number of predictor variables that can be used to reduce the Relationships between variables are not considered.
“That was the limit of analog ensemble prediction,” says Hu. “In this paper, we try to address this problem by introducing a machine learning approach that learns the complexity between predictor variables.”
Machine learning techniques take all the weather variables, such as temperature, pressure, and humidity, and transform them into a latent space, or clustered patterns that help select ideal forecasts and similarities, says science. they said.
“This approach attempts to identify the most useful features to look for to improve analog predictions,” said Hu. “Simply put, clustering the candidates yields the most accurate predictions and keeps data similarities away from less similar predictions.”
The machine learning technique overcomes the computational limitations posed by predictor weight optimization in traditional analog ensemble prediction, the scientists said.
“Machine learning has been operational for many years to make predictions faster and more accurate, but its role has been largely limited to post-processing and data preparation,” said Pennsylvania State University professor of geography, Meteorologist and atmospheric scientist Guido Cerbone said. Advisor and co-author of papers. “He’s been around for a year or so where machine learning has been used as the core of algorithms, often even replacing numerical model solutions.”
Research results published in the journal Boundary-Layer Meteorology suggest that machine learning will make more predictors available and produce more accurate predictions.
“Our work shows that machine learning models can also be used in the geosciences to probe complex features,” Hu said. “In earth science, we deal with hundreds of variables. We show that all these relationships can indeed be detected from a data set that is complex.”
George Young, professor emeritus of meteorology at Pennsylvania State University, and Luca Delle Monache, deputy director of the Western Center for Weather and Water Extremes at the Scripps Institute of Oceanography at the University of California, San Diego, also contributed.
The National Science Foundation and the Pennsylvania State Institute for Computational Data Science funded this research.
