Cloudiness is the most important meteorological factor determining the amount of solar radiation reaching the Earth's surface. The fraction of solar energy getting through the clouds is not constant. Differences in transmission of solar radiation through the clouds depend on a number of factors, which include the degree of cloudiness, the cloud genera present, cloud thickness (or height) from the base to the top of the cloud, cloud base height, microstructure and water content per unit of volume, the distribution of clouds with respect to the position of the Sun's disk in the sky, and other factors not yet fully known.
Thick low-level layered clouds may reduce solar radiation intensity by 80–90%. In certain conditions, tall convective clouds may raise its value by 10–15% when compared with cloudless weather (Monteith and Unsworth, 1988). This enhancement of solar flux at the surface is only an instantaneous effect.
In the research literature, the impact of cloudiness on solar radiation is usually considered in terms of cloud cover extent, with little regard to the cloud genera present. Yet simple measurements and observations indicate that even when the sky is overcast with high-level clouds, for example Cirrostratus, solar radiation still reaches the surface of the Earth without impediment. On the other hand, low-level and mid-level layered clouds, for example Nimbostratus, obstruct direct radiation. In both cases, the degree of cloudiness is the same, the sky is officially overcast (8/8), yet the difference in measured solar radiation intensity reaching the Earth's surface may equal up to 700 Wm−2 (Matuszko, 2009).
2. Research goals
The problem of the degree of absorption and transmission of solar radiation energy by different genera of clouds, despite its fundamental significance in climate research, has not been adequately explained thus far in the climatologic literature. Scientific papers on the subject are scarce and primarily contain descriptions of short-term experimental observations and measurements conducted at one or just a few stations (Kalitin 1950; London, 1957; Lumb, 1964; Manabe and Strickler, 1964; Kondratiev, 1965; Kostyanoy and Kurilova, 1966; Davies et al., 1975; Robinson, 1977; Suckling and Hay, 1977; Kondratiev and Binienko, 1984; Monteith and Unsworth, 1988; Mazin and Hrgian, 1989; Segal and Davis, 1992; Kondratiev et al., 1998; Kuchinke and Nunez, 1999).
The obtained results often differ from each other. The differences are due to local circumstances, synoptic situation, transparency of the atmosphere, difficulties in applying measurement methodologies, and other not yet known causes. Nevertheless, most researchers believe that clouds impact solar radiation intensity in one of two ways: usually weakening radiation, or intensifying it in certain weather conditions.
The aim of the paper is to try to determine the influence of cloud cover, of its extent and cloud genera, on solar radiation intensity measured at the Earth's surface. The results of the research conducted in Krakow (Poland) were compared to results obtained by other researchers who performed measurements in selected cities in Poland and around the world.
3. Source materials and methods
Solar radiation measurements were made in 2003–2007 using a Kipp & Zonen CM5 automatic sensor. The sensor was installed on the roof of the old astronomical observatory building in Krakow. The sensor's location allowed for free recording of solar radiation. This was confirmed by a measurement of the obstruction of the local horizon, which showed that nothing was blocking the path of solar radiation. The Kipp & Zonen CM5 sensor records solar radiation intensity in ten-minute intervals in Wm−2. Observations of the cloud cover, measured on a scale from 0 to 8, and cloud genera based on the international classification system found in the International Cloud Atlas (WMO, 1956), were performed at the same site.
In order to fully determine the impact of a particular genus of clouds on solar radiation intensity, it would be necessary to consider cases when the sky is entirely covered by this particular genus of clouds. Unfortunately, that was not always possible since some cloud genera (Cirrus, Cumulus, Cumulonimbus) usually do not cover the entire sky when present alone, i.e. when there is only one genus of clouds in the sky. For that very reason, it is difficult to identify a direct dependence of a particular cloud genus on solar radiation intensity. However, keeping certain limitations in mind, this relationship can be estimated.
First, the associations between particular cloud genera and solar radiation considering the Sun's angle of elevation over the horizon were investigated. Next, in order to limit an impact of other cloud genera on the amount of solar radiation reaching the Earth's surface, only the cases with just one cloud genus present in the sky were analysed and compared with cloudless sky conditions. For days thus selected, the flow of solar radiation intensity was recorded. This methodology allowed to maximize the objectivity of the results considering the limitations of data availability.
Hourly and daily solar radiation amounts were calculated based on intensity values measured by the CM5 sensor. In addition, a relative intensity of the solar radiation (global transmission) was determined by comparing measured values with values calculated for the upper limit of the atmosphere for a given day and time.
Basic statistical techniques were used in the research, mainly linear and quadratic regression (Haan, 2002).
Differences between solar radiation values at the upper limit of the atmosphere and the intensity of solar radiation measured at the Earth's surface in Krakow (φ = 50° 04N, λ = 19° 58E, h = 224 m above sea level) are shown in Figure 1. The figure shows that the atmosphere modifies the amount of solar radiation transmitted to the Earth's surface. The changes in radiation amount are due to the processes of absorption and scattering by gas molecules, clouds, and atmospheric aerosols.
4. Association of solar radiation with the degree of cloudiness
The solar radiation reaching the surface of the Earth is a sum of the radiation transmitted directly from the Sun and diffuse radiation coming in from the entire sky. Under clear sky conditions, the solar radiation intensity depends primarily on direct solar radiation; when the sky is overcast, it depends on scattered radiation. When the sky is partly cloudy, the intensity of direct radiation decreases and the intensity of scattered radiation increases.
The proportional contribution of the two components of solar radiation depend on the degree of cloud cover, cloud genera, and the position of clouds with respect to the Sun's disk. The position of clouds with respect to the Sun is important; however, not only the fact whether the Sun is covered by clouds or not is relevant, but also—in the case when the Sun's disk is not covered by clouds—whether the observer is located on the sunlit or shaded side of the cloud.
Measurements of solar radiation intensity and cloud cover observations performed in Krakow confirm the findings thus far (Monteith and Unsworth, 1988; Segal and Davis, 1992; Podstawczyńska, 2007). The results show that solar radiation intensity is greatest with partly cloudy skies, with clouds not covering the Sun's disk, and not when the sky is cloudless (Table I). The values in Table I are a generalisation of the relationship between solar radiation, degree of cloudiness and the height of the Sun over the horizon. The table shows the average of all measurements, regardless of time of year, hour, cloud genus, and the location of clouds relative to the Sun's disk.
Table I. Association of solar radiation [Wm−2] with the degree of cloudiness and the Sun's height above the horizon in Krakow (2004–2007)
Degree of cloudiness (in octas)
Sun's height above the horizon [ho]
ho < 20
ho > 60
Robinson (1977) as well as Monteith and Unsworth (1988) state that well-developed convective clouds may increase a global radiation value by 5–15% on average with respect to cloudless sky conditions. The largest increase in radiation intensity takes place when the sky is 50% clouded, with the horizontal and vertical dimensions of each cloud being roughly the same. Research by Segal and Davis (1992), conducted in Colorado (USA), shows an increase in solar radiation intensity of 250 Wm−2 when compared with cloudless sky results. Such a significant increase in global radiation intensity was due to reflection in the lateral parts of convective clouds, took place around noontime, and usually lasted 15–30 min. In some cases, duration of about 1 h was noted. Similar events were also observed in Krakow and Gorzow Wielkopolski (Matuszko and Soroka, 2009), but they lasted no more than 20 min.
In Krakow, the largest hourly average value of solar radiation intensity corresponds to a cloud cover of 1/8, with the Sun's angular elevation above the horizon at its maximum (Table I). Global radiation intensity decreases as cloud cover increases from 825 to 270 Wm−2 when the Sun's height is at its maximum (ho > 60), and from 247 to 62 Wm−2 when the Sun's height is lowest (ho < 20). The sharpest decreases in radiation intensity are observed when cloud cover increases from 7/8 to 8/8, namely when direct radiation ceases to reach the Earth's surface. A relatively large difference in radiation intensity can also be observed when cloud cover increases from 6/8 to 7/8, especially when the Sun's height is more than 50° (Table I). When the cloud cover is smaller and the Sun's angular elevation is low, radiation intensity decreases at a significantly lower rate due to the influence of the increase in cloud cover.
It is important to note that in Krakow the largest (over 1000 Wm−2) measured global radiation values occur when the cloud cover is equal to 3/8 or more, often 6/8 (Table II). Such situations occur only from May until July, which is the time period when the Sun's height is the largest and convective clouds dominate the cloud cover.
Table II. Cloud cover and synoptic situation on days with maximum solar radiation intensity (over 1000 Wm−2) in Krakow (2004–2007)
Date and time (UTC + 1)
Solar radiation intensity (Wm−2)
Cloud cover at noon observation time
Degree of cloudiness (octas)
On the basis of synoptic maps of Europe, 21 types of synoptic situations (circulation types) were distinguished, taking into account the direction of air masses advection and the kind of pressure pattern.
c Cyclonic. N Situations with an advection of air masses from the north.
NW From the northwest etc. Ca Central anticyclone situation (high centre).
Ka Anticyclonic wedge or ridge of high pressure.
Cc Central cyclonic.
Bc Trough of low pressure. (according to Niedźwiedź 2007).
Ac, Cu, Cb
Ci, Cs, Cu
Ci, Ac, Sc, Cu
Ci, Ac, Sc, Cu
Ac, Sc, Cu
Ci, Ac, Cu
Ac, Sc, Cu
Ac, Sc, Cu
In Krakow, the cloud cover weakens solar radiation intensity for most of the year—from January to April as well as from August to December. This is noticeable especially during the cooler half of the year, when the cloud cover is dominated by layered clouds which often cover the sky entirely, thus blocking direct radiation from reaching the Earth's surface. The process of multiple scattering of solar radiation is typical for thick clouds, increasing absorption and causing a decrease in the transmission of solar radiation (Podstawczyńska, 2007).
The thicker the cloud, the less radiation is transmitted, and the greater the resulting albedo. According to Gojsa and Shoshin (1970), when the Sun's angular elevation is low, Stratus and Stratocumulus clouds (175 m thick) transmit 37% of incident radiation, whereas clouds 900 m thick transmit only 10%. The Sun's angular elevation is more important in the case of thin clouds, namely Stratus and Stratocumulus. In this case, when the Sun's angular elevation is 50°, a substantial 56% of radiation is transmitted, and when a cloud's thickness is 900 m, only 13% of radiation is transmitted, which is not much more than when the Sun's height is 10°.
5. Differences in solar radiation intensity due to cloud genera
Different cloud genera transmit solar radiation to different degrees, shown by measurement results at many locations around the world (Kalitin, 1950; London, 1957; Lumb, 1964; Manabe and Strickler, 1964; Kondratiev, 1965; Davies et al., 1975; Suckling and Hay, 1977; Kondratiev and Binienko, 1984; Monteith and Unsworth, 1988).
Research on the impact of cloud genera on radiation transmission has shown that the interpretation of these associations is complex and accurate comparisons are difficult. Therefore, although the patterns in the results obtained at various locations around the world are similar, the exact values are somewhat different.
The hourly average values of solar radiation intensity obtained in Krakow, depending on cloud genera and the Sun's height above the horizon and compared with cloudless sky conditions (Table III), are close to the results published by Kalitin (1950), which were based on measurements performed in the Russian city of Pawłowsk during 1931–1940.
Table III. Hourly averages of solar radiation intensity (%) depending on cloud genera and the Sun's height above the horizon, compared with cloudless sky conditions in Krakow (2004–2007)
Sun's height above the horizon (ho)
ho < 20
ho > 60
Ci + Cs + Cc
Cb + Cu
Layered clouds, especially low-level layered clouds, block radiation the most. On the other hand, global radiation intensity may be very high in the case of vertical clouds, provided they do not cover the Sun's disk. However, relative to average values, it is lower than that for cloudless skies and high clouds. The results of observations and measurements performed in Krakow also confirmed a very important property of Cirrus clouds described in many earlier publications (Vowinckel and Orvig, 1962; Abakumova et al., 1989; Anikin and Šukurov, 1989). The presence of Cirrus clouds in the atmosphere produces such a large proportion of scattered radiation that the global radiation intensity is greater than it would be under cloudless sky conditions. In Krakow, the largest hourly average global radiation intensity is recorded when high-level clouds are present in the skies, primarily Cirrus clouds, and the Sun's height above the horizon is at its maximum (Table III).
A good way to show differences in global radiation intensity as dependent on the genus composition of cloud cover is to compare 24-h temporal data patterns representing the solar radiation intensity for a cloudless day, and that for a day when the sky is covered by clouds belonging to only one genus of clouds (Figure 2). This method has been applied in other studies (Robinson, 1977; Estupiñán and Raman, 1996; Kuchinke and Nunez, 1999; Podstawczyńska, 2007), where it was assumed that weather conditions on a cloudless day are close to ideal. The days were selected in such a way that they belonged to the same month in order to assure the comparability of astronomical conditions (the length of day and the Sun's height above the horizon). Nimbostratus clouds proved to be the least permeable to solar radiation in Krakow, which confirms the results obtained by Liou (1976).
Layered clouds such as Stratus, Stratocumulus, and Altostratus (Figure 2) are only slightly permeable to solar radiation, which makes them second in line in terms of weak permeability, similarly to what was described by Houghton (1954) and List (1966). The presence of vertical clouds in the sky, such as Cumulus, Cumulonimbus, and Altocumulus, leads to high values of solar radiation, provided the clouds do not cover the Sun's disk.
If a Cumulus heaped cloud is blocking the Sun, a large decrease in radiation intensity follows due to the significant thickness of the cloud. According to Reynolds et al. (1975) as well as Liou (1976), the permeability of Cumulonimbus clouds to radiation (with the sky entirely overcast) is only 3%, whereas 10–23% of solar radiation is transmitted through ‘normal’ Cumulus clouds (Houghton, 1954; London, 1957).
A temporary increase in the solar radiation value, in comparison to a base value under cloudless sky conditions, is possible in the case of vertical clouds and Cirrus clouds. This effect is observed in Krakow with Cumulus and Cumulonimbus clouds present in the sky. This also applies to Altocumulus, Stratocumulus, Cirrus, and Cirrostratus clouds. This type of situation arises when clouds do not cover the Sun's disk, direct sunrays reach the Earth's surface without obstruction, and scattered radiation is being emitted by clouds located around the visible solar disk. In order for the mirror effect to actually occur, clouds must be present in the skies that are well developed in the vertical direction. Merely the presence of Stratocumulus or Altocumulus clouds will most likely not be sufficient.
It was observed (Matuszko and Soroka, 2009) that if a cloud covers the Sun's disk, then radiation intensity decreases rapidly. When the Sun reappears from behind the clouds, then radiation intensity becomes very strong again. Radiation intensity becomes unusually high for a few minutes before and after the Sun's disk is covered by clouds.
This phenomenon, being a result of radiation diffraction on the boundary of a cloud, may be compared to brief flashes of light. Monteith (1973) explains the very high solar radiation values during periods preceding the moment of shielding of the Sun by clouds and immediately after it reappears from behind the clouds by a strong scattering of radiation by water droplets found along the boundary of a cloud. According to Robinson (1977), the concentration of water droplets within the lateral parts of Cumulus and Cumulonimbus clouds is smaller, and the process of scattering and reflection is stronger than the process of absorption of solar radiation (the silver lining effect). Kuchinke and Nunez (1999) compare the sunlit edges of clouds to lenses that direct sunrays to the surface of the Earth.
Therefore, the processes of reflection and diffraction as well as a strong scattering of radiation in the direction of the surface of the Earth cause increased global radiation intensity with respect to cloudless sky radiation intensity values. This phenomenon takes place primarily within the lateral parts of convective clouds and along the edges of convective clouds lit by direct solar radiation.
Global transmission values of solar radiation for a cloudless day and a day with a particular genus of clouds (Table IV) confirm that there are differences in solar radiation intensity. This, in turn, strongly suggests that solar radiation intensity depends on cloud genera. The global transmission of solar radiation is defined as the ratio of the solar radiation measured on the horizontal plane at the surface of the Earth to the solar radiation at the upper limit of the atmosphere. On the basis of a comparison of transmission data for selected days (cloudless and with particular clouds), it may be concluded that solar radiation is attenuated the most by Nimbostratus and Stratocumulus clouds, followed by Stratus, Altostratus, Altocumulus and Cumulus (Table IV). Cirrus clouds are characterized by the greatest transmission. The presence of Cumulonimbus, assuming that it does not cover the Sun's disk practically does not limit solar radiation getting through to the surface of the Earth. However, when the sky is entirely covered by Cumulonimbus clouds, permeability to solar radiation is very small (below 6%).
Table IV. A comparison of the global transmission of solar radiation (T) for a cloudless day (B) and a day with a selected genus of clouds: R (difference in transmission for a cloudless day and a day with selected clouds), S (ratio of transmission on a day with clouds to transmission on a cloudless day); data for Krakow (2004–2007)
It should be remembered, however, that the magnitude of the global transmission of solar radiation is impacted not only by cloud cover but also by the content of water vapour, aerosols, and other factors that are not yet fully understood. For that very reason, the results presented here describe only approximate associations due to the error caused by objective difficulties in measurement methodology and observation of the investigated phenomena.
The proper selection of pairs of days—a cloudless day and a day with a selected genus of clouds—out of the days of the same month guarantees the best pick of days with similar meteorological conditions. It would be ideal if a cloudless day were followed by a day with a particular genus of clouds. This would allow for the assumption that the transparency of the atmosphere on the two consecutive days was similar. This type of ideal two-day scenario never actually occurred during the research period.
Observations of changes in solar radiation intensity due to the presence of particular cloud genera made in Krakow during the months with the longest days with Sun's height above the horizon at its maximum (May and June), proved that at the moment when the Sun's disk becomes covered by Cirrus spissatus clouds, temporary decreases in solar radiation intensity amounted to about 100 Wm−2. The temporary decreases in solar radiation intensity for other clouds were as follows: Cirrostratus fibratus 150–300 Wm−2, Cumulus humilis 200–300 Wm−2, Cumulus mediocris 450 Wm−2, Cumulus congestus 600 Wm−2, Cumulus and Stratocumulus 600 Wm−2, Altocumulus and Cumulus 750 Wm−2, and a maximum value for Cumulonimbus, more than 800 Wm−2.
The presence of Cumulus clouds in cloud cover that is a mix of cloud genera is the deciding factor when it comes to the magnitude of increases in solar radiation intensity. This is due to the fact that the mirror effect is observed most frequently when the Cumulus cloud genus is present. An example of this scenario would be the simultaneous presence of Cumulus clouds and other clouds, such as Altocumulus or Stratocumulus, where it is Cumulus that is primarily responsible for increases in solar radiation intensity.
The presence of Nimbostratus clouds also causes a significant decrease in solar radiation in comparison to solar radiation values measured under cloudless skies. On days when Nimbostratus clouds were observed throughout the entire day, the highest recorded radiation intensity ranged from 80 to 120 Wm−2. The corresponding radiation intensity maximum for a cloudless day is over 800 Wm−2. Hence, it may be concluded that the attenuation of solar radiation intensity by the Nimbostratus cloud genus was approximately 700 Wm−2.
A very wide range of values of the attenuation of solar radiation has been noted for Stratocumulus clouds. It was observed (Matuszko and Soroka, 2009) that when the sky was entirely covered by clouds of this genus, radiation intensity ranged from 35 to 350 Wm−2. However, it is not known what cloud genera found higher in the atmosphere could have been obscured. In May of 2007, with a cloud cover of 7/8 consisting of Stratocumulus clouds, radiation intensity varied within a range of 370–820 Wm−2. A certain number of rare cases have been recorded in Gorzow Wielkopolski, where Stratocumulus clouds have caused radiation increases. This type of scenario occurred, for example, in June of 2008, with a cloud cover of 1/8 Stratocumulus castellanus and 1/8 Altocumulus perlucidus: the solar radiation intensity increased to 1020 Wm−2. An interesting daily pattern of solar radiation intensity was observed in Krakow on the 17 May 2007 (Figure 3).
The morning had been cloudless; at about 9:00 AM, Cumulus clouds developed, but because they did not shield the Sun's disk, solar radiation increased rapidly. At 10:50 AM, solar radiation reached 975 Wm−2. Soon after, a Cumulus cloud began to shield the Sun's disk. At 11:00 AM, radiation intensity decreased to 395 Wm−2. Once the cloud had passed, at 11:10 AM, radiation intensity increased to 1038 Wm−2. Subsequently, the cloud cover increased and Stratocumulus clouds emerged. By 12:00 PM, Stratocumulus clouds covered the entire sky, and radiation intensity decreased to 607 Wm−2.
As mentioned earlier, just before a cloud covers the Sun's disk as well as just after the Sun reappears from behind the given cloud, solar radiation intensity values become very high. On the basis of observations (Matuszko and Soroka, 2009) it appears that, on average, Stratocumulus clouds weaken radiation intensity by at least 400 Wm−2.
In the winter, when the Sun's height above the horizon is low, differences between the highest and the average solar radiation intensity are much smaller (not more than 200 Wm−2). This may also be caused by a lower frequency of occurrence of convective clouds during this time of year. During the cool season of the year, it is more difficult than it is in the summer to assess the role of cloud cover in the transmission of solar radiation that reaches the surface of the Earth. This is due to the fact that the measured radiation values are small, with average value of approximately 300 Wm−2 at noon under cloudless skies.
What is also typical is the uniform decrease in minimum radiation values when the sky is overcast with a steady amount of solar radiation reaching the Earth's surface for the same cloud type (layered).
Results from investigations carried out in Łódź (Podstawczyńska, 2007), similar to those in Kraków, shows that the highest values of solar radiation (over 1000 Wm−2) occurred only when convective clouds were present in the cloud cover, with Cumulus mediocris clouds being noted most frequently (65% of cases). In Łódź, the longest recorded time interval of radiation intensity over 1000 Wm−2 occurred during the morning hours (10:40–11:30 AM) and lasted approximately 40 min. Solar radiation of intensity this high occurred in Krakow twice around noon and lasted no more than 20 min (7 June 2006 and 18 May 2007).
Research from North Carolina (Estupiñán and Raman, 1996) confirmed that solar radiation reached peak values when Cumulus clouds gathered near the Sun's disk at its highest position over the horizon. According to Estupiñán and Raman (1996), the presence of vertically developed clouds was the determining factor. This is most likely why Stratocumulus clouds alone do not cause this type of effect, because their vertical dimensions are simply too small.
Decreases and increases in solar radiation values due to changes in cloud cover over Krakow on days when solar radiation intensity reached its maximum values (over 1000 Wm−2) are seen in diagrams of daily changes (Figure 3 and tables). The shape of the radiation intensity curve reflects the nature of cloud cover during the day. Large fluctuations in solar radiation intensity, disrupting natural daily patterns, indicate the development of convection in the morning hours (Figure 3(a), (c), (d), (e), (j)) and the subsequent growth of Cumulus clouds via the gradual formation of its different species (humilis, fractus, mediocris, congestus), leading up to the development of Cumulonimbus in the afternoon. Research by Robinson (1977) shows that the magnitude of scattered radiation changes in conjunction with the development of convective clouds, and it is different on the sunlit or shadowed side of a cloud.
Depending on whether the flow of direct solar radiation is covered by a cloud or not, radiation intensity suddenly falls and then climbs again. If a Cumulus cloud is thin (fractus or humilis), the fluctuations are approximately 200 Wm−2, but if the Sun's disk is temporarily covered by passing clouds that are well developed (Cumulus congestus or Cumulonimbus), the fluctuations increase up to 800 Wm−2. In the morning hours, the solar radiation curve is increasing and smooth, but in the evening hours, the curve is decreasing and uneven.
The covering of the Sun's disk by Stratocumulus and Altocumulus stratiformis perlucidus and translucidus clouds causes the irregular pattern of solar radiation recorded over time. The simultaneous appearance of several cloud genera in the sky (Cumulus, Stratocumulus, Altocumulus, Cirrus, Cirrocumulus) at different height levels (Figure 3(c), (e), (g), (i), (j), (l)) causes an irregular solar radiation pattern during the course of the day, yet a narrower fluctuation range of solar radiation intensity values than when Cumulus or Cumulonimbus clouds are alone present.
The daytime range of temporary changes in attenuation of solar radiation by particular cloud genera is very large—from 30 Wm−2 for Cirrus and Cirrostratus clouds to almost 900 Wm−2 for Cumulonimbus clouds. In many cases, it is not only the cloud genus that determines solar radiation intensity; the species or even variety of a cloud can also be a contributing factor. The largest differences in the transmission of radiation are observed for the varieties translucidus and opacus as well as for the species calvus and capillatus, humilis and congestus, fibratus and spissatus, and stratiformis and castellanus.
On days with Stratus (8/8) clouds in the sky, differences in radiation intensity are very large due to the variable thickness of Stratus clouds. Another reason for this, perhaps the most important one, is the presence of clouds above, which cannot be seen from a ground-based research station. When Cirrus and Cirrostratus clouds are present, the Sun's height above the horizon is crucial. When the Sun's height is low, Cirrus and Cirrostratus clouds significantly limit radiation. Kondratiev (1965) argued that when the Sun's height increases from 5° to 50°, the transmission of solar radiation through Cirrus clouds increases from 2 to 44%.
It should be remembered that it might not always be possible to determine the impact of a single cloud genus on solar radiation intensity, as several cloud genera can often be seen in the sky at the same time. Furthermore, clouds such as Cirrocumulus are very rare (3%) and generally not present alone.
The results obtained by analysing solar radiation data from the City of Krakow confirm well-known patterns regarding the association of solar radiation intensity, measured at the surface of the Earth, with cloud cover. The following new conclusions have also been drawn:
1.Solar radiation intensity depends on the Sun's angular elevation above the horizon, the amount of cloud cover shielding the Sun's disk, the position of clouds with respect to the Sun's disk, cloud genus, cloud species, cloud variety, as well as the transparency of the atmosphere.
2.Under cloudless skies, both annual and daily patterns of solar radiation reflect the pattern of changes in the Sun's height above the horizon. A regular pattern of solar radiation becomes disrupted by the arrival of clouds and changes in the transparency of the atmosphere.
3.The impact of different cloud genera on solar radiation intensity is easier to observe in the warmer versus the cooler part of the year due to the fact that larger amounts of solar radiation are being transmitted.
4.Clouds impact solar radiation intensity usually by weakening radiation, however, in certain weather conditions they may also intensify radiation.
5.The range of changes in attenuation of solar radiation by different cloud genera is very large: from 30 Wm−2 for Cirrus clouds to over 800 Wm−2 for Cumulonimbus clouds.
6.The highest intensity of solar radiation (over 1100 Wm−2, in Krakow and in other cities) was observed when the sky was partly cloudy, from 3/8 to 6/8, with Cumulus mediocris and congestus as well as Cumulonimbus clouds present.
7.A high (over 900 Wm−2) solar radiation intensity is also possible when the cloud cover consists of Stratocumulus and Altocumulus clouds as well as associated with them Cumulus clouds.
8.In many cases, it is not just the cloud genus that determines solar radiation intensity: cloud species and variety can be equally important. The largest differences in the transmission of solar radiation for the same cloud genus are typical of the translucidus variety, for which the intensity of solar radiation is greater than for opacus, and for the fibratus species, which slightly weakens radiation in comparison to spissatus.
9.On days when Stratus (8/8) clouds are present, radiation intensity shows a wide range (on average from 40 to 80%) due to the variable thickness of Stratus clouds; an even more important reason, however, may be the presence of clouds above, which cannot be seen from a ground-based research station.
10.When Cirrus and Cirrostratus clouds are present in the sky, the Sun's height above the horizon is a crucial factor. When the Sun's height is low (h < 20°), the transmitted radiation is largely limited; solar radiation intensity is approximately 45% lower than it would be if the Sun's height were high (h > 60°).