Towards a compound event-oriented climate model evaluation: A decomposition of the underlying biases in multivariate fire and heat stress hazards

Abstract

Abstract. Climate models' outputs are affected by biases that need to be detected and adjusted to model climate impacts. Many climate hazards and climate-related impacts are associated with the interaction between multiple drivers, i.e. by compound events. So far climate model biases are typically assessed based on the hazard of interest, and it is unclear how much a potential bias in the dependence of the hazard drivers contributes to the overall bias and how the biases in the drivers interact. Here, based on copula theory, we develop a multivariate bias assessment framework, which allows for disentangling the biases in hazard indicators in terms of the underlying univariate drivers and their statistical dependence. Based on this framework, we dissect biases in fire and heat stress hazards in a suite of global climate models by considering two simplified hazard indicators, the wet-bulb globe temperature (WBGT) and the Chandler Burning Index (CBI). Both indices solely rely on temperature and relative humidity. The spatial pattern of the hazards indicators is well represented by climate models. However, substantial biases exist in the representation of extreme conditions, especially in the CBI (spatial average of absolute bias: 21 °C) due to the biases driven by relative humidity (20 °C). Biases in WBGT (1.1 °C) are small compared to the biases driven by temperature (1.9 °C) and relative humidity (1.4 °C), as the two biases compensate each other. In many regions, also biases related to the statistical dependence (0.85 °C) are important for WBGT, which indicates that well-designed physically-based multivariate bias adjustment should be considered for hazards and impacts that depend on multiple drivers. The proposed compound event-oriented evaluation of climate model biases is easily applicable to other hazard types. Furthermore, it can contribute to improved present and future risk assessments through increasing our understanding of the biases’ sources in the simulation of climate impacts.

Larger Spatial Footprint of Wintertime Total Precipitation Extremes in a Warmer Climate

“…Following Bevacqua et al. (2019) (and, e.g., Manning et al., 2019; Villalobos‐Herrera et al., 2020), this data set is defined as , where P pres i and are the wintertime precipitation in the present and its empirical CDF, respectively; and P fut i and are the same, but for the future. Given that has a standard uniform distribution, the marginal pdfs of the variables in the data set are as in the future; the copula is a function of the variables and is, therefore, as in the present.…”

“…We estimate (1) as 100 × (SS magn − SS pres )/SS pres , where SS pres is the spatial scale extreme in the present and SS magn is the spatial scale extreme in a data set that assumes changes in precipitation magnitude, but not in spatial dependence (pairwise Spearman's correlations and tail dependencies [Bevacqua et al, 2017]). Following Bevacqua et al (2019) (and, e.g., Manning et al, 2019Villalobos-Herrera et al, 2020), this data set is defined as F F P F F P…”

Growing Threats From Unprecedented Sequential Flood‐Hot Extremes Across China

“…However, the model evaluation and bias correction for simulating unprecedented extremes, with reference to realistic observations, are inherently difficult (Lehner et al., 2018). The challenge is especially evident for multivariate events, because models not only have to reasonably reproduce univariate extremes statistics, but also need to capture their spatio‐temporal causal/statistical linkage as well as the underlying processes (Villalobos‐Herrera et al., 2020; Zscheischler et al., 2019). Whether the linkage exists is unknown, let alone the mechanism understanding; meanwhile univariate components of sequential extremes may cluster just by chance as an expression of random variability.…”

Amplified Risk of Compound Heat Stress-Dry Spells in Urban India

“…The multi-hazard framework developed in this study would be helpful to calculate compound climate risks in a warming climate and further developing an impact database such as for mapping heat-health risks in urban areas, crop insurance losses, or assessing potentials for renewable power generation. Further, results from this study would provide valuable insights to assess biases in climate model simulations and to evaluate models' credibility in precisely estimating the joint behavior of compound extremes under deep uncertainties related to climate variability and change (Brown et al 2020;Villalobos-Herrera et al 2020). The derived insights would ensure adaptive policies to maximize urban resilience in the face of climate uncertainty.…”

“…To the best of our knowledge, using high-quality gauge-based observational records, this study is the first to quantify the multivariate risk of concurrent climate extremes considering more than two climate variables covering six different hydrometeorological regions, especially over densely populated and highly invested urban locations of India. The strength of gauge-based observational records as compared to previous assessments (Hao et al, 2018;Ridder et al, 2020;Sharma and Mujumdar, 2017;De Luca et al, 2020;(Villalobos-Herrera et al 2020) is its ability to preserve local variabilities, although station-based records have certain limitations such as uncertainties owing to gaps in observations and sparse spatial coverage of station networks. However, in areas of complex terrain with large spatial heterogeneity, extremes are often underestimated in the gridded dataset or satellite-based products, especially for the intensity and frequency of extremes (Kandlikar et al 2005;King et al 2013;Timmermans et al 2019;Raymond et al 2020).…”