Results
Observed changes in outdoor days
Here, we use the state-of-the-art ERA5 reanalysis to present a snapshot of outdoor days (Fig. 1). Based on our analysis, most of the population living in the US enjoyed frequent outdoor days (on average, about 172 days per year in the continental US) over the past several decades, although significant regional differences exist (Fig 1a). In particular, the southern US areas located south of 40° N, such as Hawaii, California, and Florida, stand out with more frequent outdoor days, compared to the northern regions. From a seasonal perspective, a peak of outdoor days occurs during summer when warm temperatures prevail (on average, about 56 days per summer in the continental US; Fig. S2). Meanwhile, outdoor activities are highly limited during winter due to cold conditions (on average, about 1 day per winter in the areas located north of 40° N).
A trend analysis reveals that the unprecedented recent warming, as a consequence of anthropogenic climate change, has led to marginal changes in annual outdoor days across the US (Fig. 1b) but a significant shift in the seasonality (Fig. S3). The results of the trend analysis in annual outdoor days show a weak north-south gradient (Fig. 1b). Northern regions, including Northeast, Upper Midwest, Northern Rockies and Plains, and Northwest, show weak increasing trends of outdoor days (Fig. S4), benefiting from recent climate change. Whereas weak declining trends are found in the South and Southeast. However, it is important to note that responses of outdoor days to increasing greenhouse gas concentration are likely to vary substantially across seasons (Fig. S3). The largest reduction is found in boreal summer over the south-eastern US, while the increasing trend is evident in relatively cold months in the north-western US. These trends in outdoor days may contribute to the evident disparity in climate impacts between the northwest (e.g., Washington) and the southeast (e.g., Florida).
Projected changes in temperature and outdoor days
By utilizing NASA Earth Exchange Global Daily Downscaled Projections (NEX-GDDP-CMIP6), we project how the climate system in the US responds to externally imposed radiative forcing, in terms of temperature, assuming low and high emissions scenarios (Fig. 2). It is projected that an increase in temperature will exceed at least 5 °C across the US by 2100 under the SSP5-8.5 scenario, while the largest increases in temperature are expected in the northern US (Fig. 2a). Regionally, the Northeast, Upper Midwest, and Northern Rockies and Plains are likely to experience significant warming of 6 to 7 ℃ (Fig. 2b). A similar but relatively weak warming trend is projected in the Northwest, West, South, and Southeast. The overall trends in temperature indicate nearly uniform warming across the different climate regions in the country. Note that there is a high inter-model agreement on the sign of the change across different generations of models (Fig. 2 and Fig. S5), and the rate of warming is significantly lower under the low emissions scenario compared to the high emissions scenario (Fig. 2).
The story is different for future anticipated changes in outdoor days. The projected warming is likely to alter the seasonality of outdoor days significantly, which has important implications for the quality of life in different communities (Fig. 3). For example, statistically significant and robust decreases in outdoor days are expected especially in the middle of summer (i.e., days too hot for outdoor activities), while some more days in early spring and late fall and a significant increase in winter are projected (i.e., “broadening of the shoulders”; Figs. S6-S7). Interestingly, the balance of annual outdoor days is little changed over most US climate regions with two notable exceptions. The northwest exhibits higher increases in outdoor days, by 14 %, driven by small decreases in outdoor days during summer and relatively large increases during winter. Meanwhile, we project considerable decreases in outdoor days during warm months that exceed the projected increases during cold months, leading to significant drops of outdoor days in the southeast (-23 %), south (-19 %), and Ohio Valley (-18 %). These findings highlight the varying impacts of climate change on outdoor days across different regions.
Consequently, the nearly uniform warming translates into a significant northwest-southeast disparity in the projected change of annual outdoor days (Fig. 4). The Ohio Valley, South, and Southeast, which are relatively less prosperous by national standards (Fig. S8), could be disproportionately affected by the negative impacts of climate change, along with reduced annual outdoor days, limiting outdoor activities. At a local scale, major metropolitan areas like Chicago, Houston, and Charlotte, known for their high tourism rates, may face a shortage of outdoor days by the end of this century (Fig. S9). Note that people in these regions are relatively more skeptical about climate change and are less supportive of policies to mitigate climate change (45,46). Meanwhile, despite of global warming, the overall impact of higher temperatures leads to more, or no change, in annual outdoor days, especially in relatively wealthy states of the north-western US (Fig. S8). Achieving net-zero carbon emissions by the century’s second half (i.e., SSP1-2.6 scenario) could significantly address the issue of reduced outdoor days, thereby mitigating the risks of climate change. Failure to do so could pose a threat to the tourism and recreation industries, especially in Los Angeles, Houston, New York, Phoenix, Chicago, and Charlotte (Fig. S9). The projected trends in outdoor days from the NEX-GDDP-CMIP6 Global Climate Models (GCMs) is consistent with the observed trends (Fig. 1b), and the projected trends from CMIP5 models (Fig. S10), indicating a robust signal of climate change.
We propose an explanation for these projections. The changes in outdoor days are dictated by the position of the probability distribution of daily mean temperature relative to the thresholds defining outdoor days (Fig. S11). For example, the projected probability distributions of temperature especially in the Ohio Valley, South, and Southeast are likely to be far away from the conditions for thermal comfort, limiting outdoor activities significantly. On the other hand, the probability distribution of temperature in the northern US, in relatively cold months of the year, is projected to approach favorable conditions for outdoor days, resulting in more days.