How to Read: The ERA5 global reanalysis [Reading Guide]

This guide will help you efficiently read and understand "The ERA5 global reanalysis" paper, published in 2020 in the Quarterly Journal of the Royal Meteorological Society, using a three-pass approach.

First pass (5–10 min)

The goal of the first pass is to get a high-level overview of the paper's purpose, main contributions, and general scope.

• What to skim:
  • Title: "The ERA5 global reanalysis" immediately tells you the subject: a new global reanalysis dataset.
  • Abstract: Read this carefully. It summarizes ERA5's key features (e.g., higher resolution, hourly output, uncertainty estimates) and its improvements over its predecessor, ERA-Interim. Note the core findings like skill gain and improved fit to observations.
  • Section 1: Introduction: Understand the motivation for ERA5, the context of reanalysis, and the paper's structure. It sets the stage for why this new dataset is important.
  • Section 4: Summary and Outlook: Read the conclusion to grasp the main findings and the authors' perspective on the dataset's significance and future.
• What to notice:
  • Contributions: The paper introduces and evaluates ERA5, a new, significantly improved global reanalysis dataset. Key improvements include enhanced spatial (31 km vs. 80 km) and temporal (hourly) resolution, and the inclusion of uncertainty estimates.
  • Problem: The need for a more accurate, higher-resolution, and comprehensive global atmospheric, land surface, and ocean wave record than previous reanalyses like ERA-Interim.
  • Setting: The paper describes a global reanalysis, meaning it covers the entire Earth's atmosphere, land surface, and ocean waves, spanning from 1950 onwards (though the paper focuses on 1979 onwards).

Second pass (up to 1 hour)

Using SpatialRead in the Second Pass

• Use the Screenshot tool to capture figures, tables, diagrams, and formulae from the PDF and use AI for simple explanations.
• Ask: "Explain Figure X at a high level", "Compare Table 2 vs 3", or "What does the y-axis represent here?"
• Download SpatialRead →

In this pass, you'll delve deeper into the paper's structure, focusing on key sections, figures, and tables that highlight the system's setup and primary evaluation results.

• What sections/figures to focus on:
  • Section 2: ERA5 System Description: This section details how ERA5 was constructed.
    • Section 2.1 General setup: Understand the core components like the Integrated Forecasting System (IFS) Cy41r2, the horizontal resolution, and the number of vertical levels.
    • Section 2.2 Observing system: Pay attention to the evolution of observing systems over time, as depicted in Figure 1, which illustrates the increasing number of observations assimilated.
    • Table 1: This table provides a concise summary of ERA5's key characteristics compared to ERA-Interim, making it easy to see the advancements.
  • Section 3: Evaluation of ERA5: This is where the performance of ERA5 is assessed.
    • Section 3.1 Re-forecast skill: Figure 2 graphically demonstrates the gain in forecast skill of ERA5 over ERA-Interim, which is a crucial metric.
    • Section 3.2 Fit to observations:
      • Figures 3 and 4 show how well ERA5 fits independent radiosonde and PILOT observations for temperature, wind, and humidity. Note the improved fit in the troposphere but not necessarily the stratosphere.
      • Figure 5 illustrates the much-improved fit for ocean wave height compared to buoy data.
    • Section 3.3 Global mean characteristics:
      • Figures 8 and 9 are important for understanding precipitation performance, showing global-mean correlation with GPCP data and precipitation bias.
    • Section 3.5 Case studies: Briefly review Figures 13 and 14 to see how the enhanced resolution allows for better representation of specific weather events like hurricanes and atmospheric rivers.
• Key terms/questions to note:
  • Key terms: Reanalysis, 4D-Var, Integrated Forecasting System (IFS), data assimilation, ensemble data assimilation (EDA), bias correction, GPCP (Global Precipitation Climatology Project).
  • Questions: What specific advancements in the IFS model or data assimilation scheme contribute to ERA5's improvements? How robust are the uncertainty estimates? Are there any known limitations or areas where ERA5 performs less well?

Third pass (1–5 hours)

Using SpatialRead in the Third Pass

• Highlight complex keywords, dense paragraphs, equations, or proofs and ask for a plain-English explanation that preserves notation.
• Use AI to explain the concepts in a way that is easy to understand.
• Ask for alternative phrasings or analogies to confirm understanding without changing technical meaning.
• Connect concepts with each other using AI to build a comprehensive understanding of the paper.
• Download SpatialRead →

This is your deep reading pass. You'll critically analyze the methodology, scrutinize the results, understand the limitations, and consider potential applications or follow-up work.

• Deep reading guidance:
  • Methods (Section 2): Read Sections 2.3 (Data assimilation) and 2.4 (Model) thoroughly. Understand the principles of 4D-Var and how bias correction is applied. Pay attention to the specific physics parameterizations used in IFS Cy41r2, as these directly influence the model's output. Consider how the evolution of the observing system (Figure 1) impacts the reanalysis quality over time.
  • Results (Section 3): Go through each subsection of Section 3 in detail. For every figure, understand what it's showing, the metrics used, and the implications of the results. For example, in Section 3.2.1, note the specific altitudes where improvements are seen or not seen. In Section 3.3, analyze the global mean characteristics for different variables (temperature, wind, precipitation, etc.) and compare them to other reanalyses or observational datasets where mentioned. Critically evaluate the uncertainty estimates presented in Section 3.4 and Figure 12.
  • Limitations: While the paper highlights improvements, pay attention to any caveats or areas where ERA5 does not perform as well (e.g., the stratosphere fit in Section 3.2.1). These are crucial for understanding the dataset's appropriate use.
• Replication pointers:
  • While you cannot replicate the generation of the ERA5 dataset itself, you can replicate or extend its evaluation. The paper provides a benchmark for how ERA5 performs against various observational datasets (radiosondes, buoys, GPCP).
  • You could obtain ERA5 data (as described in Section 4) and perform your own comparisons with independent observations or other reanalysis products for specific regions, time periods, or variables not extensively covered in this paper.
  • The detailed description of the system setup in Section 2 provides a foundation for understanding the characteristics and potential biases of the data, which is essential for any research using ERA5.