ClimateNet: an expert-labelled open dataset and Deep Learning architecture for enabling high-precision analyses of extreme weather

Abstract

Abstract. Identifying, detecting and localizing extreme weather events is a crucial first step in understanding how they may vary under different climate change scenarios. Pattern recognition tasks such as classification, object detection and segmentation have remained challenging problems in the weather and climate sciences. While there exist many empirical heuristics for detecting extreme events, the disparities between the output of these different methods even for a single event are large and often difficult to reconcile. Given the success of Deep Learning (DL) in tackling similar problems in computer vision, we advocate a DL-based approach. DL, however, works best in the context of supervised learning; when labeled datasets are readily available. Reliable, labeled training data for extreme weather and climate events is scarce. We create ClimateNet – an open, community-sourced human expert-labeled curated dataset – that captures tropical cyclones (TCs) and atmospheric rivers (ARs) in high-resolution climate model output from a simulation of a recent historical period. We use the curated ClimateNet dataset to train a state-of-the-art DL model for pixel-level identification, i.e. segmentation, of TCs and ARs. We then apply the trained DL model to historical and climate change scenarios simulated by the Community Atmospheric Model (CAM5.1) and show that the DL model accurately segments the data into TCs, ARs or the background at a pixel level. Further, we show how the segmentation results can be used to conduct spatially and temporally precise analytics by quantifying distributions of extreme precipitation conditioned on event types (TC or AR) at regional scales. The key contribution of this work is that it paves the way for DL-based automated, hi-fidelity and highly precise analytics of climate data using a curated expert-labelled dataset – ClimateNet. ClimateNet and the DL-based segmentation method provide several unique capabilities: (i) they can be used to calculate a variety of TC and AR statistics at a fine-grained level; (ii) they can be applied to different climate scenarios and different datasets without tuning as they do not rely on threshold conditions; and (iii) the proposed DL method is suitable for rapidly analyzing large amounts of climate model output. While our study has been conducted for two important extreme weather patterns (TCs and ARs) in simulation datasets, we believe that this methodology can be applied to a much broader class of patterns, and applied to observational and reanalysis data products via transfer learning.

Detection of atmospheric rivers with inline uncertainty quantification: TECA-BARD v1.0.1

“…There is a growing body of literature in which researchers decompose precipitation and other meteorological processes into constituent weather phenomena, such as tropical cyclones, extratropical cyclones, fronts, mesoscale convective systems, and atmospheric rivers (e.g., Kunkel et al, 2012;Neu et al, 2013;Walsh et al, 2015;Schemm et al, 2018;Wehner et al, 2018). Research focused on atmospheric rivers (ARs) in particular has contributed a great deal to our understanding of the water cycle (Zhu and Newell, 1998;Sellars et al, 2017), atmospheric dynamics (Hu et al, 2017), precipitation variability (Dong et al, 2018), precipitation extremes (Leung and Qian, 2009;Dong 6132 T. A. O'Brien et al: AR detection with UQ: TECA-BARD v1.0.1 et al, 2018), impacts (Neiman et al, 2008;Ralph et al, 2013Ralph et al, , 2019a, meteorological controls on the cryosphere (Gorodetskaya et al, 2014;Huning et al, 2017Huning et al, , 2019, and uncertainty in projections of precipitation in future climate change scenarios (Gershunov et al, 2019b).…”

Opposite Interdecadal Trends of Summer Atmospheric Rivers over East Asia and Western North Pacific in Recent Decades

“…Empirically, p 0 is 1000 hPa and p top is 200 hPa or 300 hPa with enough pressure levels containing most of the moisture to detect ARs. Besides, another AR algorithm ClimateNet (Prabhat et al, 2020) (See Table S1) based on deep learning and trained by expert-labeled AR images is also compared at the end of the article.…”

Outlook for Exploiting Artificial Intelligence in the Earth and Environmental Sciences

“…Preparing the data (e.g., labeling) for AI exploitation is a notable challenge in some applications. This critical but often overlooked step has attracted some recent attention (e.g., Bonfanti et al 2018;Lee et al 2019;Prabhat et al 2020) and should be an emphasis of future efforts by prediction centers. This will not only provide more readily available datasets for AI exploitation, it should also in principle allow more creative ways to exploit satellite data.…”