Evaluation of hot temperature extremes and heat waves in the Mississippi River Basin

Highlights

• Changes of heat wave were analyzed in the Mississippi River Basin (MRB).

• More frequent and longer heat waves were observed in western and north-western MRB.

• The trends for heat wave frequency were mostly upward after the change-points.

• The trends for heat wave frequency were mostly downward before the change-points.

Abstract

The Intergovernmental Panel on Climate Change (IPCC) reports that the changes in extreme events are expected to be detected earlier than changes in climate averages. Hot temperature extremes in the United States cause important economical, societal, and environmental impacts. Thus, the climatology of hot extremes and heat waves (HWs) merits further examination. Spatiotemporal trends of five extreme temperature indicators were investigated for the period 1948–2017 for the Mississippi River Basin (MRB), which covers approximately 41% of the Contiguous United States. A hot extreme was identified using the 90th percentile threshold and a HW was defined as two or more consecutive hot temperature extreme days. Results suggest that the western, north-western, southern, and northern parts of the MRB are regions with a growing risk of extreme temperatures with more frequent and longer hot events. Change in the eastern part of the MRB suggests a smaller risk with a general downward trend in HWs.

Change-point analysis indicates that these climatic data exhibit nonstationarity. A significant increase happened starting in 1994 in the percentage of area with a HW longer than 10 consecutive days. Frequency and length of HWs were greater during El Niño years. However, this finding is not statistically significant. An increase in HW duration and frequency can impact water availability, human and animal health, agriculture and the economy. Results from this study assist in finding hotspots where changing extreme conditions have happened and where society may need to make adjustments related to water use, human outdoor activities and agricultural practices.

Introduction

A large number of record-breaking hot temperature extremes and heat waves (HW) have been reported in recent years all around the world (Coumou and Rahmstorf, 2012) suggesting that this increased frequency is symptomatic of ongoing global climate change. HWs are among the 10 deadliest global natural disasters (Guha-Sapir et al., 2012). Lack of cooling at night and multi-day atmospheric heat accumulation (Miralles et al., 2014) have been linked to human and livestock mortality (Anderson and Bell, 2011; CDC, 1996; Fouillet et al., 2008; Nienaber and Hahn, 2007). The connections between HWs and water scarcity (Seneviratne et al., 2006), human mortality (Basu and Samet, 2002; Patz et al., 2005), livestock health (Hahn, 1999; Thornton et al., 2009), and crop loss (Deryng et al., 2014; Lobell et al., 2012; Lobell et al., 2013) have raised serious concerns about the future impacts of climate change (Kovats and Hajat, 2008). The Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report indicated an increase in warm extreme indices for a majority of global land areas (Hartmann et al., 2013). Since 1950, this increase is significant and consistent with global climate change (Donat and Alexander, 2012; Hartmann et al., 2013).

In 2003, 2010, and 2015, European areas experienced devastating HWs that broke long-standing temperature records and caused extensive crop loss, water scarcity, wildfires and loss of life (Barriopedro et al., 2011; Garcia-Herrera et al., 2010; Krzyżewska et al., 2019; Miralles et al., 2014). Atmospheric blocking with the presence of persistent high-pressure areas (Meehl and Tebaldi, 2004) has been identified as a key synoptic weather pattern causing temperature escalation to recorded extremes. Maximum summer temperatures that were observed during 2010 in western Russia were exceptional based on the observational record that dates back to 1871 (Barriopedro et al., 2011). This HW lasted from June into August and the heat and dry conditions set the stage for extensive wildfires (Trenberth and Fasullo, 2012). Barriopedro et al. (2011) suggest that the probability of summer with a mega-HW in Europe will increase by a factor of 5–10 in the next 40 years. In China, a warming trend was found for extremely hot days and nights (Shi et al., 2018). HWs were found to be generally concurrent with warm season drought in most regions in China (Chen et al., 2019). In Pakistan, a positive trend was found for HWs with a maximum temperature greater than 40 °C (Zahid and Rasul, 2012). Stronger winds and lower relative humidity are distinctive characteristics of HWs in Pakistan (Khan et al., 2019a).

In the United States, the 1995 Chicago HW was described as the “urban inferno” (Klinenberg, 2015). Although this HW was relatively short (3 days), extreme temperatures accompanied by high humidity made for severe heat index conditions (Kaiser et al., 2007; Russo et al., 2017). Tavakol et al. (2020); Tavakol and Rahmani (2019) have documented an upward trend in severe heat index conditions for the central MRB. Results from HW studies in the United States show an increasing frequency of HW, especially over the Great Plains and eastern United States (Alexander et al., 2006; Smith et al., 2013). While the changes in the frequency of HW events have been studied, there is still limited knowledge on changes in HW length which is expected to increase in the 21st century (Meehl and Tebaldi, 2004).

Considering the impacts of HW on water availability and human and animal mortality, and the projected probabilities of more frequent, more intense, and longer hot summers in North America (Meehl and Tebaldi, 2004), it is important to understand and analyze changes in high temperatures and HWs in the United States. The objective of this study is to assess spatial patterns and temporal changes for HWs in the MRB and specify regions and times with more risk of experiencing a HW. Previous studies used different sets of data, study periods, HW definitions, and diverse temperature variables to analyze HWs in the United States (Alexander et al., 2006; DeGaetano and Allen, 2002; Lyon and Barnston, 2017; Oswald, 2018; Smith et al., 2013). Station-based temperature observations are limited by missing information and cannot perfectly represent the spatial pattern of HWs. To cope with this issue, researchers have used either interpolated (i.e., gridded) (Oswald, 2018) or model-based (i.e., reanalysis) data (Smith et al., 2013) to spatially and temporally provide a complete dataset. In this study, the National Centers for Environmental Prediction/National Center for Atmospheric Research (NCEP-NCAR) reanalysis data covering 1948–2017 were used. In addition, a change-point detection analysis was completed to determine any sudden change in HW frequency. Analysis of trends before and after the change-point will provide a better understanding of temporal change in HWs.