Micro-meteorology controlIntegrated technology of harmonic prediction and active monitoring of micro-meteorology for future autonomous society

Planned Research

A03 Impact Modeling Group

Academic Background and the Crux of the “Question”

Heat stress is a big social issue. Especially in urban areas, people experience a triad of adverse effects the heat island effect, global warming, and an aging society. The death toll related to heat stroke, mostly of the elderly, reaches approx. 900 in Japan every year (the average figure of 2007-2011 [Fujibe, Weather (2013)]). This figure is far bigger when compared to the annual average of 60 people dead and missing due to typhoons and torrential rain, although we can never belittle the damage caused by heavy rain, since more than 220 people died due to heavy rain in July 2018. The death toll from heat stroke is similar to the annual 1,000 (average of 1995 – 2009) dead and missing due to earthquakes and tsunami or tidal wave. To create a safe and secure society, it is a big challenge to reduce this heat stress and to reduce and control the risk of heat stroke.

Heat stress causes enormous damage not only to the people but also to the economy. For instance, it is feared that the increase of heat stress due to global warming would result in greater loss of workable hours during the day. According to the American Climate Prospectus (http://rhg.com/reports/climate-prospectus) in 2014, the annual loss of labor productivity related to temperature rise is estimated to be approximately 0.2% of the estimated GDP, 38 trillion US dollars (equivalent to approx. 4,100 trillion Japanese yen), which amounts to 76 billion US dollars (equivalent to approx. 8.4 trillion Japanese yen). This figure stands for the estimate of damage caused by global warming only. In urban areas where temperature increases are also due to heat island and overpopulation effects, the damage severity may increase even more. Thus, ultrahigh-definition micro-meteorological prediction information on the city block scale is extremely useful in reducing the damage caused by heat stress. Our research group input the 5 m mesh meteorological data (atmospheric temperature, humidity, amount of solar radiation, and wind velocity) calculated by the heat environment prediction simulation of A01 Harmonic Prediction Group and performed a three-dimensional human thermal response model simulation. Thus, our group succeeded in developing a risk assessment technology that takes into account the difference between the pedestrians walking on the sunny side and those walking on the shady side of the same street (July 23, 2019 Press Release: Success in developing the technology for detailed heat stroke risk assessment in urban areas – for safer and securer behavior selection. “http://www.jamstec.go.jp/j/about/press_release/20190723_2/” Kamiya et al. 2019).

Detailed simulation using human thermal models is currently performed by workstations and large supercomputers, which takes several hours to one day. It is effective for taking proactive measures or a review following a serious incident but cannot be linked to real-time services or applied to the crowd. To put it to practical use, an evaluation that normally takes three hours by supercomputers needs to be performed in milliseconds. To attain such superhigh speed performance, the research group clarifies the human time responses to micro-meteorology and develops a contracted model (low-dimensional model) capable of such evaluation.

ACGIH (American Conference of Governmental Industrial Hygienists) standards use core body temperature as the direct index of heat stress. For the core temperature, time series change is more important than instantaneous heat stress. In the meantime, it is infeasible to monitor core body temperature directly. Our group possesses an internationally recognized human thermo-physiological model. This model is capable of precisely assessing the human thermo-physiological responses to changes in external heat environment and numerically predicting the core body temperature. Our researchers are also versed in machine learning technology, and hence possesses the fundamental technologies for developing a contracted model capable of predicting human thermal responses at ultrahigh speed. Our research group is optimal and essential for realizing a real-time service to reduce heat stress of individual pedestrians.

Scientific Significance - Objective, Academic Uniqueness and Creativity

Conventionally, heat stroke risk has been evaluated at a single point in time and space, i.e., who fell down when and where. However, in reality, people fall down due to heat stroke while they are moving in the constantly changing heat environment. In this case, the history of heat stress that a person has undergone plays an important role. This research assesses heat stress, which is difficult to quantify technically or using ethical methods, by numerical simulation. Thus, our group predicts the risk of heat stroke while taking into account the constantly changing heat environment (micro-meteorology) and develops the basic model technologies to realize a future social service for risk reduction.