<p dir="ltr">Robust and actionable information regarding how heat stress will change as climate warms is essential for informing impact assessments and heat mitigation and adaptation strategies. In meeting this demand, this dissertation has two mutually reinforcing goals: to improve heat stress prediction through a more comprehensive account of human heat stress, and to advance our understanding of the driving mechanisms of model-predicted heat stress changes.</p><p dir="ltr">As the initial step in achieving the first goal, we adopt the wet-bulb globe temperature (WBGT) as our preferred metric for heat stress. Then we (i) develop a fast, scalable Python implementation of the “gold standard” physics-based WBGT model, (ii) devise a straightforward, yet effective statistical bias-correction approach, and (iii) generate a global dataset of bias-corrected heat stress prediction at fine spatial and temporal resolutions based on a CMIP6 model ensemble. </p><p dir="ltr">To achieve our second goal of understanding the driving mechanisms of WBGT changes, we take advantage of the underlying physical relationship between WBGT and the simpler, wet-bulb temperature to gain insights into WBGT by first (i) investigating the soil moisture control of wet-bub temperature under present conditions and (ii) using CMIP6 results to understand future changes of wet-bulb temperature. Then, (iii) we develop a linear sensitivity framework that is used to disentangle WBGT changes into contributions from changes in temperature, humidity, wind, solar radiation and surface pressure. This disentanglement enables us to leverage existing theories and methods to understand the driving mechanism of WBGT changes.</p><p dir="ltr">Through this work we find several noteworthy conclusions, which is explained in depth in the rest of the dissertation, but we briefly summarize here. Wide-spread positive coupling between soil moisture and wet-bulb temperature are found over previously identified land-atmosphere coupling hotspots due to the effective control of soil moisture variations on surface energy partition and boundary layer dynamics. This implies that drying-induced amplified warming may be counteracted by relative humidity reductions, and a potential mismatch between relative hotspots of warming and intensifying heat stress. We confirm this hypothesis by showing distinctly different scaling patterns (with global warming) between dry-bulb temperature and WBGT based on a CMIP6 model ensemble. Regionally amplified warming in northern hemisphere mid-latitudes and the Amazon correspond to muted increases in WBGT. The central Sahel emerges as a strong local hotspot of WBGT scaling.</p><p dir="ltr">The sensitivity framework predicts close similarity between the scaling of black globe and natural wet-bulb temperature (two major components of WBGT) and that of dry- and wet-bulb temperature, if wind speed and solar radiation changes have a minor impact. This is confirmed to be the case in a CMIP6 model ensemble, with WBGT scaling primarily influenced by temperature and humidity changes. </p><p dir="ltr">Combining these results together holistically, we reach the following conclusions. Although the widely used and empirically well validated WBGT heat stress metric is a complex function of four environmental variables, as climate changes, the changes in WBGT predicted by climate models can be mostly understood in terms of changes in near-surface air temperature and humidity. Furthermore, the linear sensitivity framework and scaling analyses developed here allow us to partially attribute the WBGT scaling pattern to regional drying or wetting trends, and associated changes in surface energy balance and boundary layer dynamics. Thus, accurate prediction of WBGT changes is to first order largely a matter of getting those temperature and humidity correct and improvements to theories and models for those fields will directly translate to improvements in WBGT prediction as well. </p>
| Identifer | oai:union.ndltd.org:purdue.edu/oai:figshare.com:article/26343658 |
| Date | 22 July 2024 |
| Creators | Qinqin Kong (19185685) |
| Source Sets | Purdue University |
| Detected Language | English |
| Type | Text, Thesis |
| Rights | CC BY 4.0 |
| Relation | https://figshare.com/articles/thesis/Toward_better_prediction_and_deeper_understanding_of_human_heat_stress/26343658 |
Page generated in 0.002 seconds