Commonly used indices disagree about the relative effect of moisture on heat stress - code supplement

Commonly used indices disagree about the relative effect of moisture on heat stress

This is a code archive supporting the article "Commonly used indices completely disagree about the relative effect of moisture on heat stress". The code contained in this archive is sufficient to reproduce the plots contained in the manuscript. The code runs in python.

Abstract

Irrigation and urban greening can mitigate extreme temperatures and reduce adverse health impacts from heat. However, some recent studies suggest these interventions could actually exacerbate heat stress by increasing humidity. These studies use different heat stress indices (HSIs), hindering intercomparisons of the relative roles of temperature and humidity.

Our novel method uses calculus of variations to compare the sensitivity of HSIs to temperature and humidity, independent of HSI units. We explain the properties of different HSIs and identify conditions under which they disagree. We highlight recent studies where the use of different HSIs could have led to opposite conclusions.

Our findings have significant implications for the evaluation of irrigation and urban greening as adaptive responses to overheating, and climate adaptation measures in general. We urge researchers to be critical in their choice of HSIs, especially in relation to health outcomes; our method provides a useful tool for making informed comparisons.

Files in this archive

• README.md
• basic_psychrometric_chart.py
  • Code for defining psychrometric charts. These are re-usable utility functions. To actually produce the charts use psychrometric_charts.py.
• environment_dump.yml
  • My environment exactly as-was at the point of archiving.
• environment_minimal.yml
  • The minimum packages required to run the code.
• heat_index.py
  • Codes for calculating heat index and some other heat stress indices.
• mtec_fields.py
  • Codes for calculating 2D field of marginal temperature equivalent changes.
• psychrometric_charts.py
  • Codes for producing the psychrometric charts appearing in the paper.
• utci.py
  • Code for calculating the UTCI polynomial.
  • This was based the code found in https://github.com/ecmwf-projects/thermofeel/blob/master/thermofeel/thermofeel.py, which has been modified under the Apache 2.0 license.
  • Additionally we apply the correct limits of validity to the calculation, as described in https://doi.org/10.7163/GPol.2013.1
• wouters.py
  • Code for producing comparisons of Wouters et al's results between different heat stress indices.
• wouters_table1.xlsx
  • Table 1 from Wouters et al, enabling wouters.py.

One piece of nessary code is archived separately due to licensing requirements:

• WetBulb_jit.py
  • Is archived at https://doi.org/10.5281/zenodo.7777698
  • This contains code for calculating wet bulb temperature using the Davies-Jones method.
  • It is based on the code https://github.com/jrbuzan/HumanIndexMod_2020 and is subject to the GNU GENERAL PUBLIC LICENSE.

How to run

These python scripts were run cell-by-cell in ipython, where the # %% indicates a cell break. Usually running as a normal script will produce the same results, but the plots may not come out the same, as there is an implicit plt.show() at the end of each cell.

Running the follow scripts is sufficient to reproduce all of the figures in the manuscript, the order is not important:

• mtec_fields.py
• psychrometric_charts.py
• wouters.py