Using Climate Model Projections

A framework of obtaining and applying climate model projections in engineering practice

  • A step-by-step process of determining, obtaining, and postprocessing climate model projections for engineering was summarized
  • A review on the types and methods of climate model projections used in the previous studies was provided
  • A case study on the accuracy, uncertainty, and recommendation of applying climate model projections was conducted

As an example result for this work, figure below presents a summarized overview of procedures needed to use climate model projections in engineering applications, which consist of three main steps: A) determining the requirements for climate variables from engineering applications (with several requirements needed); B) Obtaining climate projections (with several stages of decisions needed); and C) postprocessing the obtained climate projections if needed.

The details of this work can be found in Lai, Y., T. Lopez-Cantu, D.A. Dzombak, and C. Samaras. 2022. Framing the Use of Climate Model Projections in Infrastructure Engineering – Practices, Uncertainties, and Recommendations. Journal of Infrastructure Systems. 28 (3), 04022020. https://doi.org/10.1061/(ASCE)IS.1943-555X.0000685, which was also featured in Editor's Choice.

Glossary of terms, abbreviations, and acronyms:

  • GCM: global climate model or general circulation model, which is a type of large-scale, physics-based climate simulation models. Applied with the trajectories of forcings (i.e., the warming or cooling effects on climate) such as from greenhouse gas emissions, these GCMs can be and were used to provide long-term future climate projections. Different agencies and institutes have their own GCMs, which combine to a total of dozens of GCMs available.
  • CMIP: Coupled Model Intercomparison Project from the World Climate Research Programme and can be considered as repositories of GCM results. The phase 6 of CMIP (CMIP6) was released around 2020, while the previous phases such as CMIP3 and CMIP5 have been commonly used in engineering applications.
  • Downscaling: referred to the processes of translating large-scale, course-resolution GCM projections to finer resolution (can be both temporally and spatially; generally referred to spatial downscaling in engineering applications). Downscaling is often used to provide regional and location-specific climate model projection results. Downscaling also includes two main types of methods, using statistical methods (identifying and applying statistical relationships between historical observations and GCM simulations) and using dynamical methods (using regional climate models to run the GCM simulation results in finer resolutions).
  • NA-CORDEX: the North American component of the Coordinated Regional Downscaling Experiment (CORDEX). NA-CORDEX therefore is a repository of downscaled GCM results from using dynamical downscaling methods, i.e., using regional climate models.
  • BCCA: bias corrected constructed analogs approach, a type of statistical downscaling method.
  • LOCA: localized corrected constructed analogs approach, a type of statistical downscaling method. Note that the LOCA results were recommended for hydrological designs by a technical report conducted by National Cooperative Highway Research Program.
  • MACA: multivariate adaptive constructed analogs approach, a type of statistical downscaling method.
  • RCP: representative concentration pathways developed for CMIP5 GCM experiments and represent different trajectories of anthropogenic forcings such as the forcing from future greenhouse gas emissions. The numbers following RCP such as 4.5 and 8.5 generally refer to the end-of-century total forcing targets, i.e., 4.5 and 8.5 W/m2. In the recent CMIP6, shared socioeconomic pathways (SSPs) were used such as SSP2-4.5 and SSP5-8.5, which in addition to the forcing numbers, represent the consideration of the pathways of socioeconomic development to achieve the forcing trajectories.

Some useful related resources: