Abstract

The geographic and temporal variability of the surface–3600-m cloud frequency and cloud-base height over the contiguous United States for a 5-yr period (2008–12) and the interannual variations for a 16-yr period (2000–15) are described using information from the Automated Surface Observing System (ASOS) observations. Clouds were separated into four categories by the cloud amount reported by ASOS: few (FEW), scattered (SCT), broken (BKN), and overcast (OVC). The geographic distributions and seasonal and diurnal cycles of the four categories of surface–3600-m cloud frequency have different patterns. Cloud frequency of FEW, SCT, and BKN peaks just after noon, whereas the frequency of OVC peaks in the early morning. However, the geographic distributions and seasonal and diurnal cycles of the four categories of the surface–3600-m cloud-base height are similar. The diurnal cycles of the cloud-base height within the surface–3600-m level present a minimum in the morning and peak in the late afternoon or early evening. Cloud frequency and cloud-base height within this range are closely related to surface air temperature and humidity conditions. From 2000 to 2015, the cloud frequency in the contiguous United States showed a positive trend of 0.28% yr−1 while the cloud-base height showed a negative trend of −4 m yr−1 for the surface–3600-m level, accompanied with a positive trend of precipitation days (0.14 days yr−1). Moreover, the increase of cloud frequency and the decrease of cloud-base height were most obvious in winter in the eastern half of the contiguous United States.

1. Introduction

The global climate is largely determined by Earth’s energy budget (Trenberth et al. 2009). The absorbed solar shortwave radiation heats the planet while the emitted longwave radiation cools it. Clouds play a major role in regulating the energy flows into and out of the earth–atmosphere system. They cool the earth by reflecting solar radiation into space. They warm the earth by absorbing thermal infrared radiation from below and reemitting it back to space and the surface. During daytime, whether a cloud cools or warms the surface depends on the cloud type associated with the cloud height (Lee et al. 1997). At night, the impact of clouds on solar shortwave radiation is zero; thus, all types of clouds have a warming effect. As a result, the general magnitude and sign of the cloud radiative effect depends on the cloud amount, the cloud height (base and top), and the time of occurrence during a day (Ramanathan et al. 1989; IPCC 2013).

Our current knowledge of cloud processes is limited; thus, cloud represents one of the largest uncertainties in simulating past climate change and predicting future climate change (Cess et al. 1990; Fasullo and Trenberth 2012; Sherwood et al. 2014). Observations provide an opportunity to assess how clouds contribute and respond to climate change. The most common sources of cloud observations are from satellites and surface weather stations.

Passive satellite visible and thermal infrared sensors principally derive information regarding cloud amount, cloud optical thickness, cloud-top pressure, and other cloud properties from radiance measurements (Rossow and Schiffer 1991). These sensors have the advantage of observing cloud-top height over the entire globe (Forsythe et al. 2000). However, these satellite sensors cannot pierce through clouds; thus, assumptions on cloud thickness are required (Welch et al. 2008) to determine the cloud-base height (Hutchison 2002; Kokhanovsky and Rozanov 2005), and these assumptions introduce large uncertainties (Welch et al. 2008). Moreover, many of these satellite sensors have different spectral channels for day and night retrievals, which introduce intrinsic errors to the observations of cloud diurnal variations. New airborne active sensors, such as the Cloud Profiling Radar on board the CloudSat satellite and Cloud–Aerosol Lidar with Orthogonal Polarization on board the CALIPSO satellite, can provide multiple cloud-layer-top and cloud-layer-base heights (Kim et al. 2011). However, the samples from these sensors are spatially sparse.

Traditionally, synoptic weather reports via the global telecommunication system (GTS) from surface stations contain cloud information such as total cloud cover, the predominate cloud types for three layers (low, middle, and high), the cloud amount, and the cloud-base height of the lowest layer, according to World Meteorological Organization (WMO) codes (WMO 1974). Nevertheless, the ways cloud observations were made vary from country to country (Sun et al. 2001). For example, in the United States, historically human observers were allowed to report up to four or more cloud layers and the cloud-base height of each layer; for augmentation by human observers at the Automated Surface Observing System (ASOS) sites, up to six cloud layers and the cloud-base height of each were provided. Those rich-information observations were archived, but modifications and abridgements were made on-site to generate transmitted WMO synoptic data over GTS.

Routine observations of clouds have long-term historic records. The cloud types they report have strong indications for cloud dynamic processes. However, these datasets that generally based on human visual observations are somewhat subjective and have potential inhomogeneity related to changes in observational practice (Dai et al. 2006; Free and Sun 2013; Henderson-Sellers 1992). Moreover, the lack of illumination at night may introduce inconsistencies when analyzing the diurnal cycle.

Radiosonde data can also be used to infer cloud information, such as cloud vertical structure and cloud frequency, based on profiles of temperature, relative humidity, and pressure (Costa-Surós et al. 2014; Chernykh et al. 2001; Minnis et al. 2005; Wang and Rossow 1995; Zhang et al. 2012). However, these data are restricted by radiosonde observation frequencies (usually twice daily) and therefore are not able to depict the diurnal changes of cloud properties.

Previous studies have analyzed the climatology of cloud properties using ground-based and remote sensing methods. Stubenrauch et al. (2013) accessed 12 state-of-the-art satellite cloud datasets and analyzed the latitudinal variations, seasonal and diurnal cycles, and interannual variability of cloud amount, cloud-top location, and other cloud properties. However, the climatology of cloud-base height has not been properly analyzed via satellite observations. Warren et al. (1986, 1988) produced a database from surface land station and ship measurements and further analyzed the long-term and seasonal and diurnal variations of cloud cover, cloud types, and cloud-base height based on this database (Eastman and Warren 2014; Eastman et al. 2011; Warren et al. 2007). However, these surface databases have only four or eight daily observations and may be influenced by poor illumination at night.

For regions of the United States, studies have focused on the total cloud cover, cloud amounts, and cloud-ceiling height changes based on different ground-based and remote sensing methods (Dai et al. 2006; Free and Sun 2014; Sun 2003; Sun et al. 2015; Sun and Groisman 2004; Sun et al. 2001, 2007). Long-term interannual variations of these parameters are generally based on daytime cloud observations only.

Studies have shown that cloud microphysical characteristics (e.g., cloud particle size, ice/water content, and cloud infrared absorption coefficient) and macrophysical characteristics (e.g., low cloud amount, high cloud amount, and high cloud-top height) can substantially change (Bender et al. 2012; Brient and Bony 2012; Platt 1989; Zelinka and Hartmann 2010), which in turn introduce important climatic feedback. Cloud vertical structures (i.e., the cloud-top and cloud-base height) are also sensitive to climate change (Chepfer et al. 2014; Ockert-Bell and Hartmann 1992; Williams et al. 2015). Cloud-base height is an important feature that describes the impact of clouds in a changing climate. For example, its variations directly impact cloud radiative properties and thus affect the global radiation balance (Slingo and Slingo 1988; Viúdez-Mora et al. 2015). Moreover, cloud-base height is important for weather forecasting and aviation control (Ellrod and Gultepe 2007; Mittermaier 2012; Vislocky and Fritsch 1997). Hence, there is a need to better monitor and understand cloud-base height variability.

Beginning from the early 1990s, ASOS has gradually replaced the manned weather stations in North America as well as in some other parts of the world. ASOS contains a set of automated instruments for observing sky condition (i.e., cloud amount and cloud-base height), as well as other variables such as temperature, dewpoint, precipitation, wind, visibility, and pressure (NWS 1998).

The ceilometer introduced by ASOS has become one of the most common sources of surface cloud observations in the United States. Ceilometers from the widely distributed ASOS stations provide continuous data with high quality for the study of diurnal, seasonal, and interannual variations of cloud frequency and cloud-base height over a large region. In addition, the simultaneous observations of temperature and precipitation from ASOS can be used to examine the variability of cloud frequency and cloud-base height.

This paper uses cloud frequency and cloud-base height data collected from ASOS ceilometers from 2000 to 2015 to analyze the statistics of cloud frequency and cloud-base height over the contiguous United States. The data and methods are briefly discussed in section 2; the seasonal and diurnal variations and interannual variability results are presented in section 3; an analysis of temporal relationships between cloud properties (cloud frequency and cloud-base height) and several physically related parameters is shown in section 4; and the concluding remarks and a discussion of the results are presented in section 5.

2. Data

The data used in this study are from the ASOS 5-min dataset (NWS 1998). ASOS employs standard laser ceilometers (Vaisala models CT12K or CL31; Vaisala Inc., Woburn, Massachusetts) to observe the sky conditions. A ceilometer sends laser pulses vertically upward toward the sky and determines whether the return signals are from cloud bases (Fig. 1). Knowing the speed of light, cloud-base height can be obtained by the time delay between the launch of the laser pulse and the detection of the backscatter signal. The ASOS sky condition algorithm (NWS 1998) processes the sensor signals into 30-s sample “hits.” The algorithm calculates the cloud amount and cloud-base height values using the most recent 30 min of the 30-s data samples, and data in the last 10 min are double weighted.

Fig. 1.

Example of (left) an ASOS meteorological station and (right) the ceilometer working principle (station image is from http://www.wrcc.dri.edu/images/Station_pics/Utah/milford_ut_asos.jpg).

Fig. 1.

Example of (left) an ASOS meteorological station and (right) the ceilometer working principle (station image is from http://www.wrcc.dri.edu/images/Station_pics/Utah/milford_ut_asos.jpg).

ASOS provides sky condition information, including whether the sky is clear or cloudy. When the sky is cloudy, the information will also include the cloud amount and cloud-base height for at most three layers and up to 12 000 ft (approximately 3600 m) above the surface. With the exception of clear-sky (CLR) conditions, the cloud amounts reported by ASOS have four values: few (FEW), scattered (SCT), broken (BKN), and overcast (OVC). These values are determined by the ratio of the number of cloud hits to the total number of hits in the range of 5%–25%, 25%–50%, 50%–87%, and 87%–100% (NWS 1998).

The precision of cloud amount observed by ASOS is low; therefore, we did not use the cloud amount directly. In this research, we separated clouds into four categories by the four cloud amounts (FEW, SCT, BKN, and OVC). Subsequently, we divided the number of the cases when ASOS reported these four categories by the total number of valid observations to calculate the cloud frequency in each category; the total cloud frequency was calculated by summing the frequencies of the four categories.

In this research, we analyzed only the lowest layer of the cloud-base height because ceilometers have been reported to exhibit inherent limitations (Costa-Surós et al. 2013) that prevent them from completely describing the upper layers of clouds when multiple layers of clouds occur.

The data used in this research underwent rigorous quality control. First, we determined the valid days; we rejected the data recorded on the days when the daily valid data numbers were less than 50% of the total number of daily observations. For the analysis of interannual variations and trends, we selected 186 stations from ASOS that had abundant data for the period from 2000 to 2015 and determined whether the data met the following criterion: sufficient data for each season (more than 45 valid days in a season) for more than 10 yr within the 16-yr period. Finally, because the number of ASOS stations has increased in recent decades, we selected 719 stations with sufficient data for all seasons from 2008 to 2012 to calculate the climatology of the clouds and determined whether the data met the following criterion: sufficient data for each season (more than 45 valid days in a season) for more than 3 yr of the 5-yr period. The selected stations were evenly distributed throughout the contiguous United States.

ASOS uses a variety of sensors to identify different types of weather (e.g., rain, snow, and frozen rain), identify the obstructions to vision (e.g., mist, fog, and haze), and infer the intensity of precipitation. We also checked the precipitation, obstruction, temperature, and dewpoint datasets from the ASOS 5-min data to support the analyses of cloud frequency and cloud-base height from laser ceilometers.

Precipitation and fog may attenuate the signal of ceilometers. According to the ceilometer technical manuals (Vaisala 1989), CT12K ceilometers perform algorithms to compensate for the attenuation of the cloud-base signal caused by obstructions, such as precipitation and fog. When the cloud base is completely obscured by these obstructions, the algorithm will calculate a height of vertical visibility (VV) instead of a cloud-base height. During 2008–12, approximately 18.9%, 7.2%, 89.2%, and 1.5% of the VV measurements occurred with precipitation, mist, fog, and haze, respectively. We excluded the VV measurements under haze and included the rest of the VV measurements that occurred with precipitation, mist, and fog in the category of OVC. The annual mean occurrence frequency of VV measurements was 0.9% for all 719 stations during 2008–12. It was not comparable with annual mean cloud frequency (39%). Therefore, in general, including VV had little influence on the distributions and variations of cloud frequency and cloud-base height.

The macrocharacteristics of clouds are mainly dependent on the atmospheric temperature and humidity condition. To better analyze the geographic differences, based on the Köppen–Geiger climate classification (Kottek et al. 2006), we merged similar climate regions that have the same main climate types and precipitation types (i.e., have the same first and second letter of Köppen–Geiger climate classification) and separated the contiguous United States into six climate zones (Fig. 2; Table 1). The hot-summer Mediterranean climate (Csa) and warm-summer Mediterranean climate (Csb) are merged into one climate zone because they have obvious features of Mediterranean climate. The humid continental climate–dry cool summer (Dsb) and continental subarctic–cold dry summer (Dsc), which together contain only two stations, are also counted as the Mediterranean climate zone in this research because they both have dry summers. The tropical savanna climate (Aw) and tropical monsoon climate (Am) were merged into one climate zone because they both have dry winters, and the area of the equatorial climate in the contiguous United States is not large enough to separate it into two climate zones.

Fig. 2.

Map of six climate zones of the contiguous United States used in this study (see Table 1). The climate zones are generated based on the Köppen–Geiger climate classification. The dots are the 186 stations selected to analyze the trends from 2000 to 2015. The stations are evenly distributed.

Fig. 2.

Map of six climate zones of the contiguous United States used in this study (see Table 1). The climate zones are generated based on the Köppen–Geiger climate classification. The dots are the 186 stations selected to analyze the trends from 2000 to 2015. The stations are evenly distributed.

Table 1.

Climate zones used in this study, the corresponding climate regions of the Köppen–Geiger climate classification for each climate zone, and the number of stations selected for analyzing interannual variability and for analyzing seasonal and diurnal variability. The percentages in parentheses refer to the ratios of the numbers of selected stations to the total numbers of stations in each climate zone and over the entire contiguous United States.

Climate zones used in this study, the corresponding climate regions of the Köppen–Geiger climate classification for each climate zone, and the number of stations selected for analyzing interannual variability and for analyzing seasonal and diurnal variability. The percentages in parentheses refer to the ratios of the numbers of selected stations to the total numbers of stations in each climate zone and over the entire contiguous United States.
Climate zones used in this study, the corresponding climate regions of the Köppen–Geiger climate classification for each climate zone, and the number of stations selected for analyzing interannual variability and for analyzing seasonal and diurnal variability. The percentages in parentheses refer to the ratios of the numbers of selected stations to the total numbers of stations in each climate zone and over the entire contiguous United States.

3. Results

In this section, we show the seasonal, diurnal, and interannual variability of cloud frequency and cloud-base height for the surface–3600-m layer over the contiguous United States as well as their relationships with precipitation.

a. Seasonal distribution of cloud frequency

The geographical distributions of the average surface–3600-m total cloud frequency for 719 stations recorded during different seasons over the contiguous United States for five years (2008–12) are shown in Fig. 3. The results indicate that seasonal variability and geographical differences occurred. The total cloud frequency is higher (over 40%) in the northwestern coastal region, the Great Lakes region, and southern Florida throughout the year (Fig. 3). In arid areas, the total cloud frequency is relatively low throughout the year. The annual average surface–3600-m total cloud frequencies for the arid steppe climate zone and the arid desert climate zone are 27% and 15%, respectively, whereas those for the Mediterranean climate zone, the snow climate zone, the warm and humid climate zone, and the equatorial climate zone are 39%, 45%, 39%, and 50%, respectively. As shown in Fig. 4a, the total cloud frequency is generally higher in winter and lower in summer. In winter, the average total cloud frequency for all 719 stations is 43%, whereas the average is 33% in summer.

Fig. 3.

Mean total cloud frequency (%) for the surface–3600-m layer of 719 stations over the contiguous United States in (a) spring (March–May), (b) summer (June–August), (c) autumn (September–November), and (d) winter (December–February) during the period from 2008 to 2012.

Fig. 3.

Mean total cloud frequency (%) for the surface–3600-m layer of 719 stations over the contiguous United States in (a) spring (March–May), (b) summer (June–August), (c) autumn (September–November), and (d) winter (December–February) during the period from 2008 to 2012.

Fig. 4.

Mean seasonal and annual (ANN) cloud frequency (%) for the surface–3600-m layer in the different climate zones during the period from 2008 to 2012: (a) total cloud frequency, (b) cloud frequency of FEW, (c) cloud frequency of SCT, (d) cloud frequency of BKN, (e) cloud frequency of OVC, and (f) cloud frequency during precipitation. Climate zones I–VI are as in Fig. 2. Error bars designate the standard error of the value in a climate zone and season. Cloud frequency during precipitation in (f) is calculated by dividing the number of cases when ASOS reported precipitation and clouds by the total number of precipitation cases. The areas with slanted lines indicate the frequency (%) of the VV observations during precipitation.

Fig. 4.

Mean seasonal and annual (ANN) cloud frequency (%) for the surface–3600-m layer in the different climate zones during the period from 2008 to 2012: (a) total cloud frequency, (b) cloud frequency of FEW, (c) cloud frequency of SCT, (d) cloud frequency of BKN, (e) cloud frequency of OVC, and (f) cloud frequency during precipitation. Climate zones I–VI are as in Fig. 2. Error bars designate the standard error of the value in a climate zone and season. Cloud frequency during precipitation in (f) is calculated by dividing the number of cases when ASOS reported precipitation and clouds by the total number of precipitation cases. The areas with slanted lines indicate the frequency (%) of the VV observations during precipitation.

In Figs. 4b–e, the cloud frequencies for the four categories (FEW, SCT, BKN, and OVC) within the surface–3600-m layer have different seasonal and geographical distributions. The cloud frequencies of FEW and SCT in the equatorial climate zone located in south Florida are much higher than in the other climate zones. The cloud frequencies of FEW and SCT are generally higher in summer than in winter. The cloud frequency of BKN is also higher in the equatorial climate zone, but it is higher in winter than in summer in the equatorial climate zone, the arid steppe climate zone, and the arid desert climate zone. The cloud frequency of OVC is higher in the snow climate zone in the northeast and in the Mediterranean climate zone on the West Coast of the contiguous United States but is lower in the equatorial climate zone. The cloud frequency of OVC is much higher in winter than in summer.

Figure 4f shows the surface–3600-m cloud occurrence frequency during precipitation occurrence. It is calculated by dividing the number of cases when ASOS detected precipitation as well as clouds by the total number of precipitation cases. The areas with slanted lines indicate the frequency of VV observations during precipitation. Although the VV frequency is not compatible with cloud frequency, it accounts for an impressive proportion of precipitation. VV during precipitation occurred more in winter and in cold areas. The VV frequency during precipitation in the equatorial climate zone is very low and therefore not apparent in Fig. 4f. The cloud frequency during precipitation is under 100% mainly because a small portion of precipitation events are generated by higher-level clouds that ceilometers do not observe. Precipitation is also transported by wind and can occur when clouds are not right above the ceilometers, although this precipitation makes only a slight contribution to the total. The total cloud frequencies during precipitation are over 90% in many seasons and many parts of the contiguous United States, which confirms that clouds below 3600 m play an important role in precipitation. Not all clouds that produce precipitation are detected by ASOS ceilometers. In the arid desert zone and the equatorial climate zone, especially in summer, 20%–40% of the precipitation occurs without the detection of clouds by the CT12K ceilometers from ASOS.

b. Seasonal distribution of the cloud-base height

Geographical distributions of the average surface–3600-m cloud-base height for all categories of clouds detected by ceilometers are shown in Fig. 5; it displays data collected from 719 stations over the contiguous United States from 2008 to 2012 during different seasons. The regional means of the cloud-base height for all categories of clouds, cloud-base height for each of the four categories, and cloud-base height during precipitation for the surface–3600-m layer in the different climate zones are shown in Figs. 6a–f, respectively.

Fig. 5.

As in Fig.3, but for mean cloud-base height (m).

Fig. 5.

As in Fig.3, but for mean cloud-base height (m).

Fig. 6.

Mean seasonal and annual cloud-base height (m) for the surface–3600-m layer in different climate zones in period from 2008 to 2012: (a) cloud-base height of all clouds, (b) cloud-base height of FEW, (c) cloud-base height of SCT, (d) cloud-base height of BKN, (e) cloud-base height of OVC, (f) cloud-base height during precipitation. Climate zones I–VI are as in Fig. 2. Error bars designate the standard error of the value in a climate zone and season.

Fig. 6.

Mean seasonal and annual cloud-base height (m) for the surface–3600-m layer in different climate zones in period from 2008 to 2012: (a) cloud-base height of all clouds, (b) cloud-base height of FEW, (c) cloud-base height of SCT, (d) cloud-base height of BKN, (e) cloud-base height of OVC, (f) cloud-base height during precipitation. Climate zones I–VI are as in Fig. 2. Error bars designate the standard error of the value in a climate zone and season.

In general, the cloud-base height is lower in the east and higher in the west except on the West Coast of the contiguous United States (Fig. 5). In the arid steppe climate zone and the arid desert climate zone in the west, the annual average surface–3600-m cloud-base heights are 1884 and 2390 m, respectively. In the snow climate zone, the warm and humid climate zone, and the equatorial climate zone, which occur primarily in the eastern part of the contiguous United States, the annual average surface–3600-m cloud-base heights are 1485, 1373, and 1175 m, respectively. Along the West Coast of the contiguous United States, the cloud-base height is mainly less than 1200 m during every season (Fig. 5), which is due to the high frequencies of stratus, stratocumulus, and fog along the west coastal area (Leipper 1994; Warren et al. 1986). Lower values for the cloud-base heights are also observed in the coastal areas along the Gulf of Mexico and the Great Lakes area. As shown in Fig. 6a, most of the regions of the contiguous United States have higher cloud-base heights in summer than in winter except in the equatorial climate zone.

The average cloud-base heights of the four categories of FEW, SCT, BKN, and OVC for the surface–3600-m layer are both higher in arid areas such as the arid steppe climate zone and the arid desert climate zone and lower in more humid areas such as the tropical climate zone and the warm and humid climate zone (Figs. 6b–e). Their geographical distributions are similar and not shown in the figure. The annual average surface–3600-m cloud-base heights of all the 719 stations during 2008–12 for the four categories (FEW, SCT, BKN, and OVC) are 1941, 1802, 1672, and 1212 m, respectively. The mean cloud-base height becomes lower when cloud amount increases.

The annual average cloud-base height during precipitation at all the 719 stations during 2008–12 is 851 m. However, it is much higher in the arid steppe climate zone and the arid desert climate zone, especially in summer, when the average cloud-base heights during precipitation reach about 2000 m (Fig. 6f).

c. Diurnal variations of cloud frequency and cloud-base height

For most stations within a climate zone, the patterns of the diurnal cycle of cloud frequency and cloud-base height are similar. Therefore, we analyzed the average diurnal cycle of cloud frequency and cloud-base height from all stations in every climate zone. All the diurnal cycles are obtained based on local standard time (LST).

Figure 7 shows the diurnal cycles of cloud frequency within the surface–3600-m layer of the four categories (FEW, SCT, BKN, and OVC) for different climate zones. The diurnal frequency of FEW, SCT, and BKN is allied to the diurnal cycle of cumulus clouds (e.g., Eastman and Warren 2014, their Fig. 10). The occurrence frequencies of these types of clouds begin to increase in the early morning as the sun rises and the solar radiation heats the surface and drives convection in the boundary layer. The value peaks just after noon and decreases in the afternoon. Except in summer in the equatorial climate shown in Fig. 7d(6), the diurnal cycles of cloud frequencies of OVC generally peak in the morning, which corresponds to the diurnal cycle of stratiform clouds (e.g., Eastman and Warren 2014, their Fig. 10). The peak in the morning is associated with the diurnal peak frequency of drizzle and nonshowery precipitation in the morning (e.g., Dai 2001, their Fig. 6).

Fig. 7.

Diurnal variations of cloud frequency (%) of (a) FEW, (b) SCT, (c) BKN, and (d) OVC within the surface–3600-m layer for six climate zones (I–VI, as in Fig. 2) during 2008–12.

Fig. 7.

Diurnal variations of cloud frequency (%) of (a) FEW, (b) SCT, (c) BKN, and (d) OVC within the surface–3600-m layer for six climate zones (I–VI, as in Fig. 2) during 2008–12.

The diurnal cycles of cloud-base heights within the surface–3600-m layer have similar patterns in different climate zones for different categories (Fig. 8). Daily minima generally appear around 0900 LST (0700–1100 LST) in the morning. This time corresponds to the cumuliform clouds’ onset time. The newly and vastly generated cumulus clouds in the lower atmosphere drive down the average cloud-base height at that time. Then, clouds are lifted during the daytime and reach peaks in the late afternoon or early evening. As shown in Fig. 8, in general, the minima of the diurnal cycles of cloud-base height occur earlier and the peak values occur later during the day in summer than in winter because sunrise is earlier and sunset is later in summer. The exceptions in the equatorial climate zone may be due to few samples of stations as well as the small difference between summer and winter in a tropical area. The exceptions in the arid desert climate zone may be due to the fact that some of the clouds higher than 3600 m are not taken into account in the diurnal cycles especially in summer.

Fig. 8.

Diurnal variations of cloud-base height (m) of (a) FEW, (b) SCT, (c) BKN, and (d) OVC within the surface–3600-m layer and (e) precipitation frequencies for six climate zones (I–VI, as in Fig. 2) during 2008–12. The asterisks in the figure represent the cloud-base height value when the value of temperature minus dewpoint temperature reaches its diurnal peak value.

Fig. 8.

Diurnal variations of cloud-base height (m) of (a) FEW, (b) SCT, (c) BKN, and (d) OVC within the surface–3600-m layer and (e) precipitation frequencies for six climate zones (I–VI, as in Fig. 2) during 2008–12. The asterisks in the figure represent the cloud-base height value when the value of temperature minus dewpoint temperature reaches its diurnal peak value.

The contrasts between surface temperature and dewpoint are used to estimate the lifting condensation level: the larger the contrasts, the higher the lifting condensation level, and the higher the convective cloud-base height (Craven et al. 2002). The cloud-base height peaks later than the time when the contrast between the temperature and dewpoint reaches its daily maximum in the midafternoon (shown by the asterisks in Fig. 8).

d. Interannual variations of cloud frequency and cloud-base height

We calculated the linear trends of surface–3600-m total cloud frequency and cloud-base height using least squares regressions (Table 2; Figs. 9 and 10). As shown in Table 2, from 2000 to 2015, the total cloud frequency increased while the cloud-base height decreased. These changes in cloud frequency and cloud-base height are especially obvious in winter, which showed an average increase of 6.6% in surface–3600-m cloud frequency and an average decrease of 149 m in surface–3600-m cloud-base height for all 186 stations during the 16 studied years. As shown in Fig. 9, the increasing cloud frequency trend in winter generally occurs in the eastern part of the United States. In winter, over 50% of stations have significant (i.e., passed the two-tailed Student’s t test at the 90% level) increasing trends of cloud frequency in the snow climate zone and the warm and humid climate zone, while cloud-base height shows a decreasing trend over the eastern part of the contiguous United States (Fig. 10).

Table 2.

Seasonal and annual trends of cloud-base height (m yr−1), cloud frequency (% yr−1), and precipitation days (days yr−1) in different climate zones from 2000 to 2015. Trends in boldface are statistically significant at the 0.1 level.

Seasonal and annual trends of cloud-base height (m yr−1), cloud frequency (% yr−1), and precipitation days (days yr−1) in different climate zones from 2000 to 2015. Trends in boldface are statistically significant at the 0.1 level.
Seasonal and annual trends of cloud-base height (m yr−1), cloud frequency (% yr−1), and precipitation days (days yr−1) in different climate zones from 2000 to 2015. Trends in boldface are statistically significant at the 0.1 level.
Fig. 9.

Linear trends of total cloud frequency (% yr−1) for the surface–3600-m layer of 186 stations over the contiguous United States of (a) spring, (b) summer, (c) autumn, and (d) winter for the period from 2000 to 2015. Trends are calculated using least squares regression. The dots indicate that the trend of a station has passed the two-tailed Student’s t test at the 90% level, and the triangles indicate that the trend fails to pass.

Fig. 9.

Linear trends of total cloud frequency (% yr−1) for the surface–3600-m layer of 186 stations over the contiguous United States of (a) spring, (b) summer, (c) autumn, and (d) winter for the period from 2000 to 2015. Trends are calculated using least squares regression. The dots indicate that the trend of a station has passed the two-tailed Student’s t test at the 90% level, and the triangles indicate that the trend fails to pass.

Fig. 10.

As in Fig. 9, but for linear trends of cloud-base height (m yr−1).

Fig. 10.

As in Fig. 9, but for linear trends of cloud-base height (m yr−1).

The correlations between the trends during daytime and nighttime were tested. Figure 11 shows the scatterplots of nighttime and daytime trends of surface–3600-m total cloud frequency as well as cloud-base height. The trends of daytime and nighttime total cloud frequency seem to be consistent, with a correlation coefficient of 0.75. The same conclusion is also found for the trends of cloud-base height, with a correlation coefficient of 0.66 between daytime and nighttime trends.

Fig. 11.

(a) Scatterplots of linear trends of nighttime surface–3600-m total cloud frequency (% yr−1) as a function of linear trends of daytime surface–3600-m total cloud frequency (% yr−1). (b) Scatterplots of linear trends of nighttime surface–3600-m cloud-base height (m yr−1) as a function of linear trends of daytime surface–3600-m cloud-base height (m yr−1). Each dot represents a station for one season. The red line is the best-fit line calculated using least squares regression.

Fig. 11.

(a) Scatterplots of linear trends of nighttime surface–3600-m total cloud frequency (% yr−1) as a function of linear trends of daytime surface–3600-m total cloud frequency (% yr−1). (b) Scatterplots of linear trends of nighttime surface–3600-m cloud-base height (m yr−1) as a function of linear trends of daytime surface–3600-m cloud-base height (m yr−1). Each dot represents a station for one season. The red line is the best-fit line calculated using least squares regression.

The precipitation days that have precipitation for more than half an hour were summed within a season, and then the trends of precipitation days were calculated for a season (Table 2; Fig. 12). The results indicate that from 2000 to 2015, precipitation days increased for a large part of the contiguous United States. An increase of 2.4 precipitation days in winter occurred during the 16 years from 2000 to 2015. Figure 13 illustrates the scatterplots between the significant trends of surface–3600-m total cloud frequency and the trends of precipitation days. The figure shows that most of the stations with significant positive (negative) trends of cloud frequency also show positive (negative) trends of precipitation days. The correlation between the trends of precipitation days and total cloud frequency in summer is weak. In winter, the correlation between the trends of cloud frequency and precipitation days is 0.50 (p < 0.05), and the plots are mostly located in the first quadrant, which indicates the close connection between the increasing cloud frequency and increasing snow and rainfall days in winter.

Fig. 12.

Linear trends of precipitation days (days yr−1) for 186 stations over the contiguous United States of (a) spring, (b) summer, (c) autumn, and (d) winter for the period from 2000 to 2015. Trends are calculated using least squares regression. The dots indicate that the trend of a station has passed the two-tailed Student’s t test at the 90% level, and the triangles indicate that the trend fails to pass.

Fig. 12.

Linear trends of precipitation days (days yr−1) for 186 stations over the contiguous United States of (a) spring, (b) summer, (c) autumn, and (d) winter for the period from 2000 to 2015. Trends are calculated using least squares regression. The dots indicate that the trend of a station has passed the two-tailed Student’s t test at the 90% level, and the triangles indicate that the trend fails to pass.

Fig. 13.

Scatterplots of linear trends of precipitation days (days yr−1) as a function of linear trends of surface–3600-m total cloud frequency (% yr−1) when the linear trends of surface–3600-m total cloud frequency (% yr−1) is significant in different seasons: (a) spring, (b) summer, (c) autumn, and (d) winter. Each dot represents a station for a season. The different colors of the dots indicate the climate zones of the stations corresponding to Fig. 2. The blue line is the best-fit line calculated using least squares regression.

Fig. 13.

Scatterplots of linear trends of precipitation days (days yr−1) as a function of linear trends of surface–3600-m total cloud frequency (% yr−1) when the linear trends of surface–3600-m total cloud frequency (% yr−1) is significant in different seasons: (a) spring, (b) summer, (c) autumn, and (d) winter. Each dot represents a station for a season. The different colors of the dots indicate the climate zones of the stations corresponding to Fig. 2. The blue line is the best-fit line calculated using least squares regression.

4. Temporal relationships with physically related parameters at surface

Cloud datasets in the United States are confronted with inhomogeneity problems because of the changes to observation practices and the introduction of ASOS (Free and Sun 2013). Therefore, many studies tested the long-term variations of cloud datasets with physically related surface air parameters to identify a homogeneous climate record for cloud (Free and Sun 2014; Sun and Groisman 2004; Sun et al. 2000, 2007). Interannual variations of cloud macrophysical properties have been examined and proved to be closely related to parameters such as surface air temperature Tsfc, relative humidity RHsfc, and diurnal temperature range (DTR) (Dai et al. 2006; Sun et al. 2000, 2007). However, most of their analyses are generally calculated based on area-averaged cloud macrophysical properties and area-averaged surface air parameters in a certain country or a region. With high density of ASOS stations, we want to test the relationships at every station to see if there are apparent spatial variations over the contiguous United States.

We calculated the correlations between the detrended monthly average surface–3600-m total cloud frequency and detrended monthly average RHsfc, DTR, and Tsfc as well as the correlations between the detrended monthly average surface–3600-m cloud-base height and detrended monthly average RHsfc, DTR, and Tsfc at every station over the contiguous United States during 2000–15 in different seasons. The RHsfc was calculated from temperature and dewpoint observations from the ASOS 5-min dataset and the DTR was determined by daily maximum temperature minus the minimum temperature from the ASOS 5-min dataset.

In general, the correlations between cloud-base height and RHsfc, DTR, and Tsfc (Figs. 14b,d,f and 15b,d,f) are opposite to the correlations between cloud frequency and these parameters (Figs. 14a,c,e and 15a,c,e), which is due to the negative correlation of cloud frequency and cloud-base height within the surface–3600-m layer.

Fig. 14.

Correlations between detrended monthly mean total cloud frequency and (a) detrended monthly mean relative humidity, (c) DTR, and (e) Tsfc; correlations between detrended monthly mean cloud-base height and (b) detrended monthly mean relative humidity, (d) DTR, and (f) Tsfc for 186 stations during summer (June–August) from 2000 to 2015 over the contiguous United States. The dots indicate that correlation is significant, and the triangles indicate that the correlation is not significant at the 0.05 level.

Fig. 14.

Correlations between detrended monthly mean total cloud frequency and (a) detrended monthly mean relative humidity, (c) DTR, and (e) Tsfc; correlations between detrended monthly mean cloud-base height and (b) detrended monthly mean relative humidity, (d) DTR, and (f) Tsfc for 186 stations during summer (June–August) from 2000 to 2015 over the contiguous United States. The dots indicate that correlation is significant, and the triangles indicate that the correlation is not significant at the 0.05 level.

Fig. 15.

As in Fig. 14, but during winter (December–February).

Fig. 15.

As in Fig. 14, but during winter (December–February).

When the RHsfc is higher, there is a high possibility for more water vapor condensed as clouds and resulting in more cloud amount. The positive correlations between total cloud frequency and RHsfc are consistent among all 186 stations in both summer and winter except for a very few stations that are not significant or have negative correlations.

Nearly all 186 stations over the contiguous United States present negative correlations between cloud frequency and DTR in either summer or winter seasons. The surface–3600-m layer contains mainly low clouds, which are especially efficient in reducing the daily maximum temperature and DTR for their high capability in reflecting solar radiation. As shown in Figs. 14b,d and 15b,d, the correlations between cloud-base height and RHsfc as well as DTR are not significant in many stations along the West Coast of the contiguous United States. This is likely due to the decoupling of clouds and surface conditions (i.e., the clouds resulting from the land and sea breeze).

The correlations between cloud frequency and Tsfc are different between summer and winter. In summer, nearly all the stations of the entire contiguous United States present negative correlations between total cloud frequency and Tsfc as well as positive correlations between cloud-base height and Tsfc (Figs. 14e,f). However, in winter, many stations in the east present positive correlations between total cloud frequency and Tsfc (Fig. 15e). The geographical difference can also be found for the correlations between cloud-base height and Tsfc (Fig. 15f).

5. Conclusions and discussion

Based on ASOS observations of clouds, we analyzed the seasonal, diurnal, and interannual variations in cloud frequency and cloud-base height over the contiguous United States. The results show that ceilometers from ASOS can monitor clouds with high spatial and temporal resolution within their measurement range.

We separated clouds into four categories by their cloud amounts (FEW, SCT, BKN, and OVC). The cloud frequencies of the four categories show different geographical distributions and seasonal and diurnal variations. The categories of FEW, SCT, and BKN may contain more cumuliform clouds, whereas OVC may contain more stratiform clouds according to the result. The mean cloud-base height decreases as cloud amount increase from FEW to OVC. However, the geographical distribution and seasonal and diurnal variations of cloud-base height have similar features for the four categories.

Our results showed that the peak of cloud-base height lags the peak of the difference between temperature and dewpoint in the diurnal cycle, which may be explained by convection constantly lifting the clouds regardless of whether convection is increasing and decreasing during the daytime. Another explanation may be that the convective precipitation that has higher frequencies in the late afternoon [Figs. 8e(1)–(6)] dissipates the lower portions of the strong vertically developed clouds and keeps the upper portions of clouds at a higher level.

The surface–3600-m layer contains mainly low clouds, which are connected to surface parameters through turbulence. The surface–3600-m cloud frequency and cloud-base height are closely related to RHsfc as well as DTR, and the correlations are consistent in most of the stations over the contiguous United States in different seasons. The relationship between cloud frequency and Tsfc is negative in summer because of the cooling effect of low clouds. In winter, cloud frequency is positive in relation to Tsfc in the eastern part of the contiguous United States. A possible explanation for the inconformity in winter may be the different synoptic situations during winter. Cold outbreaks containing dry and cold arctic air routinely plunge south into many parts of the central and eastern contiguous United States (Kunkel et al. 2013), which results in lower temperatures and more clear days, causing a positive relationship between cloud frequency and Tsfc. Sun et al. (2000) found that cloud cover and Tsfc were positively correlated at midlatitudes in winter because changes in snow cover affect Tsfc by altering the surface albedo. This is consistent with our result.

From 2000 to 2015, the average total cloud frequency in the surface–3600-m layer over the contiguous United States increased while the cloud-base height decreased and was accompanied with an increase of precipitation days. The increase of cloud frequency and decrease of cloud-base height are very noticeable in the eastern part of the contiguous United States in the winter season. Previous research on long-term trends of cloud amount and cloud-base height was generally based on daytime observations. Our results show that the nighttime trends of cloud frequency and cloud-base height are consistent with the corresponding daytime trends. Sun et al. (2015) reported that cloud amount decreased at a trend of −0.40% from 1982 to 2007, and Sun et al. (2007) showed that cloud-ceiling height in the surface–3600-m level significantly increased after the early 1970s until 2003. Cloud-ceiling height is defined as the base height of the lowest layer when BKN or OVC of opaque clouds are reported. Based on this definition, we calculated the trend of cloud-ceiling height (−2 m yr−1) over the contiguous United States during 2000–2015 from ASOS ceilometer observation. This is consistent with the negative trend of cloud-base height we calculated that is also statistically significant at the 0.1 level. Combining the result of this study and earlier studies suggests that decadal variabilities in cloud properties have been ongoing over the contiguous United States: decreasing cloud amount from the 1980s to the late 1990s and then increasing up to 2015 as well as increasing cloud-base height from 1970s to the late 1990s and then decreasing up to 2015.

At present, ASOS only reports clouds under 12 000 ft (approximately 3600 m). Clouds in this vertical range play an important role in precipitation. On average, 94% of precipitation is accompanied by clouds within the surface–3600-m layer for all 719 stations from 2008 to 2012. Our results show the increase of cloud frequency is closely related to the increase of precipitation days in spring and winter. In summer when convection is strong, more than 20% of precipitation may occur higher than 3600 m, which may help to explain the weak correlation between the trends of total cloud frequency with the trends of precipitation days in summer. Therefore, the observation scope (3600 m) of the CT12K ceilometer used by ASOS may not be adequate to study clouds and their interactions with weather and climate.

Next-generation ceilometer sensors (CL31) from Vaisala replaced the old-generation ceilometer sensors (CT12K) at ASOS stations, and they provide more reliable information in the lower atmosphere up to 25 000 ft (about 7600 m). The new-generation ceilometers have the ability to observe a large part of clouds within the troposphere, thereby providing us with the opportunity to study the vertical structure of clouds and their effects on precipitation. In addition, the newest-generation ceilometers (e.g., Vaisala CL51 and Jenoptik CHM15k) can detect clouds up to 13 and 15 km. Even though their implementation is not planned for ASOS in the near future, the observations from these ceilometers may contain valuable information for the entire troposphere.

Although cloud observations from ASOS differ greatly from routine cloud observations and do not include observation of cloud types, the merits of the advantages of ASOS cloud observation cannot be neglected. The automatic, continuous, objective, quantitative, and high-density observations of cloud amount and cloud-base height data from ASOS can provide an opportunity to study cloud processes, validate models, and quantify cloud climate effects.

Acknowledgments

This study was funded by the National Natural Science Foundation of China (41525018 and 91337111) and the National Basic Research Program of China (2012CB955302). The data used here were downloaded from the National Climatic Data Center (ftp://ftp.ncdc.noaa.gov/pub/data/asos-fivemin/). The study was completed when the lead author was a Visiting Scholar at the Department of Atmospheric and Oceanic Science, University of Maryland.

REFERENCES

REFERENCES
Bender
,
F. A.
,
V.
Ramanathan
, and
G.
Tselioudis
,
2012
:
Changes in extratropical storm track cloudiness 1983–2008: Observational support for a poleward shift
.
Climate Dyn.
,
38
,
2037
2053
, doi:.
Brient
,
F.
, and
S.
Bony
,
2012
:
How may low‐cloud radiative properties simulated in the current climate influence low‐cloud feedbacks under global warming?
Geophys. Res. Lett.
,
39
,
L20807
, doi:.
Cess
,
R. D.
, and Coauthors
,
1990
:
Intercomparison and interpretation of climate feedback processes in 19 atmospheric general circulation models
.
J. Geophys. Res.
,
95
,
16 601
16 615
, doi:.
Chepfer
,
H.
,
V.
Noel
,
D.
Winker
, and
M.
Chiriaco
,
2014
:
Where and when will we observe cloud changes due to climate warming?
Geophys. Res. Lett.
,
41
,
8387
8395
, doi:.
Chernykh
,
I. V.
,
O. A.
Alduchov
, and
R. E.
Eskridge
,
2001
:
Trends in low and high cloud boundaries and errors in height determination of cloud boundaries
.
Bull. Amer. Meteor. Soc.
,
82
,
1941
1947
, doi:.
Costa-Surós
,
M.
,
J.
Calbó
,
J.
González
, and
J.
Martin-Vide
,
2013
:
Behavior of cloud base height from ceilometer measurements
.
Atmos. Res.
,
127
,
64
76
, doi:.
Costa-Surós
,
M.
,
J.
Calbó
,
J.
González
, and
C. N.
Long
,
2014
:
Comparing the cloud vertical structure derived from several methods based on radiosonde profiles and ground-based remote sensing measurements
.
Atmos. Meas. Tech.
,
7
,
2757
2773
, doi:.
Craven
,
J. P.
,
R. E.
Jewell
, and
H. E.
Brooks
,
2002
:
Comparison between observed convective cloud-base heights and lifting condensation level for two different lifted parcels
.
Wea. Forecasting
,
17
,
885
890
, doi:.
Dai
,
A.
,
2001
:
Global precipitation and thunderstorm frequencies. Part II: Diurnal variations
.
J. Climate
,
14
,
1112
1128
, doi:.
Dai
,
A.
,
T. R.
Karl
,
B.
Sun
, and
K. E.
Trenberth
,
2006
:
Recent trends in cloudiness over the United States: A tale of monitoring inadequacies
.
Bull. Amer. Meteor. Soc.
,
87
,
597
606
, doi:.
Eastman
,
R.
, and
S. G.
Warren
,
2014
:
Diurnal cycles of cumulus, cumulonimbus, stratus, stratocumulus, and fog from surface observations over land and ocean
.
J. Climate
,
27
,
2386
2404
, doi:.
Eastman
,
R.
,
S. G.
Warren
, and
C. J.
Hahn
,
2011
:
Variations in cloud cover and cloud types over the ocean from surface observations, 1954–2008
.
J. Climate
,
24
,
5914
5934
, doi:.
Ellrod
,
G. P.
, and
I.
Gultepe
,
2007
:
Inferring low cloud base heights at night for aviation using satellite infrared and surface temperature data
.
Pure Appl. Geophys.
,
164
,
1193
1205
, doi:.
Fasullo
,
J. T.
, and
K. E.
Trenberth
,
2012
:
A less cloudy future: The role of subtropical subsidence in climate sensitivity
.
Science
,
338
,
792
794
, doi:.
Forsythe
,
J. M.
,
T. H.
Vonder Haar
, and
D. L.
Reinke
,
2000
:
Cloud-base height estimates using a combination of meteorological satellite imagery and surface reports
.
J. Appl. Meteor.
,
39
,
2336
2347
, doi:.
Free
,
M.
, and
B.
Sun
,
2013
:
Time-varying biases in U.S. total cloud cover data
.
J. Atmos. Oceanic Technol.
,
30
,
2838
2849
, doi:.
Free
,
M.
, and
B.
Sun
,
2014
:
Trends in U.S. total cloud cover from a homogeneity-adjusted dataset
.
J. Climate
,
27
,
4959
4969
, doi:.
Henderson-Sellers
,
A.
,
1992
:
Continental cloudiness changes this century
.
GeoJournal
,
27
,
255
262
, doi:.
Hutchison
,
K.
,
2002
:
The retrieval of cloud base heights from MODIS and three-dimensional cloud fields from NASA’s EOS Aqua mission
.
Int. J. Remote Sens.
,
23
,
5249
5265
, doi:.
IPCC
,
2013
: Climate Change 2013: The Physical Science Basis. Cambridge University Press, 1535 pp., doi:.
Kim
,
S.-W.
,
E.-S.
Chung
,
S.-C.
Yoon
,
B.-J.
Sohn
, and
N.
Sugimoto
,
2011
:
Intercomparisons of cloud-top and cloud-base heights from ground-based Lidar, CloudSat and CALIPSO measurements
.
Int. J. Remote Sens.
,
32
,
1179
1197
, doi:.
Kokhanovsky
,
A. A.
, and
V. V.
Rozanov
,
2005
:
Cloud bottom altitude determination from a satellite
.
IEEE Trans. Geosci. Remote Sens. Lett.
,
2
,
280
283
, doi:.
Kottek
,
M.
,
J.
Grieser
,
C.
Beck
,
B.
Rudolf
, and
F.
Rubel
,
2006
:
World map of the Köppen-Geiger climate classification updated
.
Meteor. Z.
,
15
,
259
263
, doi:.
Kunkel
,
K. E.
,
L. E.
Stevens
,
S. E.
Stevens
,
L.
Sun
,
E.
Janssen
,
D.
Wuebbles
, and
J. G.
Dobson
,
2013
: Regional climate trends and scenarios for the U.S. National Climate Assessment: Part 9. Climate of the contiguous United States. NOAA Tech. Rep. NESDIS 142-9, 77 pp. [Available online at https://www.nesdis.noaa.gov/sites/default/files/asset/document/NOAA_NESDIS_Tech_Report_142-9-Climate_of_the_Contiguous_United_States.pdf.]
Lee
,
W.-H.
,
S. F.
Iacobellis
, and
R. C.
Somerville
,
1997
:
Cloud radiation forcings and feedbacks: General circulation model tests and observational validation
.
J. Climate
,
10
,
2479
2496
, doi:.
Leipper
,
D. F.
,
1994
:
Fog on the U.S. West Coast: A review
.
Bull. Amer. Meteor. Soc.
,
75
,
229
240
, doi:.
Minnis
,
P.
,
Y.
Yi
,
J.
Huang
, and
K.
Ayers
,
2005
:
Relationships between radiosonde and RUC‐2 meteorological conditions and cloud occurrence determined from ARM data
.
J. Geophys. Res.
,
110
,
D23204
, doi:.
Mittermaier
,
M.
,
2012
:
A critical assessment of surface cloud observations and their use for verifying cloud forecasts
.
Quart. J. Roy. Meteor. Soc.
,
138
,
1794
1807
, doi:.
NWS
,
1998
: Automated Surface Observing System (ASOS) user’s guide. NOAA Rep., 61 pp. [Available online at http://www.nws.noaa.gov/asos/aum-toc.pdf.]
Ockert-Bell
,
M. E.
, and
D. L.
Hartmann
,
1992
:
The effect of cloud type on Earth’s energy balance: Results for selected regions
.
J. Climate
,
5
,
1157
1171
, doi:.
Platt
,
C.
,
1989
:
The role of cloud microphysics in high-cloud feedback effects on climate change
.
Nature
,
341
,
428
429
, doi:.
Ramanathan
,
V.
,
R.
Cess
,
E.
Harrison
,
P.
Minnis
,
B.
Barkstrom
,
E.
Ahmad
, and
D.
Hartmann
,
1989
:
Cloud-radiative forcing and climate: Results from the Earth Radiation Budget Experiment
.
Science
,
243
,
57
63
, doi:.
Rossow
,
W. B.
, and
R. A.
Schiffer
,
1991
:
ISCCP cloud data products
.
Bull. Amer. Meteor. Soc.
,
72
,
2
20
, doi:.
Sherwood
,
S. C.
,
S.
Bony
, and
J.-L.
Dufresne
,
2014
:
Spread in model climate sensitivity traced to atmospheric convective mixing
.
Nature
,
505
,
37
42
, doi:.
Slingo
,
A.
, and
J.
Slingo
,
1988
:
The response of a general circulation model to cloud longwave radiative forcing. I: Introduction and initial experiments
.
Quart. J. Roy. Meteor. Soc.
,
114
,
1027
1062
, doi:.
Stubenrauch
,
C.
, and Coauthors
,
2013
:
Assessment of global cloud datasets from satellites: Project and database initiated by the GEWEX Radiation Panel
.
Bull. Amer. Meteor. Soc.
,
94
,
1031
1049
, doi:.
Sun
,
B.
,
2003
:
Cloudiness over the contiguous United States: Contemporary changes observed using ground‐based and ISCCP D2 data
.
Geophys. Res. Lett.
,
30
,
1053
, doi:.
Sun
,
B.
, and
P. Ya.
Groisman
,
2004
:
Variations in low cloud cover over the United States during the second half of the twentieth century
.
J. Climate
,
17
,
1883
1888
, doi:.
Sun
,
B.
,
P. Ya.
Groisman
,
R. S.
Bradley
, and
F. T.
Keimig
,
2000
:
Temporal changes in the observed relationship between cloud cover and surface air temperature
.
J. Climate
,
13
,
4341
4357
, doi:.
Sun
,
B.
,
P. Ya.
Groisman
, and
I. I.
Mokhov
,
2001
:
Recent changes in cloud-type frequency and inferred increases in convection over the United States and the former USSR
.
J. Climate
,
14
,
1864
1880
, doi:.
Sun
,
B.
,
T. R.
Karl
, and
D. J.
Seidel
,
2007
:
Changes in cloud-ceiling heights and frequencies over the United States since the early 1950s
.
J. Climate
,
20
,
3956
3970
, doi:.
Sun
,
B.
,
M.
Free
,
H. L.
Yoo
,
M. J.
Foster
,
A.
Heidinger
, and
K.-G.
Karlsson
,
2015
:
Variability and trends in U.S. cloud cover: ISCCP, PATMOS-x, and CLARA-A1 compared to homogeneity-adjusted weather observations
.
J. Climate
,
28
,
4373
4389
, doi:.
Trenberth
,
K. E.
,
J. T.
Fasullo
, and
J.
Kiehl
,
2009
:
Earth’s global energy budget
.
Bull. Amer. Meteor. Soc.
,
90
,
311
323
, doi:.
Vaisala
,
1989
: Technical manual: Laser ceilometer CT12K. Vaisala Rep., 304 pp. [Available from Vaisala Company, 100 Commerce Way, Woburn, MA 01801.]
Vislocky
,
R. L.
, and
J. M.
Fritsch
,
1997
:
An automated, observations-based system for short-term prediction of ceiling and visibility
.
Wea. Forecasting
,
12
,
31
43
, doi:.
Viúdez‐Mora
,
A.
,
M.
Costa‐Surós
,
J.
Calbó
, and
J.
González
,
2015
:
Modeling atmospheric longwave radiation at the surface during overcast skies: The role of cloud base height
.
J. Geophys. Res. Atmos.
,
120
,
199
214
, doi:.
Wang
,
J.
, and
W. B.
Rossow
,
1995
:
Determination of cloud vertical structure from upper-air observations
.
J. Appl. Meteor.
,
34
,
2243
2258
, doi:.
Warren
,
S. G.
,
C. J.
Hahn
,
J.
London
,
R. M.
Chervin
, and
R. L.
Jenne
,
1986
: Global distribution of total cloud cover and cloud types over land. NCAR Tech. Note NCAR/TN-273+STR, 29 pp., doi:.
Warren
,
S. G.
,
C. J.
Hahn
,
J.
London
,
R. M.
Chervin
, and
R. L.
Jenne
,
1988
: Global distribution of total cloud cover and cloud type amounts over ocean. NCAR Tech. Note NCAR/TN-317+STR, 42 pp., doi:.
Warren
,
S. G.
,
R. M.
Eastman
, and
C. J.
Hahn
,
2007
:
A survey of changes in cloud cover and cloud types over land from surface observations, 1971–96
.
J. Climate
,
20
,
717
738
, doi:.
Welch
,
R. M.
,
S.
Asefi
,
J.
Zeng
,
U. S.
Nair
,
Q.
Han
,
R. O.
Lawton
,
D. K.
Ray
, and
V. S.
Manoharan
,
2008
:
Biogeography of tropical montane cloud forests. Part I: Remote sensing of cloud-base heights
.
J. Appl. Meteor. Climatol.
,
47
,
960
975
, doi:.
Williams
,
A. P.
,
R. E.
Schwartz
,
S.
Iacobellis
,
R.
Seager
,
B. I.
Cook
,
C. J.
Still
,
G.
Husak
, and
J.
Michaelsen
,
2015
:
Urbanization causes increased cloud base height and decreased fog in coastal Southern California
.
Geophys. Res. Lett.
,
42
,
1527
1536
, doi:.
WMO
,
1974
: Manual on codes. WMO Publ. 306, Vol. 1, 348 pp. [Available from World Meteorological Organization, Avenue Giuseppe-Motta 41, CP2300, 1211 Geneva 2, Switzerland.]
Zelinka
,
M. D.
, and
D. L.
Hartmann
,
2010
:
Why is longwave cloud feedback positive?
J. Geophys. Res.
,
115
,
D16117
, doi:.
Zhang
,
J.
,
H.
Chen
,
J.
Bian
,
Y.
Xuan
,
Y.
Duan
, and
M.
Cribb
,
2012
:
Development of cloud detection methods using CFH, GTS1, and RS80 radiosondes
.
Adv. Atmos. Sci.
,
29
,
236
248
, doi:.

Footnotes

© 2017 American Meteorological Society. For information regarding reuse of this content and general copyright information, consult the AMS Copyright Policy (www.ametsoc.org/PUBSReuseLicenses).