Abrupt changes in environmental temperature, wind and humidity can lead to great threats to human life safety. The Gansu marathon disaster of China highlights the importance of early warning of hypothermia from extremely low appar-ent temperature (AT). Here a deep convolutional neural network model together with a statistical downscaling framework is developed to forecast environmental factors for 1 to 12 h in advance to evaluate the effectiveness of deep learning for AT prediction at 1 km resolution. The experiments use data for temperature, wind speed and relative humidity in ERA-5 and the results show that the developed deep learning model can predict the upcoming extreme low temperature AT event in the Gansu marathon region several hours in advance with better accuracy than climatological and persistence forecast-ing methods. The hypothermia time estimated by the deep learning method with a heat loss model agrees well with the observed estimation at 3-hour lead. Therefore, the developed deep learning forecasting method is effective for short-term AT prediction and hypothermia warnings at local areas.

Apparent Temperature Forecasting; Deep Learning; Neural Network; Spatiotemporal Forecasting; AI Applications

1. Gasparrini A, Guo Y, Hashizume M, et al. Mortality risk attributable to high and low ambient temperature: A multicountry observational study. The Lancet 2015; 386(9991): 369–375. doi: 10.1016/S0140-6736(14)62114-0.

2. Zhao Q, Guo Y, Ye T, et al. Global, regional, and national burden of mortality associated with non-optimal ambient temperatures from 2000 to 2019: A three-stage modelling study. The Lancet Planetary Health 2021; 5(7): e415–e425. doi: 10.1016/S2542-5196(21)00081-4.

3. Cannistraro G, Cannistraro M. Hypothermia risk, monitoring and environment control in operating rooms. International Journal of Heat & Technology 2016; 34(2): 165–171. doi: 10.18280/ijht.340202.

4. Bieler D. 21 die in Chinese ultramarathon suddenly struck by plunging temperatures, wind and hail [Report]. BBC News; 2021.

5. Taskforce G.P.J.I. Investigation report on public safety responsibility incidents of Baiyin Jingtai’s “5.22” Yellow River Stone Forest 100 km cross-country race [Report]. People’s Government of Gansu Province; 2021.

6. Chiu CH, Vagi SJ, Wolkin AF, et al. Evaluation of the National Weather Service extreme cold warning experiment in North Dakota. Weather, Climate, and Society 2014; 6: 22–31. doi: 10.1175/WCAS-D-13-00023.1.

7. Steadman RG. A universal scale of apparent temperature. Journal of Applied Meteorology and Climatology 1984; 23(12): 1674–1687. doi: 10.1175/1520-0450(1984)023<1674:AUSOAT>2.0.CO;2.

8. Brown DJ, Brugger H, Boyd J, Paal P. Accidental hypothermia. New England Journal of Medicine 2012; 367: 1930–1938. doi: 10.1056/NEJMra1114208.

9. Salman AG, Kanigoro B, Heryadi Y. Weather forecasting using deep learning techniques. In: 2015 International Conference on Advanced Computer Science and Information Systems (ICACSIS); 2015 Oct 10–11; Depok. New York: IEEE; 2016. p. 281–285.

10. Shi X, Gao Z, Lausen L, et al. Deep learning for precipitation nowcasting: A benchmark and a new model. Advances in Neural Information Processing Systems 2017; 30: 5622–5632. doi: 10.48550/arXiv.1706.03458.

11. Xu L, Chen N, Chen Z, et al. Spatiotemporal forecasting in earth system science: Methods, uncertainties, predictability and future directions. Earth-Science Reviews 2021; 222: 103828. doi: 10.1016/j.earscirev.2021.103828.

12. LeCun Y, Bengio Y, Hinton G. Deep learning. Nature 2015; 521: 436–444. doi: 10.1038/nature14539.

13. Reichstein M, Camps-Valls G, Stevens B, et al. Deep learning and process understanding for data-driven Earth system science. Nature 2019; 566: 195–204. doi: 10.1038/s41586-019-0912-1.

14. Xiao C, Chen N, Hu C, et al. A spatiotemporal deep learning model for sea surface temperature field prediction using time-series satellite data. Environmental Modelling & Software 2019; 120: 104502. doi: 10.1016/j.envsoft.2019.104502.

15. Zheng G, Li X, Zhang RH, Liu B. Purely satellite data–driven deep learning forecast of complicated tropical instability waves. Science Advances 2020; 6(29): eaba1482. doi: 10.1126/sciadv.aba1482.

16. Ham YG, Kim JH, Luo JJ. Deep learning for multi-year ENSO forecasts. Nature 2019; 573: 568–572. doi: 10.1038/s41586-019-0912-1.

17. Xu L, Chen N, Zhang X, Chen Z. A data-driven multi-model ensemble for deterministic and probabilistic precipitation forecasting at seasonal scale. Climate Dynamics 2020; 54: 3355–3374. doi: 10.1007/s00382-020-05173-x.

18. Hagras H. Toward human-understandable, explainable AI. Computer 2018; 51(9): 28–36. doi: 10.1109/MC.2018.3620965.

19. Zhang Q, Zhu S. Visual interpretability for deep learning: A survey. Frontiers of Information Technology & Electronic Engineering 2018; 19(1): 27–39. doi: 10.1631/FITEE.1700808.

20. Bauer P, Thorpe A, Brunet G. The quiet revolution of numerical weather prediction. Nature 2015; 525: 47–55. doi: 10.1038/nature14956.

21. Slingo J, Palmer T. Uncertainty in weather and climate prediction. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 2011; 369: 4751–4767. doi: 10.1098/rsta.2011.0161.

22. Wahl S. Uncertainty in mesoscale numerical weather prediction: Probabilistic forecasting of precipitation [PhD thesis]. Bonn: Rheinische Friedrich-Wilhelms-Universität Bonn; 2015.

23. Ayzel G, Scheffer T, Heistermann M. RainNet v1.0: A convolutional neural network for radar-based precipitation nowcasting. Geoscientific Model Development 2020; 13(6): 2631–2644. doi: 10.5194/gmd-13-2631-2020.

24. Ravuri S, Lenc K, Willson M, et al. Skilful precipitation nowcasting using deep generative models of radar. Nature 2021; 597: 672–677. doi: 10.1038/s41586-021-03854-z.

25. Frame JM, Kratzert F, Klotz D, et al. Deep learning rainfall–runoff predictions of extreme events. Hydrology and Earth System Sciences 2022; 26: 3377–3392. doi: 10.5194/hess-26-3377-2022.

26. Schltz MG, Betancourt C, Gong B, et al. Can deep learning beat numerical weather prediction? Philosophical Transactions of the Royal Society A 2021; 379(2194): 20200097. doi: 10.1098/rsta.2020.0097.

27. USGS 30 ARC-second global elevation data, GTOPO30 [Report]. USGS Earth Resources Observation and Science (EROS) Center; 1996. doi: 10.5065/A1Z4-EE71.

28. An D, Du Y, Berndtsson R, et al. Evidence of climate shift for temperature and precipitation extremes across Gansu Province in China. Theoretical and Applied Climatology 2020; 139(5): 1137–1149. doi: 10.1007/s00704-019-03041-1.

29. Li C, Wang R. Recent changes of precipitation in Gansu, Northwest China: An index-based analysis. Theoretical and Applied Climatology 2017; 129: 397–412. doi: 10.1007/s00704-016-1783-0.

30. Hersbach H, Bell B, Berrisford P, et al. The ERA5 global reanalysis. Quarterly Journal of the Royal Meteorological Society 2020; 146(730): 1999–2049. doi: 10.1002/qj.3803.

31. Kingma DP, Ba J. Adam: A method for stochastic optimization. ArXiv 2014; 1412.6980. doi: 10.48550/arXiv.1412.6980.

32. Murphy AH. Climatology, persistence, and their linear combination as standards of reference in skill scores. Weather and Forecasting 1992; 7(4): 692–698. doi: 10.1175/1520-0434(1992)007<0692:CPATLC>2.0.CO;2.

33. Wilks DS. Statistical methods in the atmospheric sciences. Cambridge: Academic press; 2011.

34. Schoof JT, Pryor SC. Downscaling temperature and precipitation: A comparison of regression-based methods and artificial neural networks. International Journal of Climatology 2001; 21(7): 773–790. doi: 10.1002/joc.655.

35. Skamarock WC, Klemp JB, Dudhia J, et al. A description of the advanced research WRF version 2 [Report]. National Center for Atmospheric Research Boulder; 2005. doi: 10.5065/D68S4MVH.

36. Kain JS. The Kain–Fritsch convective parameterization: An update. Journal of Applied Meteorology 2004; 43(1): 170–181. doi: 10.1175/1520-0450(2004)043<0170:TKCPAU>2.0.CO;2.

37. Chen S, Sun W. A one-dimensional time dependent cloud model. Journal of the Meteorological Society of Japan. Ser. II 2002; 80(1): 99–118. doi: 10.2151/jmsj.80.99.

38. NCEP FNL operational model global tropospheric analyses, continuing from July 1999 [Report]. Research Data Archive at the National Center for Atmospheric Research, Computational and Information Systems Laboratory; 2000. doi: 10.5065/D6M043C6.

39. Huth R. Statistical downscaling of daily temperature in central Europe. Journal of Climate 2002; 15(13): 1731–1742. doi: 10.1175/1520-0442(2002)015<1731:SDODTI>2.0.CO;2.

40. Cena K, Monteith JL. Transfer processes in animal coats. II. Conduction and convection. Proceedings of the Royal Society of London Series B Biological Sciences 1975; 188(1093): 395–411. doi: 10.1098/rspb.1975.0027.

41. Hanania NA, Zimmerman JL. Accidental hypothermia. Critical Care Clinics 1999; 15(2): 235–249. doi: 10.1016/S0749-0704(05)70052-X.

42. Kang J, Rong Y. Modeling and simulation of load heating in heat treatment furnaces. Journal of Materials Processing Technology 2006; 174(1–3): 109–114. doi: 10.1016/j.jmatprotec.2005.03.037.

43. Bernard V, Staffa E, Mornstein V, Bourek A. Infrared camera assessment of skin surface temperature—Effect of emissivity. Physica Medica 2013; 29(6): 583–591. doi: 10.1016/j.ejmp.2012.09.003.

44. Osczevski RJ. The basis of wind chill. Arctic 1995; 48(4): 372–382. doi: 10.14430/arctic1262.

45. Woodson WE, Tillman B, Tillman P. Human factors design handbook: Information and guidelines for the design of systems, facilities, equipment, and products for human use. New York: McGraw-Hill; 1992. p. 861.

46. American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. Thermal comfort. In: 2005 ASHRAE handbook: Fundamentals. American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.; 2005.

47. Shi R, Zhang J, Fang B, et al. Runners’ metabolomic changes following marathon. Nutrition & Metabolism 2020; 17: 19. doi: 10.1186/s12986-020-00436-0.

48. Frankenfield D, Roth-Yousey L, Compher C, Group EAW. Comparison of predictive equations for resting metabolic rate in healthy nonobese and obese adults: A systematic review. Journal of the American Dietetic Association 2005; 105(5): 775–789. doi: 10.1016/j.jada.2005.02.005.

49. Shi X, Chen Z, Wang H, et al. Convolutional LSTM network: A machine learning approach for precipitation nowcasting. In: Proceedings of the 28th International Conference on Neural Information Processing Systems; 2015 Dec 7–12; Montreal, Canada. Cambridge: MIT Press; 2015. p. 802–810.

50. Xu L, Chen N, Zhang X, Chen Z. An evaluation of statistical, NMME and hybrid models for drought prediction in China. Journal of Hydrology 2018; 566: 235–249. doi: 10.1016/j.jhydrol.2018.09.020.

51. Ghahramani Z. Probabilistic machine learning and artificial intelligence. Nature 2015; 521: 452–459. doi: 10.1038/nature14541.

52. Gal Y, Ghahramani Z. Dropout as a Bayesian approximation: Representing model uncertainty in deep learning. In: 33rd International Conference on Machine Learning: PMLR; 2016 Jun 19–24; New York. Cambridge: JMLR; 2016. p. 1050–1059.

53. Fente DN, Singh DK. Weather forecasting using artificial neural network. In: 2018 Second International Conference on Inventive Communication and Computational Technologies (ICICCT); 2018 Apr 20–21; Coimbatore. New York: IEEE; 2018. p. 1757–1761.

54. Kumar R. Decision tree for the weather forecasting. International Journal of Computer Applications 2013; 76(2): 31–34. doi: 10.5120/13220-0620.

55. Vitart F, Ardilouze C, Bonet A, et al. The subseasonal to seasonal (S2S) prediction project database. Bulletin of the American Meteorological Society 2017; 98(1): 163–173. doi: 10.1175/BAMS-D-16-0017.1.

56. Hewage P, Trovati M, Pereira E, Behera A. Deep learning-based effective fine-grained weather forecasting model. Pattern Analysis and Applications 2021; 24(4): 343–366. doi: 10.1007/s10044-020-00898-1.

57. Hossain M, Rekabdar B, Louis SJ, Dascalu S. Forecasting the weather of Nevada: A deep learning approach. In: 2015 International Joint Conference on Neural Networks (IJCNN); 2015 Jul 12–17; Killarney. New York: IEEE; 2015. p. 1–6.

58. Chakraborty S, Tomsett R, Raghavendra R, et al. Interpretability of deep learning models: A survey of results. In: 2017 IEEE SmartWorld, Ubiquitous Intelligence & Computing, Advanced & Trusted Computed, Scalable Computing & Communications, Cloud & Big Data Computing, Internet of People and Smart City Innovation; 2017 Aug 4–8; San Francisco. New York: IEEE; 2017. p. 1–6.

59. Wedi NP, Polichtchouk I, Dueben P, et al. A baseline for global weather and climate simulations at 1 km resolution. Journal of Advances in Modeling Earth Systems 2020; 12(11): e2020M–e2192M. doi: 10.1029/2020MS002192.