• Abera, T. A., J. Heiskanen, P. K. E. Pellikka, and E. E. Maeda, 2020: Impact of rainfall extremes on energy exchange and surface temperature anomalies across biomes in the Horn of Africa. Agric. For. Meteor., 280, 107779, https://doi.org/10.1016/j.agrformet.2019.107779.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Allen, M. R., and W. J. Ingram, 2002: Constraints on future changes in climate and the hydrologic cycle. Nature, 419, 228232, https://doi.org/10.1038/nature01092.

    • Search Google Scholar
    • Export Citation
  • An, K. J., Y. F. Lam, S. Hao, T. E. Morakinyo, and H. Furumai, 2015: Multi-purpose rainwater harvesting for water resource recovery and the cooling effect. Water Res., 86, 116121, https://doi.org/10.1016/j.watres.2015.07.040.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Anderson, S. P., A. Hinton, and R. A. Weller, 1998: Moored observations of precipitation temperature. J. Atmos. Oceanic Technol., 15, 979986, https://doi.org/10.1175/1520-0426(1998)015<0979:MOOPT>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Betts, A. K., and J. H. Ball, 1998: FIFE surface climate and site-average dataset 1987–89. J. Atmos. Sci., 55, 10911108, https://doi.org/10.1175/1520-0469(1998)055<1091:FSCASA>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Boer, G. J., 1993: Climate change and the regulation of the surface moisture and energy budgets. Climate Dyn., 8, 225239, https://doi.org/10.1007/BF00198617.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bristow, K. L., and G. S. Campbell, 1984: On the relationship between incoming solar radiation and daily maximum and minimum temperature. Agric. For. Meteor., 31, 159166, https://doi.org/10.1016/0168-1923(84)90017-0.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bui, A., F. Johnson, and C. Wasko, 2019: The relationship of atmospheric air temperature and dew point temperature to extreme rainfall. Environ. Res. Lett., 14, 074025, https://doi.org/10.1088/1748-9326/ab2a26.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Byers, H. R., H. Moses, and P. J. Harney, 1949: Measurement of rain temperature. J. Atmos. Sci., 6, 5155, https://doi.org/10.1175/1520-0469(1949)006<0051:MORT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Chen, G., R. Lan, W. Zeng, H. Pan, and W. Li, 2018: Diurnal variations of rainfall in surface and satellite observations at the monsoon coast (South China). J. Climate, 31, 17031724, https://doi.org/10.1175/JCLI-D-17-0373.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, X., K. Zhao, and M. Xue, 2014: Spatial and temporal characteristics of warm season convection over Pearl River Delta region, China based on 3 years of operational radar data. J. Geophys. Res. Atmos., 119, 12 44712 465, https://doi.org/10.1002/2014JD021965.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chiara, L., J. T. Allen, and M. K. Tippett, 2016: Relationships between hourly rainfall intensity and atmospheric variables over the contiguous united states. J. Climate, 29, 31813197, https://doi.org/10.1175/JCLI-D-15-0331.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cong, R., and M. Brady, 2012: The interdependence between rainfall and temperature: Copula analyses. Sci. World J., 2012, 405675, https://doi.org/10.1100/2012/405675.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dai, A., K. E. Trenberth, and T. R. Karl, 1999: Effects of clouds, soil moisture, precipitation, and water vapor on diurnal temperature range. J. Climate, 12, 24512473, https://doi.org/10.1175/1520-0442(1999)012<2451:EOCSMP>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Das Bhowmik, R., B. Suchetana, and M. Lu, 2019: Shower effect of a rainfall onset on the heat accumulated during a preceding dry spell. Sci. Rep., 9, 7011, https://doi.org/10.1038/s41598-019-43437-7.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dong, S., S. Yang, X. Yan, T. Zhang, Y. Feng, and P. Hu, 2020: The most predictable patterns and prediction skills of subseasonal prediction of rainfall over the Indo-Pacific regions by the NCEP Climate Forecast System. Climate Dyn., 54, 27592775, https://doi.org/10.1007/s00382-020-05141-5.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gao, X., Q. Zhu, Z. Yang, J. Liu, H. Wang, W. Shao, and G. Huang, 2018: Temperature dependence of hourly, daily, and event-based precipitation extremes over China. Sci. Rep., 8, 17564, https://doi.org/10.1038/s41598-018-35405-4.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Geerts, B., 2003: Empirical estimation of the monthly-mean daily temperature range. Theor. Appl. Climatol., 74, 145165, https://doi.org/10.1007/s00704-002-0715-3.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gosnell, R., C. W. Fairall, and P. J. Webster, 1995: The sensible heat of rainfall in the tropical ocean. J. Geophys. Res., 100, 18 43718 442, https://doi.org/10.1029/95JC01833.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hu, P., W. Chen, S. Chen, Y. Liu, R. Huang, and S. Dong, 2019: Relationship between the South China Sea summer monsoon withdrawal and September–October rainfall over southern China. Climate Dyn., 54, 114, https://doi.org/10.1007/s00382-019-05026-2.

    • Search Google Scholar
    • Export Citation
  • Jiang, H., Q. Deng, G. Zhou, D. Hui, D. Zhang, S. Liu, G. Chu, and J. Li, 2013: Responses of soil respiration and its temperature/moisture sensitivity to precipitation in three subtropical forests in southern China. Biogeosciences, 10, 39633982, https://doi.org/10.5194/bg-10-3963-2013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kamburova, P. L., and F. H. Ludlam, 1966: Rainfall evaporation in thunderstorm downdraughts. Quart. J. Roy. Meteor. Soc., 92, 510518, https://doi.org/10.1002/qj.49709239407.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ketzler, G., 2014: The diurnal temperature cycle and its relation to boundary-layer structure during the morning transition. Bound.-Layer Meteor., 151, 335351, https://doi.org/10.1007/s10546-013-9898-7.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Knupp, K. R., and W. R. Cotton, 1985: Convective cloud downdraft structure: An interpretive survey. Rev. Geophys., 23, 183215, https://doi.org/10.1029/RG023i002p00183.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kollet, S. J., I. Cvijanovic, D. Schuettemeyer, R. M. Maxwell, A. F. Moene, and P. Bayer, 2009: The influence of rain sensible heat and subsurface energy transport on the energy balance at the land surface. Vadose Zone J., 8, 846857, https://doi.org/10.2136/vzj2009.0005.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lee, M., C. Ho, and J. Kim, 2010: Influence of tropical cyclone landfalls on spatiotemporal variations in typhoon season rainfall over South China. Adv. Atmos. Sci., 27, 443454, https://doi.org/10.1007/s00376-009-9106-3.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lenderink, G., and E. van Meijgaard, 2008: Increase in hourly precipitation extremes beyond expectations from temperature changes. Nat. Geosci., 1, 511514, https://doi.org/10.1038/ngeo262.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, T, Y. Zeng, and X. Fu, 2018: Velocity ratio of wind-driven rain and its application on a transmission tower subjected to wind and rain loads. J. Perform. Constr. Facil., 32, 04018065, https://doi.org/10.1061/(ASCE)CF.1943-5509.0001210.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pan, C., X. Wang, L. Liu, D. Wang, and H. Huang, 2019: Characteristics of heavy storms and the scaling relation with air temperature by event process-based analysis in South China. Water, 11, 185, https://doi.org/10.3390/w11020185.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Panwar, A., M. Renner, and A. Kleidon, 2020: Imprints of evaporative conditions and vegetation type in diurnal temperature variations. Hydrol. Earth Syst. Sci., 24, 49234942, https://doi.org/10.5194/hess-24-4923-2020.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Penman, H. L., 1948: Natural evaporation from open water, bare soil and grass. Proc. Roy. Soc. London, 193A, 120145, https://doi.org/10.1098/rspa.1948.0037.

    • Search Google Scholar
    • Export Citation
  • Rooney, G. G., N. Lipzig, and W. Thiery, 2018: Estimating the effect of rainfall on the surface temperature of a tropical lake. Hydrol. Earth Syst. Sci., 22, 63576369, https://doi.org/10.5194/hess-22-6357-2018.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Schroeer, K., and G. Kirchengast, 2018: Sensitivity of extreme precipitation to temperature: the variability of scaling factors from a regional to local perspective. Climate Dyn., 50, 39813994, https://doi.org/10.1007/s00382-017-3857-9.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Srivastava, R. C., 1987: A model of intense downdrafts driven by the melting and evaporation of precipitation. J. Atmos. Sci., 44, 17521774, https://doi.org/10.1175/1520-0469(1987)044<1752:AMOIDD>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vargas Zeppetello, L. R., D. S. Battisti, and M. B. Baker, 2020: A new look at the variance of summertime temperatures over land. J. Climate, 33, 54655477, https://doi.org/10.1175/JCLI-D-19-0887.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wasko, C., W. T. Lu, and R. Mehrotra, 2018: Relationship of extreme precipitation, dry-bulb temperature, and dew point temperature across Australia. Environ. Res. Lett., 13, 074031, https://doi.org/10.1088/1748-9326/aad135.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wei, N., Y. Dai, M. Zhang, L. Zhou, D. Ji, S. Zhu, and L. Wang, 2014: Impact of precipitation-induced sensible heat on the simulation of land-surface air temperature. J. Adv. Model. Earth Syst., 6, 13111320, https://doi.org/10.1002/2014MS000322.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yin, D., M. L. Roderick, G. Leech, F. Sun, and Y. Huang, 2014: The contribution of reduction in evaporative cooling to higher surface air temperatures during drought. Geophys. Res. Lett., 41, 78917897, https://doi.org/10.1002/2014GL062039.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, L., R. Zhang, T. Hong, Y. Zhang, and Q. Meng, 2018: Impact of post-rainfall evaporation from porous roof tiles on building cooling load in subtropical China. Appl. Therm. Eng., 142, 391400, https://doi.org/10.1016/j.applthermaleng.2018.07.033.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, S., C. Meurey, and J.-C. Calvet, 2019: Identification of soil-cooling rains in southern France from soil temperature and soil moisture observations. Atmos. Chem. Phys., 19, 50055020, https://doi.org/10.5194/acp-19-5005-2019.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhao, S., and S. Yang, 2014: Dynamical prediction of the early season rainfall over southern China by the NCEP Climate Forecast System. Wea. Forecasting, 29, 13911401, https://doi.org/10.1175/WAF-D-14-00012.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • View in gallery

    Location and topographic map of the study area, Guangdong province, distribution of 86 meteorological stations and four representative stations in four climatic regions: North, Central, Southeast, and Southwest.

  • View in gallery

    A 16-yr regional and monthly mean occurrence proportion (%) at all levels of rainfall intensity over four climatic regions. Note: for display, the annotated number in each cell is rounded to the nearest whole number. The sum of the original values at each rainfall intensity level is 100% (the same for Fig. 3).

  • View in gallery

    A 16-yr and spatial mean occurrence proportion (%) of rainfall events at each hour to daily total for each rainfall intensity category over convective and monsoon rainfall systems. Note: a figure for typhoon rainfall was not provided since it did not show regular patterns.

  • View in gallery

    The 16-yr mean precipitation of monsoon rainfall, convective rainfall, typhoon rainfall, and total rainfall spatial distribution (mm) during summer.

  • View in gallery

    The annual summer precipitation of each rainfall type (mm) for 2005–20 over four climatic regions. The dashed lines are trend lines of each type of rainfall with a corresponding trend equation, determination coefficient, F value, and P value on top of each panel. All chart elements with the same color represent the features of the same climatic region.

  • View in gallery

    (top) Diurnal mean curves and (bottom) hourly mean variation of Ta at four representative stations in each climatic region during summer. Note: numbers in parentheses following each group name in the legend represent sample-day counts. The total sample counts of each station is 1472. The vertical dashed lines indicate the time when the Ta minimum and maximum appear during NR days; the vertical solid marks indicate that during other rainfall days.

  • View in gallery

    Scatterplot of hourly Ta variation vs rainfall intensity, and histogram of hourly Ta variation over four rainfall intensity levels in four climatic regions during three summer type rainfalls Note: the horizontal solid lines indicate the 90th percentiles of hourly Ta variation samples in each group while the horizontal dashed lines indicate the 10th percentiles (the same in Fig. 8).

  • View in gallery

    Scatterplot of hourly Ta variation vs rainfall intensity over different RH levels preceding the rainfall in three types of summer rainfall. Note: the numbers at the bottom of each panel represents the sample size of the corresponding scatters in the same color.

  • View in gallery

    Scatterplot of hourly Ta variation vs RH preceding the rainfall, during (top) convective rainfall, (middle) monsoon rainfall, and (bottom) typhoon rainfall. Note: different color scatters imply the RH level following the rainfall.

  • View in gallery

    Scatterplot of hourly Ta variation vs hourly RH variation, during (top) convective rainfall, (middle) monsoon rainfall, and (bottom) typhoon rainfall.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 142 136 32
PDF Downloads 117 113 25

Quantifying the Rainfall Cooling Effect: The Importance of Relative Humidity in Guangdong, South China

View More View Less
  • 1 aGuangdong Climate Center, Guangzhou, China
  • | 2 bKey Laboratory of South China Sea Meteorological Disaster Prevention and Mitigation of Hainan Province, Haikou, China
  • | 3 cGuangzhou Climate and Agro-meteorology Center, Guangzhou, China
  • | 4 dSouthern Marine Science and Engineering Guangdong Laboratory (Zhuhai), Guangdong Province Key Laboratory for Climate Change and Natural Disaster Studies, School of Atmospheric Sciences, Sun Yat-sen University, Guangzhou, China
Open access

Abstract

In this study, we used hourly observations to investigate the cooling effect of summer rainfall on surface air temperature (Ta) in a subtropical area, Guangdong province, South China. Data were categorized step-by-step by rainfall system (convection, monsoon, and typhoon), daily rainfall amount, and relative humidity (RH) level. Moreover, the average hourly Ta variation due to solar radiation was removed from all observations before statistical analysis. The results showed that the linear relationship between hourly Ta variation and rainfall intensity did not exist. However, the cooling effect of rainfall on Ta variation was dominant. In addition, convective rainfall does cause a greater temperature drop than the other two rainfall systems. After further partitioning all samples by RH level preceding the rainfall, the relationship between hourly Ta variation and rainfall intensity became distinctive. When RH was below 70%, rainfall-induced cooling became more substantial and scaled linearly with event intensity, but when RH exceeded 70%, the rainfall cooling effect was generally restrained by the RH increase. A strong correlation between hourly Ta variation and RH level preceding the rainfall suggests the importance of RH on the rainfall cooling effect.

© 2022 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Yu Zhang, zy2202604@163.com

Abstract

In this study, we used hourly observations to investigate the cooling effect of summer rainfall on surface air temperature (Ta) in a subtropical area, Guangdong province, South China. Data were categorized step-by-step by rainfall system (convection, monsoon, and typhoon), daily rainfall amount, and relative humidity (RH) level. Moreover, the average hourly Ta variation due to solar radiation was removed from all observations before statistical analysis. The results showed that the linear relationship between hourly Ta variation and rainfall intensity did not exist. However, the cooling effect of rainfall on Ta variation was dominant. In addition, convective rainfall does cause a greater temperature drop than the other two rainfall systems. After further partitioning all samples by RH level preceding the rainfall, the relationship between hourly Ta variation and rainfall intensity became distinctive. When RH was below 70%, rainfall-induced cooling became more substantial and scaled linearly with event intensity, but when RH exceeded 70%, the rainfall cooling effect was generally restrained by the RH increase. A strong correlation between hourly Ta variation and RH level preceding the rainfall suggests the importance of RH on the rainfall cooling effect.

© 2022 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).

Corresponding author: Yu Zhang, zy2202604@163.com

1. Introduction

Rainfall has complicated interactions with meteorological factors, including air temperature, relative humidity, wind speed, etc. (Pan et al. 2019). Among these factors, the impact of air temperature on precipitation extremes has been extensively investigated (Allen and Ingram 2002; Lenderink and van Meijgaard 2008; Gao et al. 2018; Schroeer and Kirchengast 2018; Wasko et al. 2018) based on the theoretical basis of the Clausius–Clapeyron equation. Boer (1993) established the idea that the change in the amount of tropospheric water vapor could be understood from the Clausius–Clapeyron equation, indicating that the capacity of the air to hold water vapor increases as the temperature increases generally by about 7% °C−1. However, based on many related studies over different regions and periods, Pan et al. (2019) concluded that the scaling rates between air temperature and precipitation extremes are not necessarily close to 7% °C−1. Some scaling rates were reported positive but less or greater than that, some scaling rates were reported negative unexpectedly.

On the other hand, concerning the impact of rainfall on temperature change, it is generally believed that rainfall leads to a decrease of surface air temperature, through (i) direct rain heat flux, and (ii) evaporative cooling before the surface air gets saturated during precipitation (Rooney et al. 2018). Regarding the direct rain heat flux, i.e., the precipitation-induced sensible heat (HPR), previous studies stated that it can be extremely large and have a noticeable effect on the local surface air temperature during heavy precipitation events (Gosnell et al. 1995; Anderson et al. 1998; Kollet et al. 2009), due to the much lower temperature of raindrops than the ambient air temperature at the start of a rainfall event (Byers et al. 1949). However, HPR is commonly neglected in current land surface models because of its small magnitude on long time scales (Wei et al. 2014). With regard to the evaporative cooling effect, which is dependent on the wet-bulb depression, studies indicate that the amount of water available in precipitation directly affects surface air temperature (Betts and Ball 1998; Yin et al. 2014) through energy transfer from the surface air to the ground for evaporation when the air is unsaturated. The cooling influence of rainfall is tempered by extremely high relative humidity values, which limit the total evaporation period (Vargas Zeppetello et al. 2020).

The process whereby rainfall brings down the temperature is known as the cooling effect. Due to such effects, negative correlations between temperature and rainfall have been reported in many studies. For example, Cong and Brady (2012) found that there were negative correlations between these two variables during April–July and September in Scania, Sweden. Abera et al. (2020) discovered that surface air temperature anomalies negatively and consistently responded to rainfall events in both wet (−2.33 K over grassland in March, April, and May) and dry seasons (−2.48 K over grassland in October, November, and December) over all vegetation types. Soil-cooling rain was defined (Zhang et al. 2019) as rainfall events triggering a drop in soil temperature. The cooling effect also occurs over water bodies. Rooney et al. (2018) stated that days with heavy rainfall witness a stronger reduction in lake surface temperature by the end of the day, approximately 0.3 K higher than days with light to moderate rainfall. Das Bhowmik et al. (2019) defined this weather phenomenon as “the shower effect” because of a general sense of immediate heat relief brought by the rainfall, resembling a cold shower on a hot dry day. Essentially, the surface air temperature can be sharply altered by precipitation on an hourly to daily time scale (Wei et al. 2014).

Due to the advantage of such a rainfall cooling effect, it can be applied to people’s daily life especially during summer. For example, harvested rainwater is utilized for temperature reduction at the top of buildings (An et al. 2015). Experimental results from Zhang et al. (2018) showed that the evaporation process can decrease the maximal values of the external and internal roof surface temperatures by up to 6.4° and 3.2°C, respectively. Using the principle of evaporative cooling, the evaporative cooler has been the subject of numerous U.S. patents in the twentieth century.

Evidence of the significance of rain effects on land surface air temperature, particularly in the tropics, has already started to emerge (Wei et al. 2014). In addition, southern China frequently witnesses the persistence of extremely high temperature or heat waves during summer. Rainfall following a dry spell can be the most anticipated weather phenomenon with possibilities to bring down the air temperature. However, the cooling effect of precipitation on air temperature has not been well investigated in southern China, especially on hourly time scales. In light of this and based on hourly observations, we conducted an investigation into the relationship between hourly temperature variation and summer rainfall over different types of rainfall synoptic systems in southern China.

2. Study area and data

a. Study area

Guangdong Province, located in southern China, is highly influenced by a subtropical and tropical monsoon climate. Its area cover is depicted on the map in Fig. 1. With the Nanling Mountains in its north and being adjacent to the South China Sea, it is characterized by complex terrain with higher elevation in the north and lower elevation in the south. Four climatic regions—North, Central, Southeast, and Southwest—occur in Guangdong defined by their distinctive climatic features. Due to the distribution of terrain and certain atmospheric circulation, which provides a superb situation for intensive rainfall, the frequency, intensity, and persistence of rainstorms in Guangdong are high. There are two concentrated rain periods in Guangdong with one from April to June and the other from July to September, namely, the first rainy season and second rainy season, respectively (Zhao and Yang 2014). Rainfall during summer (June–August) also accounts for nearly 50% of the annual precipitation, and it mainly comes from the summer monsoon, tropical typhoons, and convective activities (Lee et al. 2010; Hu et al. 2019; Dong et al. 2020). The 16-yr and spatial mean monthly occurrence proportions for all rainfall intensity levels (Fig. 2) show that the North climatic region witnesses a major rainfall peak during May and June. The Central and Southeast witness a major and minor peak, respectively, in May–June, and in August, while the Southwest witnesses double peaks in June and August. Additionally, the mean of the annual total rainfall events for each rainfall intensity category (Table 1) shows that more events occur for small rainfall intensities and the number decreases by rainfall intensity. Spatially, a rainfall intensity of 1–4.9 mm h−1 for the North ranks highest (270.1 times) while the Southwest ranks last (186.6 times). For rainfall intensities greater than 10 mm h−1, the annual total number of rainfall events in the North was the lowest (17.1 times for 10–14.9 mm h−1).

Fig. 1.
Fig. 1.

Location and topographic map of the study area, Guangdong province, distribution of 86 meteorological stations and four representative stations in four climatic regions: North, Central, Southeast, and Southwest.

Citation: Journal of Hydrometeorology 23, 6; 10.1175/JHM-D-21-0155.1

Fig. 2.
Fig. 2.

A 16-yr regional and monthly mean occurrence proportion (%) at all levels of rainfall intensity over four climatic regions. Note: for display, the annotated number in each cell is rounded to the nearest whole number. The sum of the original values at each rainfall intensity level is 100% (the same for Fig. 3).

Citation: Journal of Hydrometeorology 23, 6; 10.1175/JHM-D-21-0155.1

Table 1

A 16-yr and spatial mean of annual total rainfall events for each rainfall intensity category in four climatic regions (unit: number of times).

Table 1

b. Hourly observational weather data

There are 86 national meteorological observatories in Guangdong Province (Fig. 1) with 27 observatories in North, 29 in Central, 13 in Southeast, and 17 in Southwest. Up to 2005, all meteorological data were obtained through manual observation four times per day. Since then, manual observations have been replaced by continuous automatic weather stations. Even so, all the measured data have been audited manually both real time and monthly. Therefore, the quality of the hourly data of 2005–20 from all 86 national meteorological observatories used in this study is highly reliable. The hourly data we use here consist of accumulated precipitation (P; mm), averaged air dry bulb temperature (Ta; °C), and relative humidity (RH; %) both preceding and following the rainfall in June, July, and August during 2005–20. This accounts for 1472-day samples for each station. The daily accumulated precipitation of each station is computed based on hourly precipitation. Among the variables, Ta and RH are measured at screen level (2 m) while P is on the ground. We used rainfall intensity to indicate hourly precipitation in the following sections of this paper. Hourly Ta variation (°C h−1) is defined as the result of Ta following the rainfall minus Ta preceding the rainfall, and the same for hourly RH variation (% h−1).

To determine whether the precipitation–temperature relationship differs in each synoptic system, all data samples were first divided into three groups according to rainfall system, including convection, monsoon, and tropical typhoon. Afterward, data were further stratified into five rainfall groups (No Rain, NR; Light Rain, LR; Moderate Rain, MR; Heavy Rain, HR; and Storm, S) based on daily accumulated precipitation (Table 2). Due to the solar radiation, the temperature in all rainfall groups still showed single-peak diurnal variation (see Fig. 6). The temperature in the NR group was not influenced by rainfall events, therefore the average diurnal cycle of the temperature in this group can be regarded as the background diurnal cycle due to solar radiation, which was removed from all the other rainfall groups.

Table 2

Categorized rainfall groups based on daily rainfall amounts (mm).

Table 2

In addition, the 10th and 90th percentiles of hourly Ta variation were used to express its distribution features in different conditions.

c. Tropical typhoon information

Tropical typhoon information is used here to distinguish typhoon system rainfall from the other two types. After typhoon samples were filtered, the rest of the data obtained in June were marked as monsoon data, and all the others obtained during July and August were marked as convective data (Pan et al. 2019).

d. Evaporative cooling mechanism

Rainfall generally leads to a wet ground surface and evaporation from this wet surface can be expressed by the Penman equation (Penman 1948):
LυE=s(Q*G)s+γ+ρcpra[esat(T¯a)e¯a]s+γ,
where LυE is the latent heat flux and latent heat can be extracted only by a phase change, s is the slope of the saturated vapor pressure versus temperature curve (Pa K−1), Q* is the net radiation (W m−2), G is the soil heat flux (W m−2), γ is the psychrometric constant (Pa K−1), ρ is the density of air (kg m−3), cp is the specific heat capacity of air at constant pressure (J kg−1 K−1), ra is the aerodynamic resistance for heat and water vapor (s m−1), esat(T¯a) and e¯a is the saturated vapor pressure (Pa) and the vapor pressure (Pa) respectively, both at the surface air temperature.

The first term on the right-hand side of Eq. (1) is called the radiation term because it describes the evaporation due to energy input by radiation. The second term is called the aerodynamic term because it depends explicitly on the turbulent transport and the atmospheric moisture conditions through esat(T¯a)e¯a [vapor pressure deficit (VPD)]. VPD shows that even if there is no net input of energy by Q*G, evaporation can still proceed. For such cases, the latent heat from the surface to the atmosphere occurs when water evaporates, and then the energy required for evaporation will be extracted from the air to the ground through a negative sensible heat flux, leading to a lower surface air temperature. This can provide a good explanation to air cooling through evaporation when the surface air does not get saturated, i.e., when the RH is still low.

3. Results and analysis

a. Temporal and spatial characteristics of rainfall

Before discussing the precipitation–temperature relationship, we analyzed the summer diurnal pattern of all rainfall intensity categories over convective, monsoon, and typhoon rainfall. This was followed by an analysis of annual variation and spatial distribution of these three types of rainfall in order to have a basic understanding of temporal and spatial characteristics of summer rainfall in our study area.

1) Diurnal distribution of each summer rainfall system

The 16-yr and spatial mean occurrence proportions (unit: %) of rainfall events at each hour (Fig. 3) clearly show that in regions other than Southeast, convective rainfall has an obvious rainfall peak in the afternoon, i.e., between 1500 and 1800 LT than during the rest of a day during summer, reflecting that an unstable atmosphere leading to convective precipitation tends to occur in the hot summer afternoons. This result is comparable to Chen et al. (2014), who reported that convection occurs most frequently around 1400–1500 LT in the same study area. Similar diurnal pattern can also be seen from monsoon rainfall to a certain extent in most regions except for Southeast. Figure 3 also shows that it is least possible that monsoon and convective rainfall occurs around midnight (2200–0400 LT). These findings are consistent with conclusions from Chen et al. (2018) who also found an afternoon peak of rainfall in Guangdong based on hourly rain gauge data for 1998–2014. In contrast, typhoon rainfall does not show a distinctive diurnal pattern (not presented), because it depends on large-scale circulation and has little dependence on the radiation diurnal cycle.

Fig. 3.
Fig. 3.

A 16-yr and spatial mean occurrence proportion (%) of rainfall events at each hour to daily total for each rainfall intensity category over convective and monsoon rainfall systems. Note: a figure for typhoon rainfall was not provided since it did not show regular patterns.

Citation: Journal of Hydrometeorology 23, 6; 10.1175/JHM-D-21-0155.1

2) Spatial characteristics of each rainfall system

In Fig. 4, we can see that summer monsoon rainfall primarily appears in the southeastern coastal region, central area, and a small southwestern coastal region where the seasonal rainfall amount exceeds 350 mm. In contrast, monsoon rainfall in the Leizhou Peninsula in the utmost Southwest is less than 200 mm. The convective rain belt mainly lies along the coastal area from Southeast to Southwest, and also with a rainfall amount over 350 mm. Convective rainfall gradually decreases from the coastal area to the inland region according to the distance from the ocean, indicating that convective rainfall is closely related to the moist air transported from the ocean onto the continent under intense surface heating. The spatial distribution of typhoon rainfall to some extent resembles that of the convective rainfall. The rain also concentrates in the coastal area (200 mm) and decreases over the inland region (100 mm), implying that the influence of typhoons is rapidly weakened after making landfall. Overall, summer total rainfall peaks in three local regions, including two local coastal areas and one to the east of Pearl River Estuary.

Fig. 4.
Fig. 4.

The 16-yr mean precipitation of monsoon rainfall, convective rainfall, typhoon rainfall, and total rainfall spatial distribution (mm) during summer.

Citation: Journal of Hydrometeorology 23, 6; 10.1175/JHM-D-21-0155.1

3) Annual characteristics of each rainfall system

Figure 5 presents the annual summer precipitation due to each rainfall system during the past 16 years for four climatic regions. In the monsoon rainfall panel, the four significance levels (P value) of the trend lines are close to or less than 0.05, indicating a significant downtrend in monsoon rainfall for 2005–20. The past years also witnessed a significant uptrend in convective rainfall over the North climatic region, possibly indicating climate warming in this region. With regard to convective rainfall over the other regions and typhoon rainfall, no significant trend in annual variation was observed.

Fig. 5.
Fig. 5.

The annual summer precipitation of each rainfall type (mm) for 2005–20 over four climatic regions. The dashed lines are trend lines of each type of rainfall with a corresponding trend equation, determination coefficient, F value, and P value on top of each panel. All chart elements with the same color represent the features of the same climatic region.

Citation: Journal of Hydrometeorology 23, 6; 10.1175/JHM-D-21-0155.1

b. Diurnal variation pattern of air temperature

Overland, the diurnal temperature variation is mainly shaped by the absorption of solar radiation (Bristow and Campbell 1984; Dai et al. 1999; Panwar et al. 2020). Therefore, the solar radiation effect cannot be ignored for a better understanding of the cooling effect of rainfall. To distinguish the solar radiation contribution to hourly Ta variation during rainfall, the 16-yr-mean 24-h Ta for NR days and other four rainfall groups (Table 3) at each station were computed and the diurnal curves of Ta were plotted (Fig. 6). The diurnal cycles of Ta at all stations present a high similarity to each other. Only plots of four representative stations are shown here, but the spatial mean parameters of Ta over each rainfall group are listed in Table 3. The sample size of each rainfall group in each climatic region shows that during summer in the past 16 years, days without rainfall (NR) and days with light rainfall (LR) were dominant, accounting for around 45% and 30%, respectively. The sample size decreases by the daily rainfall total.

Fig. 6.
Fig. 6.

(top) Diurnal mean curves and (bottom) hourly mean variation of Ta at four representative stations in each climatic region during summer. Note: numbers in parentheses following each group name in the legend represent sample-day counts. The total sample counts of each station is 1472. The vertical dashed lines indicate the time when the Ta minimum and maximum appear during NR days; the vertical solid marks indicate that during other rainfall days.

Citation: Journal of Hydrometeorology 23, 6; 10.1175/JHM-D-21-0155.1

Table 3

Spatial mean parameters of Ta over each rainfall group. Note: T_Max stands for maximum Ta (°C); T_Min for minimum Ta (°C); DTR for diurnal temperature range (°C); H_Max for the hours when T_Max appears (LT); H_Min for the hours when T_Min appears (LT); N is the sample size.

Table 3

The gray curves in Fig. 6 are the 16-yr-mean Ta diurnal cycle of NR days in summer. Their single-peak structures demonstrate the diurnal variation of net radiation, i.e., the sum of shortwave solar radiation and longwave terrestrial radiation. Similar single-peak structures also exist in other groups, with a smaller diurnal temperature range (DTR) as rainfall increases. This indicates that solar radiation is still dominant in determining the diurnal Ta variations, even though insolation is reduced due to cloud cover that generates rainfall and that Ta variation is disturbed by rainfall under rainy weather conditions. Geerts (2003) claimed that the correlation coefficient between the monthly mean DTR and the total cloudiness was −0.70. Similarly, here we found that DTR during NR days was always largest as compared to those during the rainfall days, and it generally and consistently decreased from NR to LR, MR, HR, and S. The spatial mean DTR during NR days reached 9.0°C in the North, and 6.2°–7.2°C in other regions. The spatial mean DTR during S (storm) days was between 1.5° and 2.9°C.

The curves in Fig. 6 also show that the Ta of each group is very close to each other at the start of the day, but that by the end of the day Ta in rainfall days experienced a reduction of approximately 1°–3°C as compared to NR days. Additionally, Ta on rainfall days was consistently lower than that of NR days all day long, which could be the result of multiday precipitation driving down the temperature.

During NR days, the Ta maximum mainly appears at 1400–1500 LT, similar to Ketzler’s (2014) study who reported that the temperature maximum appears at about 1500 solar time. Peak daytime temperature mostly occurs 1–2 h earlier in rainfall days than in NR days. This effect is probably due to the overwhelming amount of precipitation that falls during the afternoon (see Fig. 3). In contrast, the time when the Ta minimum appears is rather steady, which is mostly at 0600 LT regardless of weather conditions (rainfall group), implying again that less precipitation appears in the morning and does not disturb the average effect of the solar radiation.

From the plot of hourly mean Ta variation in Fig. 6, it is clear that Ta generally rises in the morning until early afternoon and with a maximum rising rate (1.4°–1.8°C h−1) generally occurring at the beginning of the Ta rising period. Ta mainly drops from afternoon until evening, but from midnight till early morning it drops rather slowly (0.2°C h−1) or is steady.

c. Relationship between hourly temperature variation and rainfall intensity

As described above, solar radiation is always dominant in determining Ta variations, even under rainfall weather conditions. The average diurnal cycle of Ta in NR days reflects the background influence of solar radiation on hourly temperature variation. Therefore we used the average hourly Ta variation in NR days to remove such influence from solar radiation.

Based on hourly Ta variation without radiation effect, the scatters of rainfall intensity versus it under each rainfall system over four climatic regions were plotted (Fig. 7), along with the stacked histograms of hourly Ta variation at four rainfall intensity levels. As found in Fig. 7, the scatters do not indicate a linear relationship between hourly Ta variation and rainfall intensity. Additionally, scatters not only distribute below the 0°C h−1 horizontal line, but also partly above it. Table 4 provides the fraction of the total observations of such points, which is around 30%, but the fractions drop to less than 15% for samples where Ta variation is greater than 0.5°C h−1. This phenomenon might be due to random solar radiation variability because when the sun comes out quickly after a rainfall, the temperature could rise quickly. If the radiation effect is greater than the rainfall cooling effect, a positive Ta variation could occur. Additionally, there might be other contributing factors influencing local temperature.

Fig. 7.
Fig. 7.

Scatterplot of hourly Ta variation vs rainfall intensity, and histogram of hourly Ta variation over four rainfall intensity levels in four climatic regions during three summer type rainfalls Note: the horizontal solid lines indicate the 90th percentiles of hourly Ta variation samples in each group while the horizontal dashed lines indicate the 10th percentiles (the same in Fig. 8).

Citation: Journal of Hydrometeorology 23, 6; 10.1175/JHM-D-21-0155.1

Table 4

The fraction (%) of total observation over each hourly Ta variation (°C h−1) range in four climatic regions during three summer type rainfalls.

Table 4

Regarding the data distribution range, negative hourly Ta variations can range from 0°C h−1 to below −10°C h−1, while the positive hourly Ta variation range is much narrower, generally within 0°–2°C h−1. This stresses the dominance of the cooling effect of rainfall on air temperature. Concerning rainfall intensity, the histograms show that rainfall intensity weaker than 10 mm h−1 (green and blue bars) mainly induces hourly Ta variation within ±2°C h−1. A Ta decrease greater than 2°C h−1 is more frequently caused by rainfall intensity greater than 10 mm h−1 (red and purple bars).

Last, with regard to different rainfall systems, it should be pointed out that there are more scatters in each range of negative Ta variation during convective rainfall than during the other two types, as reflected by obvious fraction differences among different rainfall types (Table 4). This probably implies that convective rainfall tends to cause a greater surface temperature drop than the others.

d. Relationship between hourly temperature variation and relative humidity

As explained in the introduction, evaporative cooling is one of the main sources leading to a decrease in surface air temperature. But once the atmosphere is saturated with water vapor, the evaporation process and its cooling effect stop, and both rain and air will approach the air’s original wet-bulb temperature (Rooney et al. 2018). To investigate the role of the evaporative cooling effect on determining the Ta variation, we categorized all hourly samples into six groups based on the preceding humidity (RH) level recorded before the occurrence of rainfall: 41%–50%, 51%–60%, 61%–70%, 71%–80%, 81%–90%, and 91%–100%, respectively. Based on this categorization scatters of hourly Ta variation versus rainfall intensity were plotted again (Fig. 8). As can be seen in Fig. 8, the sample sizes (numbers at each panel’s bottom) at RH lower levels were relatively smaller, but with an increase of RH, the sample size of all rainfall groups also increased. Below RH 80% over convective and monsoon rainfall, the samples of LR days were dominant, sample sizes decreased from the LR group to the MR, HR, and S groups. In contrast, when RH was above 90%, the samples of the S group became dominant and the sample size of the LR group became the smallest. However, typhoon rainfall is an exception and sample size of the LR day is always the smallest compared to other rainfall days, probably implying that typhoons tend to bring strong rainfall weather instead of light ones.

Fig. 8.
Fig. 8.

Scatterplot of hourly Ta variation vs rainfall intensity over different RH levels preceding the rainfall in three types of summer rainfall. Note: the numbers at the bottom of each panel represents the sample size of the corresponding scatters in the same color.

Citation: Journal of Hydrometeorology 23, 6; 10.1175/JHM-D-21-0155.1

In Fig. 8, although hourly Ta variation and rainfall intensity remained linearly independent over each RH level, their relationship became distinctive. Figure 8 and Table 5 indicate that when RH was below 80%, very few samples (less than 10%) were distributed above the 0°C h−1 line, and that hourly Ta variation tended to become negatively greater from the LR group to the MR, HR, and the S groups. When RH increased, more scatters shifted above the 0°C h−1 line, and hourly Ta variation tended to get closer to certain narrow spans around its median, indicating that the rainfall cooling effect is generally restrained by RH increase. Moreover, the higher the RH, the greater the possibility of higher rainfall intensity. Though some scatters appear above the 0°C h−1 line, it is apparent that over RH 61%–70%, when rainfall intensity exceeded 10 mm h−1, a positive hourly Ta variation did not occur. Over RH 71%–80%, when rainfall intensity was over 30 mm h−1, a positive hourly Ta variation almost disappeared. For RH 81%–90%, when rainfall intensity was larger than 50 mm h−1, a positive hourly Ta variation did not occur. Over RH 91%–100%, when rainfall intensity was over 50 mm h−1, a positive hourly Ta variation only occasionally occurred. These results indicate that RH and rainfall intensity are both critical to hourly Ta variation.

Table 5

The 90th and 10th percentiles of hourly Ta variation (0°C h−1) of three rainfall types and mean over different RH levels preceding the rainfall, in different rainfall groups. Note: the 90th and 10th stand for the 90th and 10th percentile of hourly Ta variation, respectively.

Table 5

From the 90th percentiles (solid line) and 10th percentiles (dashed line) of hourly Ta variation and Table 5, both lines shift from a lower position to a higher position (RH 41%–50% is an exception due to small sample size) as RH increases. In general, the rainfall cooling effect is damped by the increase in RH. It should be pointed out that the positive Ta variation became stagnant after RH exceeded 70%, even though when RH is below 50%, the positive Ta variation no longer existed. Accordingly, it can be inferred that the RH preceding rainfall might play a more important role in determining the hourly Ta variation than rainfall intensity.

The phenomenon that increased RH depresses the rainfall cooling effect can be verified by the scatterplot of hourly Ta variation versus initial RH preceding the rainfall (Fig. 9), as the hourly Ta variation increased according to RH level increase. After dividing the samples into four groups based on RH level following the rainfall, it is clear that for RH level 96%–100% (the rightmost panel in Fig. 9), the correlation between Ta variation and the initial RH is the weakest as compared to other groups. This result might be because that there are relatively more samples with initial RH close to 100% in this group. When the atmosphere is saturated with water vapor or close to saturation, Ta variation is not closely dependent on RH itself, while other factors are more important. Similar characteristics exist among these three different rainfall types.

Fig. 9.
Fig. 9.

Scatterplot of hourly Ta variation vs RH preceding the rainfall, during (top) convective rainfall, (middle) monsoon rainfall, and (bottom) typhoon rainfall. Note: different color scatters imply the RH level following the rainfall.

Citation: Journal of Hydrometeorology 23, 6; 10.1175/JHM-D-21-0155.1

The analysis in Fig. 9 shows that the RH level preceding the rainfall significantly influenced the hourly Ta variation. However, Eq. (1) indicates that the potential evaporation which induces evaporative cooling leading to the temperature decrease is proportional to VPD. That is VPD determines the theoretical limit of the evaporation, while the actual evaporation should be reflected by the actual RH increase. To verify this, the scatters of hourly Ta variation versus hourly RH variation were plotted (Fig. 10). The scatters showed that after rainfall the RH of a large quantity of samples reached a high level of 81%–100%, where the air generally becomes saturated. However, RH does not necessarily increase during rainfall, RH decrease also occurs, probably because the rain removes water vapor through air condensation and deposits it on the surface. Therefore, RH variation is a combined effect of evaporation and air condensation during the rainfall process. Figure 10 shows that over each RH level preceding the rainfall, a strong negative linear correlation between hourly Ta variation and RH variation exists. However, the correlation coefficients became smaller as the initial RH increased. When the initial RH levels reached 91%–100%, the correlation between hourly Ta variation and RH variation became rather weak. The linear regression models were fit over different rainfall types, and the determination coefficients of these models ranged from 0.569 to 0.683. It can be inferred that hourly Ta variation reflecting the rainfall cooling effect may be indicated by hourly RH variation.

Fig. 10.
Fig. 10.

Scatterplot of hourly Ta variation vs hourly RH variation, during (top) convective rainfall, (middle) monsoon rainfall, and (bottom) typhoon rainfall.

Citation: Journal of Hydrometeorology 23, 6; 10.1175/JHM-D-21-0155.1

4. Summary and discussion

We investigated the cooling effect of summer rainfall on surface air temperature based on hourly observed precipitation (P), air temperature (Ta), and relative humidity (RH) from 2005 to 2020 in a subtropical area, Guangdong province, southern China. Data were initially categorized by summer rainfall system (monsoon, typhoon, and convection), for which temporal and spatial distribution characteristics were analyzed to give a basic understanding of summer rainfall in the study area.

To understand the radiation effect on Ta diurnal variation, data were further divided into groups NR, LR, MR, HR, and S based on daily rainfall amount. Similar single-peak structures of Ta diurnal curves in each group indicated that solar radiation was dominant in determining the hourly temperature variations. The Ta maximum (curve peak) appeared at about 1500 solar time during NR days, but appeared 1–2 h earlier during rainfall days. The Ta minimum (curve trough) mostly appeared at 0000 LT, regardless of the weather condition (rainfall group).

The average Ta diurnal cycle was removed from all observations to avoid the radiation effect. However, the linear relationship between hourly Ta variation and rainfall intensity did not exist. Scatters were distributed mostly below the 0°C h−1 horizontal line and partly above it. A wide temperature range (from 0°C h−1 to below −10°C h−1) where negative hourly Ta variations spread, was in contrast with a narrow range (0°–2°C h−1) where positive Ta variation distribution stressed the dominance of the cooling effect of rainfall on air temperature. Convective rainfall tended to cause greater surface temperature drops than the other two rainfall systems.

After further partitioning all samples by RH level (41%–50%, 51%–60%, 61%–70%, 71%–80%, 81%–90%, and 91%–100%), the relationship between hourly Ta variation and rainfall intensity became noticeable. When RH was below 70%, the hourly Ta variation tended to become negatively greater as rainfall intensity increased. When the RH was above 70%, more scatters shifted above the 0°C h−1 line, indicating that the rainfall cooling effect was generally restrained by the RH increase. Over certain rainfall intensities at each RH level, positive hourly Ta variation did not occur. This indicates that RH and rainfall intensity both play a critical role in determining the hourly Ta variation, though the former can be more important. Moreover, the strong correlation between hourly Ta variation and RH level preceding the rainfall implies the importance of RH on rainfall cooling effect. Additionally, the hourly Ta variation due to such cooling effects can be indicated by hourly RH variation.

The interactions between surface air temperature, rainfall, and relative humidity are extremely complicated and have a myriad of processes (Jiang et al. 2013; Chiara et al. 2016; Li et al. 2018; Bui et al. 2019) that are outside the scope of this paper. For example, in summer thunderstorms, convective downdrafts could be caused by atmospheric cooling and falling precipitation. These downdrafts which originate at high levels could transport cold and dense air to the surface (Kamburova and Ludlam 1966; Knupp and Cotton 1985; Srivastava 1987; Rooney et al. 2018). These additional dynamic effects are of important impacts on surface temperature change but are not discussed in this study due to the limitations of our data.

Based on relevant studies on the relationship between precipitation and Ta variation, a certain quantitative connection between hourly Ta variation and hourly precipitation is expected. Based on such relationships, precipitation data quality control aided by Ta could be made. However, we found that it was difficult to establish a quantitative relationship between these two variables and that relative humidity is a key mitigating factor controlling the strength of rainfall-induced cooling.

Acknowledgments.

Wei Liu was supported by the China Scholarship Council as a visiting scholar at Wageningen University during 2018/19. He is sincerely grateful for the knowledge and coding skills that Dr. Arnold Moene taught him, which enabled the research. The authors highly appreciate the anonymous reviewers whose comments and suggestions greatly improved this manuscript. Moreover, they express their deepest thanks to Dr. Bert Heusinkveld and Dr. Simon Berkowicz for critical readings of the manuscript. The authors declare that they have no conflict of interest. This study was funded by the South China Sea Meteorological Disaster Prevention and Mitigation of Hainan Province (SCSF202009), Special Funds for Promoting High-Quality Economic Development - Marine Economic Development Project from the Department of Natural Resources of Guangdong Province (GDOE[2019]A11), the Agricultural Science and Technology Innovation and Promotion Project of Guangdong Province (2020KJ102), and the Research Project of the Guangdong Meteorological Service (GRMC2017Z04).

Data availability statement.

The hourly observational data are not open access due to meteorology regulations in China, but the statistical results are available on request from the first author, Wei Liu.

REFERENCES

  • Abera, T. A., J. Heiskanen, P. K. E. Pellikka, and E. E. Maeda, 2020: Impact of rainfall extremes on energy exchange and surface temperature anomalies across biomes in the Horn of Africa. Agric. For. Meteor., 280, 107779, https://doi.org/10.1016/j.agrformet.2019.107779.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Allen, M. R., and W. J. Ingram, 2002: Constraints on future changes in climate and the hydrologic cycle. Nature, 419, 228232, https://doi.org/10.1038/nature01092.

    • Search Google Scholar
    • Export Citation
  • An, K. J., Y. F. Lam, S. Hao, T. E. Morakinyo, and H. Furumai, 2015: Multi-purpose rainwater harvesting for water resource recovery and the cooling effect. Water Res., 86, 116121, https://doi.org/10.1016/j.watres.2015.07.040.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Anderson, S. P., A. Hinton, and R. A. Weller, 1998: Moored observations of precipitation temperature. J. Atmos. Oceanic Technol., 15, 979986, https://doi.org/10.1175/1520-0426(1998)015<0979:MOOPT>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Betts, A. K., and J. H. Ball, 1998: FIFE surface climate and site-average dataset 1987–89. J. Atmos. Sci., 55, 10911108, https://doi.org/10.1175/1520-0469(1998)055<1091:FSCASA>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Boer, G. J., 1993: Climate change and the regulation of the surface moisture and energy budgets. Climate Dyn., 8, 225239, https://doi.org/10.1007/BF00198617.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bristow, K. L., and G. S. Campbell, 1984: On the relationship between incoming solar radiation and daily maximum and minimum temperature. Agric. For. Meteor., 31, 159166, https://doi.org/10.1016/0168-1923(84)90017-0.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bui, A., F. Johnson, and C. Wasko, 2019: The relationship of atmospheric air temperature and dew point temperature to extreme rainfall. Environ. Res. Lett., 14, 074025, https://doi.org/10.1088/1748-9326/ab2a26.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Byers, H. R., H. Moses, and P. J. Harney, 1949: Measurement of rain temperature. J. Atmos. Sci., 6, 5155, https://doi.org/10.1175/1520-0469(1949)006<0051:MORT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Chen, G., R. Lan, W. Zeng, H. Pan, and W. Li, 2018: Diurnal variations of rainfall in surface and satellite observations at the monsoon coast (South China). J. Climate, 31, 17031724, https://doi.org/10.1175/JCLI-D-17-0373.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chen, X., K. Zhao, and M. Xue, 2014: Spatial and temporal characteristics of warm season convection over Pearl River Delta region, China based on 3 years of operational radar data. J. Geophys. Res. Atmos., 119, 12 44712 465, https://doi.org/10.1002/2014JD021965.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Chiara, L., J. T. Allen, and M. K. Tippett, 2016: Relationships between hourly rainfall intensity and atmospheric variables over the contiguous united states. J. Climate, 29, 31813197, https://doi.org/10.1175/JCLI-D-15-0331.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cong, R., and M. Brady, 2012: The interdependence between rainfall and temperature: Copula analyses. Sci. World J., 2012, 405675, https://doi.org/10.1100/2012/405675.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dai, A., K. E. Trenberth, and T. R. Karl, 1999: Effects of clouds, soil moisture, precipitation, and water vapor on diurnal temperature range. J. Climate, 12, 24512473, https://doi.org/10.1175/1520-0442(1999)012<2451:EOCSMP>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Das Bhowmik, R., B. Suchetana, and M. Lu, 2019: Shower effect of a rainfall onset on the heat accumulated during a preceding dry spell. Sci. Rep., 9, 7011, https://doi.org/10.1038/s41598-019-43437-7.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Dong, S., S. Yang, X. Yan, T. Zhang, Y. Feng, and P. Hu, 2020: The most predictable patterns and prediction skills of subseasonal prediction of rainfall over the Indo-Pacific regions by the NCEP Climate Forecast System. Climate Dyn., 54, 27592775, https://doi.org/10.1007/s00382-020-05141-5.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gao, X., Q. Zhu, Z. Yang, J. Liu, H. Wang, W. Shao, and G. Huang, 2018: Temperature dependence of hourly, daily, and event-based precipitation extremes over China. Sci. Rep., 8, 17564, https://doi.org/10.1038/s41598-018-35405-4.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Geerts, B., 2003: Empirical estimation of the monthly-mean daily temperature range. Theor. Appl. Climatol., 74, 145165, https://doi.org/10.1007/s00704-002-0715-3.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Gosnell, R., C. W. Fairall, and P. J. Webster, 1995: The sensible heat of rainfall in the tropical ocean. J. Geophys. Res., 100, 18 43718 442, https://doi.org/10.1029/95JC01833.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hu, P., W. Chen, S. Chen, Y. Liu, R. Huang, and S. Dong, 2019: Relationship between the South China Sea summer monsoon withdrawal and September–October rainfall over southern China. Climate Dyn., 54, 114, https://doi.org/10.1007/s00382-019-05026-2.

    • Search Google Scholar
    • Export Citation
  • Jiang, H., Q. Deng, G. Zhou, D. Hui, D. Zhang, S. Liu, G. Chu, and J. Li, 2013: Responses of soil respiration and its temperature/moisture sensitivity to precipitation in three subtropical forests in southern China. Biogeosciences, 10, 39633982, https://doi.org/10.5194/bg-10-3963-2013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kamburova, P. L., and F. H. Ludlam, 1966: Rainfall evaporation in thunderstorm downdraughts. Quart. J. Roy. Meteor. Soc., 92, 510518, https://doi.org/10.1002/qj.49709239407.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ketzler, G., 2014: The diurnal temperature cycle and its relation to boundary-layer structure during the morning transition. Bound.-Layer Meteor., 151, 335351, https://doi.org/10.1007/s10546-013-9898-7.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Knupp, K. R., and W. R. Cotton, 1985: Convective cloud downdraft structure: An interpretive survey. Rev. Geophys., 23, 183215, https://doi.org/10.1029/RG023i002p00183.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kollet, S. J., I. Cvijanovic, D. Schuettemeyer, R. M. Maxwell, A. F. Moene, and P. Bayer, 2009: The influence of rain sensible heat and subsurface energy transport on the energy balance at the land surface. Vadose Zone J., 8, 846857, https://doi.org/10.2136/vzj2009.0005.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lee, M., C. Ho, and J. Kim, 2010: Influence of tropical cyclone landfalls on spatiotemporal variations in typhoon season rainfall over South China. Adv. Atmos. Sci., 27, 443454, https://doi.org/10.1007/s00376-009-9106-3.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Lenderink, G., and E. van Meijgaard, 2008: Increase in hourly precipitation extremes beyond expectations from temperature changes. Nat. Geosci., 1, 511514, https://doi.org/10.1038/ngeo262.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Li, T, Y. Zeng, and X. Fu, 2018: Velocity ratio of wind-driven rain and its application on a transmission tower subjected to wind and rain loads. J. Perform. Constr. Facil., 32, 04018065, https://doi.org/10.1061/(ASCE)CF.1943-5509.0001210.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Pan, C., X. Wang, L. Liu, D. Wang, and H. Huang, 2019: Characteristics of heavy storms and the scaling relation with air temperature by event process-based analysis in South China. Water, 11, 185, https://doi.org/10.3390/w11020185.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Panwar, A., M. Renner, and A. Kleidon, 2020: Imprints of evaporative conditions and vegetation type in diurnal temperature variations. Hydrol. Earth Syst. Sci., 24, 49234942, https://doi.org/10.5194/hess-24-4923-2020.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Penman, H. L., 1948: Natural evaporation from open water, bare soil and grass. Proc. Roy. Soc. London, 193A, 120145, https://doi.org/10.1098/rspa.1948.0037.

    • Search Google Scholar
    • Export Citation
  • Rooney, G. G., N. Lipzig, and W. Thiery, 2018: Estimating the effect of rainfall on the surface temperature of a tropical lake. Hydrol. Earth Syst. Sci., 22, 63576369, https://doi.org/10.5194/hess-22-6357-2018.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Schroeer, K., and G. Kirchengast, 2018: Sensitivity of extreme precipitation to temperature: the variability of scaling factors from a regional to local perspective. Climate Dyn., 50, 39813994, https://doi.org/10.1007/s00382-017-3857-9.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Srivastava, R. C., 1987: A model of intense downdrafts driven by the melting and evaporation of precipitation. J. Atmos. Sci., 44, 17521774, https://doi.org/10.1175/1520-0469(1987)044<1752:AMOIDD>2.0.CO;2.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Vargas Zeppetello, L. R., D. S. Battisti, and M. B. Baker, 2020: A new look at the variance of summertime temperatures over land. J. Climate, 33, 54655477, https://doi.org/10.1175/JCLI-D-19-0887.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wasko, C., W. T. Lu, and R. Mehrotra, 2018: Relationship of extreme precipitation, dry-bulb temperature, and dew point temperature across Australia. Environ. Res. Lett., 13, 074031, https://doi.org/10.1088/1748-9326/aad135.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Wei, N., Y. Dai, M. Zhang, L. Zhou, D. Ji, S. Zhu, and L. Wang, 2014: Impact of precipitation-induced sensible heat on the simulation of land-surface air temperature. J. Adv. Model. Earth Syst., 6, 13111320, https://doi.org/10.1002/2014MS000322.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Yin, D., M. L. Roderick, G. Leech, F. Sun, and Y. Huang, 2014: The contribution of reduction in evaporative cooling to higher surface air temperatures during drought. Geophys. Res. Lett., 41, 78917897, https://doi.org/10.1002/2014GL062039.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, L., R. Zhang, T. Hong, Y. Zhang, and Q. Meng, 2018: Impact of post-rainfall evaporation from porous roof tiles on building cooling load in subtropical China. Appl. Therm. Eng., 142, 391400, https://doi.org/10.1016/j.applthermaleng.2018.07.033.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhang, S., C. Meurey, and J.-C. Calvet, 2019: Identification of soil-cooling rains in southern France from soil temperature and soil moisture observations. Atmos. Chem. Phys., 19, 50055020, https://doi.org/10.5194/acp-19-5005-2019.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Zhao, S., and S. Yang, 2014: Dynamical prediction of the early season rainfall over southern China by the NCEP Climate Forecast System. Wea. Forecasting, 29, 13911401, https://doi.org/10.1175/WAF-D-14-00012.1.

    • Crossref
    • Search Google Scholar
    • Export Citation
Save