Abstract. Creating spatially coherent rainfall patterns with high temporal resolution from data with lower temporal resolution is necessary in many geoscientific applications. From a statistical perspective, this presents a high- dimensional, highly under-determined problem. Recent advances in machine learning provide methods for learning such probability distributions. We test the usage of Generative Adversarial Networks (GANs) for estimating the full probability distribution of spatial rainfall patterns with high temporal resolution, conditioned on a field of lower temporal resolution. The GAN is trained on rainfall radar data with hourly resolution. Given a new field of daily precipitation sums, it can sample scenarios of spatiotemporal patterns with sub-daily resolution. While the generated patterns do not perfectly reproduce the statistics of observations, they are visually hardly distinguishable from real patterns. Limitations that we found are that providing additional input (such as geographical information) to the GAN surprisingly lead to worse results, showing that it is not trivial to increase the amount of used input information. Additionally, while in principle the GAN should learn the probability distribution in itself, we still needed expert judgment to determine at which point the training should stop, because longer training leads to worse results.
In hydrology, it is often necessary to infer from a daily sum of precipitation a possible distribution over the day – for example how much it rained in each hour. In principle, for a given daily sum, there are endless possibilities. However, some are more likely than others. We show that a method from artificial intelligence called GAN can learn what a typical distribution over the day looks like.
In hydrology, it is often necessary to infer from a daily sum of precipitation a possible...
- Printer-friendly version
- Supplement
