Artificial intelligence achieves easy-to-adapt nonlinear global temperature reconstructions using minimal local data

Abstract

Understanding monthly-to-annual climate variability is essential for adapting to future climate extremes. Key ways to do this are through analysing climate field reconstructions and reanalyses. However, producing such reconstructions can be limited by high production costs, unrealistic linearity assumptions, or uneven distribution of local climate records. Here, we present a machine learning-based non-linear climate variability reconstruction method using a Recurrent Neural Network that is able to learn from existing model outputs and reanalysis data. As a proof-of-concept, we reconstructed more than 400 years of global, monthly temperature anomalies based on sparse, realistically distributed pseudo-station data and show the impact of different training data sets. Our reconstructions show realistic temperature patterns and magnitude reproduction costing about 1 hour on a middle-class laptop. We highlight the method’s capability in terms of mean statistics compared to more established methods and find that it is also suited to reconstruct specific climate events. This approach can easily be adapted for a wide range of regions, periods and variables.

Inteligencia Artificial - Chatgpt: un estudio bibliométrico y su aplicación en el caso de Gases de Efecto Invernadero

“…Por su parte ChatGPT, funciona utilizando el modelo de lenguaje GPT (Generative Pre-trained Transformer) de OpenAI o Modelo de Lenguaje Grande que crea texto rastreando miles de millones de palabras de datos de entrenamiento y aprendiendo cómo las palabras y frases se relacionan entre sí (Stokel-Walker & Van Noorden, 2023). ChatGPT tiene el potencial de ser aplicado en diferentes campos de estudio, destacándose, entre otros, los siguientes: educación (Huh, 2023;Lo, 2023), ciencias de la salud (Cifarelli & Sheehan, 2023;Archibald & Clark, 2023;Alser & Waisberg 2023), periodismo (Pavlik, 2023), variabilidad climática (Wegmann, 2023), redacción y publicación científica (Else, 2023;Thorp, 2023;van Dis et al, 2023;Ali & Djalilian, 2023;Anderson et al, 2023), biología (Agathokleous et al, 2023), biotecnología (Holzinger et al, 2023), simulaciones demográficas (Guliyev, 2023); y tecnologías de impresión 3D (Badini et al, 2023). En adición a ello, este leguaje tiene el potencial de identificar prioridades de investigación y proporcionar un punto de partida para futuras investigaciones (Lahat et al, 2023).…”

Towards variance-conserving reconstructions of climate indices with Gaussian process regression in an embedding space

“…As in many disciplines, machine learning methods have successfully gained traction in the climate reconstruction community (e.g. Michel et al, 2020;Zhang et al, 2022;Wegmann and Jaume-Santero, 2023). Here, we explore the potential of the non-linear supervised learning method Gaussian process regression (GPR) for climate index reconstructions.…”