INTERNATIONAL JOURNAL OF ENERGY RESEARCH

Int. J. Energy Res. 2007; 31:109–118

Published online 1 August 2006 in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/er.1229

A new formulation for solar irradiation and sunshine

duration estimation

Ahmet Duran S- ahin*,y

Meteorology Department, Energy Group, Aeronautic and Astronautics Faculty,
I_stanbul Technical University, Maslak 34469, I_stanbul

SUMMARY

In solar engineering and atmospheric sciences solar irradiation and sunshine duration estimations are very
important for different problems. Classical approaches that based on Angström equation for estimating
global solar irradiation in terms of sunshine duration are abundant in the literature. Most of them include
linear and lesser extent slightly nonlinear relationships between these two variables. Parameters are
determined invariably by the least squares technique leading to regression lines or curves as models. None
of these models provides within year variations in the parameters and they are all very rigid in applications
yielding single global solar irradiation estimation for a given data set. In this paper, relation between
extraterrestrial variables (length of day and solar irradiation) ratio and terrestrial variables (measured
sunshine duration and solar irradiation) ratio is taken into account. This is a novel approach and has no
restrictive assumptions. Decreasing amount of extraterrestrial variables ratio can be evaluated by
subtracting measured terrestrial variables ratio (sunshine duration/total solar irradiation) from theoretical
variables ratio. Differences between these ratios give atmospheric effects on length of day and
extraterrestrial solar irradiation. The differences are represented by a parameter which helps to estimate
easily the seasonal average parameter values without the application of the least square method. The main
idea of this paper is to suggest not only a novel but also a practical formulation for solar irradiation and
sunshine duration estimations. Its application is presented for three stations in Turkey and it is seen that
better estimations are evaluated by proposed method than Angström method. Copyright # 2006 John
Wiley & Sons, Ltd.

KEY WORDS: extraterrestrial irradiation; length of day; solar irradiation; solar energy; sunshine duration

1. INTRODUCTION

Solar irradiation and sunshine duration records depend on combined effects of astronomical
and meteorological events. The astronomic effects on the solar energy variables, namely

*Correspondence to: Ahmet Duran S- ahin, Meteorology Department, Energy Group, Aeronautic and Astronautics
Faculty, I_ stanbul Technical University, Maslak 34469, I_stanbul.
y
E-mail: sahind@itu.edu.tr

Received 28 June 2005

Revised 5 October 2005
Copyright # 2006 John Wiley & Sons, Ltd. Accepted 29 May 2006
110 - AHIN
A. D. S

extraterrestrial solar irradiation, HE and sunshine duration SE are deterministically calculable

by mathematical expressions depending on the average distance of the sun, longitude, latitude,
zenith and declination angles at different locations and seasons of the year. The mathematical
expressions for estimation procedure can be found in any renewable energy text books (Iqbal,
1983; Duffie and Beckman, 1991). They show definite periodicities without random behaviours.
Meteorological events effects on the solar energy calculations introduce random behaviours.
Meteorological solar irradiation (terrestrial) and sunshine duration ST variables have
randomness in their temporal and spatial evolutions due to the following reasons:
(1) The astronomical extraterrestrial irradiation and sunshine duration are shortened due to
meteorological and atmospheric events which are measured at a solar station as terrestrial
solar irradiation HT and sunshine duration ST. In other words, ST 5 SE and HT 5 HE.
(2) The shortening effect is not definite but might be in the form of random amounts during a
day or month depending on the climate and weather conditions.
Consequently, ratios of terrestrial solar energy variables to extraterrestrial counterparts as
HT/HE and ST/SE assume values between zero and one in a rather random manner depending
on the cloud cover percentage of the period concerned. In practice, the ratio HT/HE rarely
exceed 0.9 though it is possible that ST/SE ratio can approach one during perfectly clear sky.
Furthermore, it is logically obvious that these ratios are directly proportional to each other. In
practice, the ST measurements are comparatively easier and economical than HT measurements,
and therefore, many researches have proposed various statistical expressions in order to
estimate the latter from the former.
The first systematic proposal was in the form of linear expression as suggested by
Angström(1924). His formula has been used in practical applications for many years to estimate
the daily, monthly and annual global solar radiation amount, HT, from the comparatively
simple measurements of sun shine duration, ST, according to
HT ST
¼aþb ð1Þ
HE SE
In this equation a corresponds to the relative diffuse radiation and on an overcast day, whereas
(a+b) corresponds to the relative cloudless-sky global irradiation (Gueymard et al., 1995). As
explained above, although HT and ST vary temporally in a random manner, HE and SE have
deterministic values and the question is whether the model parameters, a and b also vary
temporally and randomly at a given station. In most applications so far in the literature, a and b
are considered as constants for the time period used in the application of Equation (1).
However, it is shown by S- ahin and S- en (1998) that a and b also change spatially and temporally.
The coefficients are estimated from available solar irradiation and sunshine duration data at a
location by frequent use of statistical regression technique. However, in such an approach there
are implied assumptions as given by different researchers (S- ahin and S- en, 1998; S- en and S- ahin,
2001; S- en et al., 2001; S- ahin et al, 2001).
A detailed historical evolution of Angström equation was explained by Martinez-Lozano
et al. (1984). Further criticisms are presented by Gueymard et al. (1995) and accordingly some
authors suggested new alternative methods (Jain, 1990; S- ahin and S- en, 1998; S- en, 2001;
Suehrcke, 2000).
On the other hand, Gopinathan (1988), Rietveld (1978), Sabbagh et al. (1977), Swartman and
Ogunlade (1967), Dogniaux and Lemonie (1983) have considered additional meteorological

Copyright # 2006 John Wiley & Sons, Ltd. Int. J. Energy Res. 2007; 31:109–118
DOI: 10.1002/er
 NEW FORMULATION FOR SOLAR IRRADIATION 111

factors to Equation (1) for the purpose of increasing the accuracy in the parameter estimates.
Although each one of these studies refined the parameter estimates but they all depend on the
average parameter values obtained by the least squares method, and therefore, there are still
remaining errors although smaller than the Angström’s model. Furthermore, Ögelman et al.
(1984) have adopted the incorporation of the standard deviation of the sunshine duration for a
better estimation of the model parameters, Soler (1986, 1990) has shown that monthly variations
of a+b are meteorologically sound and similar for different locations.
The main purpose of this paper is to suggest a new method based on a simple concept with
physically plausible estimations for sunshine duration and solar irradiation. Atmospheric effect
on solar irradiation and sunshine duration can be calculated easily by the proposed method.

2. PROPOSED METHOD

Through the classical approaches it is difficult to find atmospheric effect to extraterrestrial solar
irradiation and length of day. According to the suggestion in this paper, extraterrestrial
variables ratio ðSE =HE Þ is assumed to have a reduction by amount Re due to cloud, dust,
humidity, etc. Such a reduction in sunshine duration and solar irradiation are measured on the
horizontal surface. In other words, there is a relation between extraterrestrial and terrestrial
ratios due to atmospheric effect on these ratios. This reduction effect can be expressed as
SE ST
ð1
Re Þ ¼ ð2Þ
HE HT
where Re represents reduction amount of extraterrestrial ratio. The reduction factor results from
Equation (2) as
ððSE =HE Þ
ðST =HT ÞÞ ST =HT
Re ¼ ¼1
ð3Þ
SE =HE SE =HE
Given the astronomical calculations of SE and HE together with measurements of ST and HT the
reduction amount can be calculated easily from Equation (3) easily. If Re is known then
terrestrial sunshine duration (ST) and solar irradiation (HT) can be evaluated as
HT SE ð1
Re Þ
ST ¼ ð4Þ
HE
or
ST HE
HT ¼ ð5Þ
SE ð1
Re Þ
The suggested method has the following advantages:
1. Atmospheric effect to extraterrestrial solar components can be explained easily. In other
words, reduction amount in solar irradiation or length of day can be evaluated by
proposed method.
2. Angström equation parameters (a and b) need for each period (month, day or hour) a long
time measurements for each station. However, this method provides reduction parameter
evaluation for each month, day or hour. In other words, atmospheric effects to
extraterrestrial solar variables monitored for each period.

Copyright # 2006 John Wiley & Sons, Ltd. Int. J. Energy Res. 2007; 31:109–118
DOI: 10.1002/er
112 - AHIN
A. D. S

42.0 BLACK SEA

ISTANBUL
ANKARA
40.0
Aegean
Sea
38.0
ADANA

36.0
MEDITERRANEAN SEA SYRIA IRAQ
26.0 28.0 30.0 32.0 34.0 36.0 38.0 40.0 42.0 44.0
LONGITUDE (Degree)

Figure 1. Map of stations location.

3. In the proposed method there is no need for least square technique parameter estimation
procedure.

3. DATA

In this paper three stations are considered in Turkey (Figure 1). Monthly mean values of solar
irradiation and sunshine duration for 11 year are used in calculation; hence each station has 132
monthly mean data. Monthly mean values are estimated from hourly measurement data. All
data are measured with classical actinographs by State Meteorological Service which is one of
the members of ECMWF (DMI, 2005).
Adana, is located in the shoutern part of Turkey. Mediterranean Sea effects occur more
effectively. Moderate, severe and extreme drought magnitudes are seen at high level sunshine
duration and around high solar irradiation values at this region. These extreme values show that
sunshine duration is more effective in this area than solar irradiation. It is known that the
highest solar energy potential occurs in the southern part of Turkey. Almost in each day of year,
water is heated with solar collectors and hence solar power plants can be built in the
Mediterranean areas. Hence, solar variables and parameters are very important for this station.
Ankara is located in the central part of Turkey, there are variations depending on the
continental climate effects where topographic conditions mostly affect the rainfall occurrences.
Topographical climatic effects will be highest especially in summer months. Generally, in this
semi-arid region, moderate and severe droughts occurred in the past. Continental effects rise to
high level degree at this region.
Istanbul, has both continental and maritime climatic effects. It is located in northwestern part
of Turkey. Due to maritime effect cloud amounts are high and sunshine duration values are
lower than southern and central regions of Turkey (S- en and S- ahin, 2001; S- aylan et al., 2002).

4. APPLICATION

First of all, Re parameters of each month at each station are calculated and it is seen that some
Res are negative. As can be concluded from Equation (3) when extraterrestrial ratio is higher

Copyright # 2006 John Wiley & Sons, Ltd. Int. J. Energy Res. 2007; 31:109–118
DOI: 10.1002/er
 NEW FORMULATION FOR SOLAR IRRADIATION 113

than terrestrial ratio, a positive Re value occurs, otherwise a negative Re value is obtained.
Generally, in Adana station terrestrial ratios are higher than extraterrestrial ratios except for
few months. At this station sunshine duration values are high but solar irradiations are not as
high as expected. As a result, terrestrial ratios are higher than extraterrestrial ratios.
In Ankara station, most of the characteristic Res have negative values. These ratios also show
atmospheric effect to extraterrestrial ratio. In other words, in some months, extraterrestrial ratio
is reduced 60 and 40% received horizontal surface. Under all circumstances Re values represent
atmospheric effect irrespective of their signs. Positive and negative values must be considered for
comparison between terrestrial and extraterrestrial ratios. Monthly mean values indicate that in
the first and the last two months of the year, terrestrial ratios are higher than extraterrestrial
ratios. On the contrary, in other months terrestrial ratios are higher than extra terrestrial ratios.
Hence, negative Re values occur during eight months in this station.
In contrast to Adana and Ankara stations, majority Re values of Istanbul are positive
(Figure 2). Four months atmospheric effects to extraterrestrial ratios are higher than 0.6 ratio
value. It is estimated that average monthly terrestrial ratios values are higher than
extraterrestrial ratios during seven months (S- en and S- ahin, 2001). In this figure polynomial
connections occur below the straight line of extraterrestrial ratios.
By using of Re values, measured terrestrial variables (sunshine duration and solar irradiation)
can be estimated with Equations (4) and (5). Although Angström parameters a and b are
constant, in the proposed method there are different Re values for each month with a sequence
of Re values it is possible to make probabilistic estimations. This provides an opportunity for the
temporal prediction of Re values and solar irradiation reductions amount. If Re value of each
month is used for estimation then there might be very little error. For optimal usage, constant
Re value must be considered with minimum estimation error. In this paper, positive and negative
Re values are estimated. Hence, average positive Rep value is calculated from positive Re values
and average negative Ren is calculated from negative Re values. All stations have Rep and Ren
values in addition, Agström parameters are presented in Table I. Other variables are taken into
account directly. If Re value is positive then Rep value is taken into account for terrestrial
sunshine duration and solar irradiation estimations from Equations (4) and (5), otherwise Ren is

0.8

0.6

0.4

0.2
Re

0
1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97 103 109 115 121 127 133 139
-0.2

-0.4

-0.6

-0.8
Time (month)

’
Figure 2. Monthly Re values of Istanbul station.

Copyright # 2006 John Wiley & Sons, Ltd. Int. J. Energy Res. 2007; 31:109–118
DOI: 10.1002/er
114 - AHIN
A. D. S

Table I. Angström equation parameters and Re values for all stations.

Angström New Met.HT
Angs. HT
New Met. ST
Angs. ST

Re equation Measured HT Measured HT Measured ST Measured ST
Station Rep Ren a b R2 R2 R2 R2
Adana 0.125
0.269 0.331 0.286 0.91 0.94 0.84 0.57
Ankara 0.240
0.266 0.304 0.336 0.94 0.94 0.94 0.89
I_stanbul 0.241
0.266 0.288 0.366 0.87 0.97 0.88 0.87

used. In this paper, for three mentioned stations ST and HT values are estimated by Angström
and proposed methods.
Adana station HT estimations by proposed method are compared by measurements through
the regression technique on the basis of the coefficient of determination (R2). The same
procedure is also used for terrestrial variables estimation through Angström equation. Proposed
method and Angström equation estimations of HT with high R2 are given in Table I. It is
observed that Angström equation estimations are better than proposed method in Adana
station. On the other hand, if time series comparison is considered then generally measured data
are represented better except at maximum and minimum values. At maximum values
overestimations occur, but at minimum values underestimations exist in Adana station. The
same procedure is applied for sunshine duration estimation in Adana, and it is seen that
proposed method estimation of sunshine duration is better than Angström equation. Especially,
sunshine duration estimation is not meaningfully represented by Angström equation. In other
words, R2 attached with the Angström equation is not meaningful (Table I). However, in time
series comparison Angström equation appears weaker for terrestrial sunshine duration
estimation. There is a physical problem about the Angstrom equation estimation. In one of
the months, Angström equation estimates sunshine duration as zero which is not physically
possible. In other words, during one month absolute closed conditions could not be observed at
these latitudes especially in this station (Figure 3).
Estimations in Ankara station are made by two methods. It is seen that terrestrial solar
irradiation by Angström equation estimation is better than the proposed method. Like Adana,
both methods have R2 values higher than 0.94 and at maximum values of some months
overestimation occur by proposed method. It should be remembered that, R2 value of the
proposed method is 0.94 which means that correlation coefficient between measured and
estimated value is 0.96. This result is very representative for estimation purposes. In addition to
HT, measured sunshine durations, ST values are also estimated by both methods. As a result, it
is understood that better estimations result through the proposed method. R2 value of the
proposed method is higher than Angström equation results (Table I). Time series approach
shows that the proposed method is more representative than Angström approach. Similar to
Adana station one month has physically impossible result by Angström equation (Figure 4).
Estimations in Istanbul by both methods are compared and terrestrial solar irradiation
measurements are estimated. It is seen that R2 value for Angström method estimation is 0.97.
This means that correlation coefficient of this representation is 0.98 which is a good result.
However, the proposed method has R2 value as 0.87 which is also a good representation, but not
better than Angström approach. In the case of time series graphics, regression technique gives
some misleading information. All maximum values of Angström estimations are higher than

Copyright # 2006 John Wiley & Sons, Ltd. Int. J. Energy Res. 2007; 31:109–118
DOI: 10.1002/er
 NEW FORMULATION FOR SOLAR IRRADIATION 115

14

12

10
Sunshine Duration

Measured ST
2 Est.ST
ANG.Est.ST

0
1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 101 106 111 116 121 126
Time (Month)

Figure 3. Sunshine duration data comparison by suggested method and Angström

equation for measured data in Adana

14.00
Measured ST
Est. ST
12.00 ANG.Est.ST

10.00
Sunshine Durat

8.00

6.00

4.00

2.00

0.00
1.00 10.00 19.00 28.00 37.00 46.00 55.00 64.00 73.00 82.00 91.00 100.00 109.00 118.00 127.00

Time (month)

Figure 4. Sunshine duration data comparison by suggested method and Angström

equation for measured data in Ankara

measurements. In other words, over-estimations occur in this station’s HT values (Figure 5(a)).
Physically impossible values are estimated by Angström equation. Finally, measured sunshine
duration, ST values are estimated by both methods. It is observed that the proposed method
gives better results than the Angström approach (Table I). Furthermore, Angström estimations
result in some physically impossible values. In this station, some sunshine duration estimations
have negative values (Figure 5(b)).

Copyright # 2006 John Wiley & Sons, Ltd. Int. J. Energy Res. 2007; 31:109–118
DOI: 10.1002/er
116 - AHIN
A. D. S

Measured HT
900
Est. HT
ANG.Est.HT
800

700

600
Solar Irradiat

500

400

300

200

100

0
1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97 103 109 115 121 127
(a) Time (Month)

14
Measured ST
Est. ST
12
ANG.Est.ST

10

8
Sunshine Durat

0
1 7 13 19 25 31 37 43 49 55 61 67 73 79 85 91 97 103 109 115 121 127

-2
(b) Time (Month)

Figure 5. (a) Solar irradiation data comparison by suggested method and Angström equation for
measured data in I_stanbul; and (b) sunshine duration data comparison by suggested method and Angström
equation for measured data in I_stanbul.

In addition to this, for the accuracy of the proposed and classical Angström models mean bias
error (MBE), root mean square error (RMSE) and relative error (RE) are used. The following
results are obtained from the comparison of these models. It is not easy to see differences
between estimated and measured values by using MBE except for I_stanbul. In Adana and

Copyright # 2006 John Wiley & Sons, Ltd. Int. J. Energy Res. 2007; 31:109–118
DOI: 10.1002/er
 NEW FORMULATION FOR SOLAR IRRADIATION 117

Table II. Statistical error parameters for models comparison.

Station Error (%) New Met. HT Angs. HT New Met. ST Angs. ST.
Adana MBE 0.361 0.404 0.171
0.921
RMSE 12.627 8.786 12.436 26.019
RE 9.556 7.513 9.556 20.867

Ankara MBE 2.649 0.782
1.666
0.815

RMSE 12.802 5.566 11.877 84.206
RE 12.016 4.886 12.015 14.182

I_stanbul MBE 10.323 42.869 18.065
0.056

RMSE 21.869 52.178 26.019 116.745
RE 16.614 27.198 20.758 21.026

Ankara MBE values are approximately equal to each other but in I_stanbul MBE of HT values
that estimated from Angström equation, is 42.86% that is very high for engineering approaches.
When we look at RMSE values, it is seen that as a result of summation of square differences,
these errors are higher than MBE and generally bigger than 10% except errors of estimated HT
values for Adana and Ankara by Angström equation. But other RMSE values are very high for
HT and ST that is estimated by Angström equation, and it is clearly seen that Angström
equation is not a good approach for these parameters. Especially in I_stanbul unacceptable errors
are estimated by classical approach. One of the other comparison methods is the relative error
(RE) approach that is a very useful tool for engineering calculation. It is seen that for all station,
RE values of Angström equation are higher than errors of new proposed method except HT
values in Adana and Ankara that are estimated by Angström equation (Table II).

5. CONCLUSION

In this paper, a novel method is suggested for solar irradiation and sunshine duration
estimations. In addition, atmospheric effects to extraterrestrial solar irradiation and length of
day can also be evaluated by this method. The proposed method is compared with Angström
equation, and it is found that the former method has a set of advantages. Both methods results
are compared and some physical impossibilities are observed in the Angström estimations. At
all stations, minimum values are estimated by Angström equation as nearly zero values or less
than zero. Istanbul station solar irradiation estimations approximate to extraterrestrial solar
irradiation. These results show that Angström estimates might lead to unreal conclusions. In
addition, the proposed method does not use least square method, and there are no constant
parameters or procedural restrictions and assumptions.

NOMENCLATURE

a, b =Angström coefficients
ECMWF =European Center of Medium Weather Forecast

Copyright # 2006 John Wiley & Sons, Ltd. Int. J. Energy Res. 2007; 31:109–118
DOI: 10.1002/er
118 - AHIN
A. D. S

HE =extraterrestrial solar irradiation (cal cm
2- day
1)

HT =measured terrestrial solar irradiation (cal cm
2-day
1)
MBE =mean bias error
Re =reduction ratio of extraterrestrial ratio
RE =relative error (%)
RMSE =root mean square error
SE =length of day (h)
ST =measured sunshine duration (h)

REFERENCES
Angström A. 1924. Solar and terrestrial radiation. Quarterly Journal of the Royal Meteorological Society 50:121–125.
DMI. 2005. http://www.meteor.gov.tr/2005/genel/meteorolojikaletler/radyasyonolc.htm
Dogniaux R, Lemonie M. 1983. Classification of radiation sites in terms of different indices of atmospheric transparency.
In Proceedings of the EC Contactor’s Meeting on Solar Radiation Data, Solar Energy R of D in the EC, series F, vol. 2.
Reidel, Dortrecht, 94–105.
Duffie JA, Beckman WA. 1991. In Solar Engineering of Thermal Processes (2nd edn). Wiley: New York.
Gopinathan KK. 1988. A general formula for computing the coefficients of the correlation connecting global solar
radiation to sunshine duration. Solar Energy 41(6):499–502.
Gueymard C, Jindra P, Estrada-Cajigal V. 1995. A critical look at recent interpretations of the Angström approach and
its future in global solar radiation prediction. Solar Energy 54(5):357–363.
Iqbal M. 1983. An Introduction do Solar Radiation. Academic Press: Toronto.
Jain PC. 1990. A model for diffuse and global irradiation on horizontal surfaces. Solar Energy 45(5):301–308.
Martinez-Lozano JA, Tena F, Onrubia JE, Delarubia J. 1984. The historical evolution of the Angström formula and its
modifications: review and biography. Agriculture and Forest Meteorology 33(2–3):109–128.
Ögelman H, Ecevit A, Tas¸demiroǧlu E. 1984. Method for estimating solar radiation from bright sunshine data. Solar
Energy 33(6):619–625.
Rietveld MR. 1978. A new method for estimating the regression coefficients in the formula relating solar radiation to
sunshine. Agricultural Meteorology 19:243–252.
Sabbagh JA, Saying AAM, El-Salam EMA. 1977. Estimation of the total solar radiation from meteorological data.
Solar Energy 19:307–311.
Soler A. 1986. On the monthly variations in the atmospheric transmission for cloudless skies as inferred from the
correlation of daily global radiation with hours of sunshine for Spain. Solar Energy 37:253–238.
Soler A. 1990. Monthly specific Rietveld’s correlations. Solar and Wind Technology 7(2–3):305–306.
Suehrcke H. 2000. On the relationship between duration of sunshine and solar radiation on the earth’s surface:
Angström’s equation revisited. Solar Energy 68(5):417–425.
Swartman RK, Ogunlade O. 1967. Solar radiation estimates from common parameters. Solar Energy 11:170–172.
S- ahin AD, Kadioǧlu M, S- en Z. 2001. Monthly clearness index values of Turkey by harmonic analysis approach. Energy
Conversion and Management 42:933–940.
S- ahin AD, S- en Z. 1998. Statistical analysis of the Angström formula coefficients and application for Turkey. Solar
Energy 62:29–38.
S- aylan L, S- en O, Toros, H, Arısoy A. 2002. Solar energy potenial for heating and cooling systems in big cities of Turkey.
Energy Conversion and Management 43:1829–1837.
S- en Z. 2001. Angström equation parameters estimation by unrestricted method. Solar Energy 71(2):95–107.
S- en Z, Öztopal A, S- ahin AD. 2001. Application of genetic algorithm for determination of Angström equation
coefficients. Energy Conversion and Management 42:217–231.
S- en Z, S- ahin AD. 2001. Solar irradiation polygon concept and application in Turkey. Solar Energy 68(1):57–68.

Copyright # 2006 John Wiley & Sons, Ltd. Int. J. Energy Res. 2007; 31:109–118
DOI: 10.1002/er