5 Methods to Calculate Solar Power Generation from Regional Sunshine Hours

When calculating photovoltaic generation, many practitioners first approximate annual kWh from the system capacity in kW. This is a very convenient entry point, but when you want to reflect regional differences in a more tangible way, thinking in terms of sunshine duration can be helpful. In particular, when you want to get a rough sense of whether an area tends to be sunny or cloudy, or how much sun it receives seasonally, sunshine duration is an intuitively easy-to-understand indicator. The Japan Meteorological Agency defines sunshine duration as "the time during which direct solar irradiance at the ground is at least 0.12 kW/㎡ (0.01 kW/ft²)," and it is an observed value resulting from the effects of weather and clouds. ([Japan Meteorological Agency][1])

However, the first point to grasp here is that sunshine duration is not the same as power generation itself. Sunshine duration is information about "how many hours the sun shone," and it is not the solar irradiance energy amount that directly affects photovoltaic power generation. The Japan Meteorological Agency organizes instantaneous solar radiation in kW/㎡ (kW/ft^2) and cumulative radiation in MJ/㎡ (MJ/ft^2), treating them as quantities separate from sunshine duration. In other words, when calculating power generation using a region’s sunshine duration, you should not directly equate it to kWh; instead, you need to reinterpret it through system capacity and correction factors. ([Japan Meteorological Agency][2])

On the other hand, that does not mean sunshine duration is useless. The commentary for NEDO’s standard meteorological database shows that, for locations without solar radiation observations, solar radiation is compiled using a sunshine–radiation model that estimates solar radiation from observed sunshine duration. In other words, sunshine duration is not the final answer for power generation calculations, but it has sufficient value as an important intermediate piece of information leading to solar radiation. In practice, it is important to use it with an understanding of this role. ([NEDO][3])

This article organizes five methods for estimating solar photovoltaic generation using local sunshine hours. They are: a rough method based on annual sunshine hours; a method that accumulates month by month; a method that reflects weather differences using sunshine rate; a method that improves accuracy by converting sunshine hours to solar irradiance; and a method that applies regional corrections using actual measured values. To make this understandable without diagrams, I will explain in words what to connect and in what order. ([NEDO][4])

Table of Contents

• Clarify the assumptions for calculations based on sunshine hours

• Method 1: Estimate from annual sunshine hours

• Method 2: Accumulate power generation using monthly sunshine hours

• Method 3: Use the sunshine ratio to correct for cloudy-day effects

• Method 4: Convert sunshine hours into solar radiation to improve accuracy

• Method 5: Apply regional corrections using actual performance data

• Common misconceptions in sunshine-hours-based calculations

• Summary

Clarify the assumptions when calculating from sunshine hours

When considering power generation from sunshine hours, the first thing to clarify is the difference between kW and kWh. kW indicates the size of the system, in other words the system capacity. For example, expressions like a 5 kW system or a 10 kW system refer to how much capability the system has. On the other hand, kWh is the amount of electricity actually generated. If a 5 kW system generates at full capacity for 1 hour ideally, that is 5 kWh, and for 2 hours it is 10 kWh. In other words, you only get the generated energy in kWh by multiplying the system capacity in kW by how many hours it can generate. ([気象庁][2])

If we translate this relationship into sunshine hours, it becomes clear that sunshine hours can be used as a guideline for "hours during which power can be generated." However, it should be noted that one hour of sunshine does not necessarily equal one hour of the system operating at full output. Sunshine hours refer to the time direct sunlight strikes the ground, and depending on solar elevation, season, cloud cover, and atmospheric conditions, the solar energy received during the same 1 hour can vary. That is why, rather than determining generation solely from sunshine hours, it is necessary to consider "how to convert sunshine hours into equipment-equivalent generation hours." ([気象庁][1])

Also, while sunshine duration is very easy to understand when looking at regional differences, it does not automatically account for differences in equipment conditions. Even within the same region, power generation differs between a favorable south-facing surface and east- and west-facing surfaces, and at sites with shading, power generation drops even if sunshine duration is sufficient. In other words, sunshine-duration-based calculations are useful as an entry point, but ultimately you need to add conditions such as orientation, tilt, shading, and losses to convert them into site-specific values for each installation. ([NEDO][4])

Furthermore, the Japan Meteorological Agency’s “sunshine duration” is treated, as part of the results of regional meteorological observations, not only as measurements from surface meteorological observations but, in some cases, as estimated values derived from meteorological satellites. In other words, while regional sunshine duration data are convenient as an indicator of regional differences in practical work, it is better to understand that they are somewhat a different layer of information than the amount of solar radiation actually received by the installation surface itself. Clarifying this point can also reduce the risk of over-relying on calculations based on sunshine duration. ([Japan Meteorological Agency][5])

Method 1: Estimating from annual sunshine hours

The first method is a way to estimate power generation from the annual hours of sunshine. This is the simplest approach and is suitable when you want to grasp the outline of regional differences. The idea is to regard the longer the hours of sunshine in a region for a given system capacity in kW, the greater the likely power generation, and determine the rough level of annual generation. Rather than directly converting hours of sunshine to kWh, it is practical to use the relative amount of annual sunshine as a criterion for selecting an annual generation guideline per kW. ([Japan Meteorological Agency][1])

For example, even with the same system capacity of 5 kW, in areas with long sunshine hours and many clear days the annual generation per 1 kW tends to be estimated higher. Conversely, in regions with frequent cloudy weather or snowfall and short sunshine hours, the annual generation per 1 kW should be estimated more conservatively. The formula here takes the form: Annual generation (kWh) = system capacity (kW) × the estimated annual generation per 1 kW appropriate to local conditions. In other words, it is easier to understand sunshine hours as a clue for determining the regional coefficient rather than as a number to directly calculate generation. ([NEDO][6])

The advantage of this method is that it lets you quickly grasp the overall picture. In initial consultations and when comparing system sizes, it’s important to have a sense of roughly how much a 5 kW system will produce annually and how much a 10 kW system will. At that stage, using whether a region has many hours of sunlight or not as one axis of judgement makes the approach far more practical than proceeding with a uniform nationwide figure. ([気象庁][1])

However, this method also has its limitations. The length of sunshine alone cannot adequately represent seasonal differences, cloud thickness, the contribution of diffuse light, or the orientation and tilt of the equipment. In other words, it is useful as an initial estimate at the entry stage, but it should not be used directly as the final value for proposals or as the definitive figure for financial projections. In practice, it is easiest to first grasp the overall outline with this method and then proceed to a monthly or solar-irradiance–based approach. ([NEDO][3])

Method 2: Aggregate power generation by monthly sunshine hours

The second method is to aggregate power generation using monthly sunshine hours. Annual sunshine hours alone may reveal the outline of the system, but seasonal differences become obscured. Because photovoltaic output varies considerably across spring, summer, autumn, and winter, if you calculate from sunshine hours it is much more practical in practice to break them down by month. The Japan Meteorological Agency’s statistics also provide sunshine hours as monthly long-term averages, and these are easy to combine with NEDO’s monthly average solar irradiance data. ([Japan Meteorological Agency][5])

In this method, the length of sunshine hours in each month is used as a reference indicator for that month’s generation-equivalent hours. For example, spring and autumn tend to be relatively stable and longer, the rainy season tends to be shorter, and in winter the sunshine hours themselves tend to be short. This is combined with the installed capacity to form the monthly generation profile, and the 12 months are summed to return to an annual value. In other words, this brings the estimate closer to generation that reflects monthly trends than to an annual lump-sum approximation. (Japan Meteorological Agency [5])

This way of thinking is important because monthly power generation is directly tied to the value of the installation itself. In summer, not only is generation high, but the overlap with air-conditioning loads may make self-consumption easier. In spring and autumn, even if generation is high, surpluses may increase if demand is low. In winter, daylight hours are short and generation may drop considerably. In other words, when viewed month by month, it becomes much easier to understand how the installation is used. ([Japan Meteorological Agency][1])

Also, monthly sunshine duration is useful for roughly capturing trends in cloudiness and seasonal characteristics. Under the Meteorological Agency's definition, sunshine duration is the time during which direct solar radiation exceeds a certain threshold, so it tends to be shorter when there are many cloudy days or precipitation. Therefore, even just looking at differences in monthly sunshine duration makes it easier to grasp seasonal variations in power generation. Of course, solar irradiance itself is more precise, but as an initial assessment it is sufficiently effective. ([Meteorological Agency][1])

Method 3: Correct cloud-induced bias using the sunshine ratio

The third method is to correct for cloud-related deviations using the sunshine rate. When information on sunshine duration is available, looking at it not just as an absolute value but as how much sunlight occurred relative to the theoretical day length or a reference makes the effects of cloudy skies and clouds much easier to read. This is the idea behind the sunshine rate. The Japan Meteorological Agency also indicates that it sometimes uses the previous 1-hour sunshine rate to distinguish between clear and cloudy conditions. In other words, the sunshine rate is convenient as an intermediate indicator to connect weather conditions to power generation. ([Japan Meteorological Agency data][7])

In this method, you assess how sunny a region or month tended to be by dividing the actual sunshine hours for a period by the reference sunshine hours expected for that period. If the sunshine ratio is high, it’s reasonable to conclude that the month was relatively sunny and that power generation can be expected to be higher. If the sunshine ratio is low, cloudy conditions were more frequent, and it may be better to be slightly conservative compared with the input values. In other words, it’s helpful to think of the sunshine ratio as a simple correction that reflects the “degree of sunny versus cloudy conditions” in power generation. ([Japan Meteorological Agency][1])

The advantage of this method is that it makes differences between years and between months easy to read, even for the same region. For example, even in regions that have long sunshine duration on average, if the rainy season is longer in a given year the sunshine ratio will fall, so it's better not to overestimate the power generation. Conversely, in years with more sunny weather the result may be better than the input value. In other words, using the sunshine ratio makes it easy to reflect not only the long-term average but also any weather bias during that period. ([Japan Meteorological Agency][5])

However, there is a caveat here as well. The sunshine rate well represents the presence or absence of clouds and how likely it is to be sunny, but it does not automatically correct for equipment orientation, shading, temperature, or losses. Therefore, the sunshine rate should be used as a simple adjustment to fine-tune power generation, and equipment conditions should be considered separately. In practice, it is easy to apply by slightly raising or lowering the baseline monthly generation according to whether that month’s sunshine rate is high or low. ([Japan Meteorological Agency data][7])

Method 4: Improve accuracy by converting sunshine duration into solar radiation

The fourth method is to improve accuracy by converting sunshine duration into solar irradiance. This approach adds another step of accuracy compared with linking sunshine duration directly to power generation. The NEDO standard meteorological database explicitly states that, for locations without solar irradiance observations, it uses a sunshine–irradiance model that estimates solar irradiance from observed sunshine duration. In other words, rather than relying solely on sunshine duration, converting once into the realm of solar irradiance is considered more practical for estimating power generation. ([NEDO][3])

The point of this method is to reinterpret the "intensity of light," which is hard to grasp from sunshine duration alone, into solar irradiance. Sunshine duration indicates the periods when direct solar radiation exceeded a certain threshold, so it tells you how long the sun shone. However, even within the same one-hour period, the strength of solar radiation and the sun's altitude can differ. Therefore, by moving from sunshine duration to an estimate of solar irradiance, the kWh obtained when multiplied by an installation's kW capacity comes much closer to actual on-site values. In other words, it is easy to understand if you think of sunshine duration as an initial indicator and solar irradiance as a bridge to generation output. ([NEDO][3])

In practice, if you can use NEDO’s monthly average solar irradiance or data broken down by azimuth and tilt, using those will yield higher accuracy than estimating from sunshine duration alone. Even if you only have sunshine duration data on hand, adopting the idea of converting that into solar irradiance lets you move beyond the simplistic view that “longer sunny hours means more power generation.” In other words, sunshine duration is convenient, but linking it to solar irradiance is important to make it more practical. ([NEDO][4])

Also, once this conversion can be made, it becomes easier to organize how to handle orientation and tilt conditions. Sunshine duration itself is a meteorological condition of the site, but what determines power generation is the irradiance on the tilted surface actually received by the installation. By progressing from sunshine duration to solar irradiance, and then adjusting for slope conditions, differences between installations can be explained much more easily. This applies equally to houses, factories, and warehouses. ([NEDO][4])

Method 5 Regionally adjust using actual performance data

The fifth method is to apply regional adjustments using actual performance data. The method of estimating power generation from sunshine hours is very convenient as a starting point, but it may not fully capture locally specific weather patterns or on-site conditions. A strong approach is to use the performance records of existing facilities or nearby similar installations as correction material. In practice, this extra step considerably increases the reliability of the estimates. ([NEDO][3])

For example, suppose a system that was expected theoretically to produce around 10,000kWh per year actually produced about 9,000kWh in practice. The difference may include factors such as occurrence of cloudy weather, shading patterns, high-temperature losses, soiling, and operational conditions. If, when viewed by month, only a particular season is significantly lower, you might want to give stronger consideration to that region’s specific weather tendencies or winter conditions. In other words, actual performance becomes a strong corrective input for bridging the gap between theoretical values and field conditions. ([NEDO][3])

The advantage of this method is that it can reflect regional differences as real-world values rather than as averages. Even within the same prefecture, cloud formation and snow conditions can vary between coastal and inland areas, and between plains and mountainous regions. Databases of sunshine hours and solar radiation provide a strong foundation, but by layering on-site performance records on top of them, estimates can be brought much closer to the actual kWh. In other words, this performance adjustment can be considered the final refinement that brings sunshine-hours–based calculations closer to practical application. ([NEDO][6])

Also, adjustments based on actual performance are very effective for internal explanations. Rather than simply showing theoretical values, being able to say, "For equipment in the same region and for the same use there has been a tendency to drop by about this much in this season, so we are taking a slightly conservative view this time," makes the numbers more persuasive. Add the realism of actual performance to the theoretical values for sunshine hours and solar radiation data. This combination is extremely powerful in practice. ([NEDO][3])

Common misconceptions in calculations based on sunshine hours

One common misconception when estimating electricity generation based on sunshine duration is to assume that the longer the sunshine duration, the greater the generation will be. Sunshine duration can serve as an indicator of how long it was sunny, but even for the same 1-hour period the solar irradiance and the sun's altitude differ, so the generated output will not be the same. For that reason, while sunshine duration is useful as an initial indicator, directly converting it to kWh tends to result in coarse estimates. ([気象庁][1])

The next most common mistake is attempting to explain all regional differences solely by sunshine hours. In reality, power generation is determined by a combination of orientation, tilt, shading, temperature, snowfall, and equipment conditions. Sunshine hours are convenient for assessing regional differences, but they do not automatically account for differences in equipment. It is especially dangerous to link sunshine hours directly to equipment value for roofs with multiple surfaces or sites with shading. ([NEDO][4])

Also, it is a common mistake to stop at looking only at the annual sunshine hours. The strength of sunshine hour data is that it makes differences by month and season easy to read, so it is a waste to use only the annual aggregate. In particular, for projects that consider self-consumption or surplus, the substance of the equipment’s value becomes hard to see unless you look at summer–winter and spring–autumn differences. In other words, if you are thinking in terms of sunshine hours, you should at least break it down by month. ([Japan Meteorological Agency][5])

Summary

To calculate solar power generation from regional sunshine duration, it becomes easier to use if you organize the approach into five concepts: first, understand what sunshine duration indicates as an index; next, estimate from annual sunshine duration; then accumulate using monthly sunshine duration; correct for cloudiness differences with the sunshine ratio; convert sunshine duration into solar irradiance to improve accuracy; and finally apply regional corrections using actual performance data. Sunshine duration is not the generation itself, but it is a very easy-to-understand indicator for reading regional and seasonal differences. ([Japan Meteorological Agency][1])

The important thing is not to treat sunshine hours as a cure-all on their own. Only by adopting the idea of converting them to solar irradiance, accounting for orientation (azimuth) and tilt, and finally correcting for shading and losses does it become easier to convert equipment kW into realistic kWh. In other words, sunshine hours are an excellent starting point, but in practice it is essential to connect them to the subsequent equipment conditions. ([NEDO][4])

Also, if you truly want to improve this calculation accuracy, it is important to accurately capture the on-site conditions. If the orientation of the roof surface, the positions of surrounding obstacles, and the elevation differences are ambiguous, corrections for shading and layout applied to the theoretical values derived from sunlight hours become coarse. In particular, winter shading and differences in usable area vary greatly depending on whether the relative positions on site have been accurately determined. ([NEDO][6])

In that respect, the LRTK, an iPhone-mounted high-precision GNSS positioning device, is highly effective as a means of accurately grasping positional relationships on site. Because it makes it easier to accurately record the positions of roof edges and obstacles in the field, it becomes easier to link to power generation estimates that take into account orientation, shading, and layout conditions. If you want to calculate solar power generation as truly usable figures from regional sunshine hours, properly capturing on-site conditions with means such as LRTK becomes a major advantage in practice. ([NEDO][3])