Heat / Health Warning Systems: Development, Implementation, and Intervention Activities

There is an increasing awareness that heat is a major killer in many larger urban areas, and many municipalities have taken renewed interest in how they deal with oppressive heat. The implementation of sophisticated heat/health watch warning systems (HHWWS) is becoming more widespread, and these systems are becoming an important mechanism to save lives. One primary consideration in HHWWS development is the knowledge that response to heat varies through time and space. The more elaborate systems consider not only the intensity of heat, but the variability of the summer climate, which is closely related to urban population vulnerability. Thus, thresholds that induce negative health responses vary from one city to another, as well as over the season cycle at any one city. Warning system development involves a clear and consistent nomenclature (e.g. heat advisory, excessive heat warning), coordination between the agency issuing the warning and other stakeholders, public awareness of the system, targeted intervention procedures, and evaluation of effectiveness. This chapter describes these attributes in greater detail. Over the course of recent decades, significant heat waves (e.g., North America in 1980 and 1995, Europe in 1976 and 2003, East Asia in 2004) have resulted in significant loss of life and exposed considerable weaknesses in the infrastructure of heat wave mitigation plans and human adaptation to oppressive weather (Klinenberg 2002). In response to these heat events, many municipalities around the world have taken renewed interest in how they deal with the oppressive heat. In this chapter, we discuss the mechanisms for the development and implementation of heat/health Chapter 3 Heat / Health Warning Systems: Development, Implementation, and Intervention Activities Laurence S. Kalkstein, Scott C. Sheridan, and Adam J. Kalkstein K.L. Ebi et al. (eds.), Biometeorology for Adaptation to Climate Variability and Change, 33 © Springer Science + Business Media B.V. 2009 L.S. Kalkstein Center for Climatic Research, University of Delaware, Newark, DE 19716, USA S.C. Sheridan Department of Geography, Kent State University, Kent, OH 44242, USA A.J. Kalkstein Department of Geography, Arizona State University, Tempe, AZ 85287, USA 34 L.S. Kalkstein et al. watch warning systems (HHWWS), one of the key methods by which heat events are forecast and their effects are mitigated. We begin by describing the details by which thermal stress is evaluated in current HHWWS and the process by which warning criteria are determined. We then discuss the real-time development of HHWWS along with the “message delivery” to the public, heat mitigation strategies, and checking the effectiveness of HHWWS. 3.1 The Evaluation of Thermal Stress There is robust literature (Kovats and Koppe 2005) associating what is generally termed “oppressive” heat with some negative health consequence. However, the means by which “oppressive” is defined varies widely (Watts and Kalkstein 2004); accordingly, the HHWWS that have been developed across the world in recent years have utilized a diversity of methods. Each of these methods has their respective strengths and weaknesses. The utilization of a temperature threshold is perhaps the simplest of all methods. However, as outdoor temperature alone is significantly correlated with human mortality during excessive heat events (EHEs), temperature is considered by some to be a fairly reliable indicator. Moreover, the sole utilization of temperature has a further advantage in that it is the most commonly measured of all meteorological variables and thus is available for more locations. A number of nations, including Spain (Ministero de Sanidad y Consumo 2005), France (Pascal et al. 2006), the United Kingdom (UK Department of Health 2005), and Portugal (Paixao and Nogueira 2002), utilize maximum and/or minimum temperature thresholds in determining heat stress (Fig. 3.1). An extension of the temperature threshold is the utilization of an “apparent temperature” that takes into account humidity (and wind speed in certain cases) as well as temperature. Several different formulations of the apparent temperature exist, including the Heat Index (Steadman 1984), used widely in the USA and Australia, and the Humidex (Masterton and Richardson 1979), developed in Canada. These indices are especially useful in locations where summer absolute humidity levels can vary widely, hence their widespread use in North America. Thresholds can then be developed as with temperature; the 40.6°C threshold of heat index across much of the USA is a prime example (Watts and Kalkstein 2004). Another method of assessing meteorological conditions for application to the heat-health issue involves the classification of weather types, or air masses. The philosophy behind this “synoptic” methodology is to classify an entire suite of meteorological variables and thus holistically categorize the atmospheric situation at a given moment for a particular location or region (Yarnal 1993). This categorization when applied to heat is usually based upon surface weather variables, although upper atmospheric variables may also be incorporated. By categorizing the atmosphere into one of several internally homogeneous groups, other factors, such as solar radiation, wind speed, and cloud cover are inherently accounted for. For example, as a building’s “heat load”, as expressed by solar radiation income, has been associated with variability in human mortality, cloud cover or a some direct measure of solar radiation can be an important inclusion (Koppe and Jendritzky 2005). In synoptic approaches, discrete categories are created rather than a meteorological threshold 3 Heat / Health Warning Systems 35 value along the continuum of a continuous variable (e.g., temperature); the result is a determination of “oppressive” synoptic categories that are historically associated with negative health outcomes. The synoptic-based systems generally require meteorological data that is more comprehensive than the temperatureor apparent temperature-based models, including hourly surface data for a number of variables. A number of systems employ the synoptic methodology. Most notable are around 20 of the newer HHWWS across the USA (Sheridan and Kalkstein 2004), that incorporate the Spatial Synoptic Classification (SSC, Sheridan 2002). Several systems in Italy (Michelozzi and Nogueira 2004), Canada, South Korea, and China (Tan et al. 2003) also utilize the SSC. BOSTON 0 20 40 60 80 100 120 140 160 180 5PM Temp (°C) D ea th s DALLAS 0 10 20 30 40 50 60 70 80