CCdownscaling: A Python package for multivariable statistical climate model downscaling

Future climate projections are made with global numerical models whose spatial resolution often exceed 100s of km2. These scales are too large to resolve many weather events, leaving a gap between the climate information needed to understand the impact of climate change on many human activities, and the information that can be provided by global models. Regional climate projections generated using statistical downscaling methods can provide an essential bridge between global climate models and the high spatial resolution data needed. As the demand for localized climate information continues to grow, new software tools are necessary to provide downscaled climate information. In this article, we describe CCdownscaling, a software package that provides multiple statistical climate downscaling methods to the station scale, including the Self Organizing Maps method. CCdownscaling includes several evaluation metrics for assessing the skill of downscaled climate information in various applications, and we demonstrate these features on an example dataset.

© This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/