The 6 basics for handling MET data in the PVSyst manual

Table of Contents

• Roles You Should Know Before Handling MET Data in PVSyst

• MET data basics 1: Check solar radiation and air temperature separately

• MET Data Basics 2: Understanding the Differences Between Monthly Data and Hourly Data

• MET Data Basics 3: Confirm the Offset Between the Observation Site and the Planned Site

• MET Data Basics 4: Don't Overlook Missing or Anomalous Values

• MET data basics 5: Be mindful of converting to solar irradiance on tilted surfaces

• MET Data Basics 6: Examining the connection with simulation results

• Workflow for using MET data in practice while reading the PVSyst manual

• Common mistakes in MET data verification and how to prevent them

• Summary

Roles to Know Before Handling MET Data in PVSyst

One of the items that people who start reading the PVSyst manual tend to stumble over first is the handling of MET data. In photovoltaic system simulations you enter many settings such as module capacity, inverter specifications, azimuth, tilt, wiring losses, and shading conditions. However, the foundation of the energy yield is the meteorological conditions. No matter how precisely you enter the system parameters, if the assumptions for solar irradiance or temperature are off, the resulting annual and monthly energy yields can differ significantly from reality.

MET data refers to meteorological data used in simulations. It mainly includes information such as solar irradiance, air temperature, and, in some cases, wind speed and humidity. In photovoltaic power generation, solar irradiance is the input energy for generation, and air temperature affects module temperature and output degradation. Therefore, MET data should not be regarded as mere auxiliary information but as the starting point for power generation calculations.

When reading the PVSyst manual's explanation of MET data, it is important not just to follow "which screen to import it on" but to be aware of "which calculations that data will be used in." For example, if the annual energy production is lower than expected, you may be quick to suspect loss settings or equipment selection, but in reality the cause can also be the location, period, units, or interpolation method of the meteorological data you imported.

In solar power planning, the basis for expected generation is required in various situations, such as business feasibility assessments, design comparisons, explanations to financial institutions, proposals to clients, and post-construction performance comparisons. At those times, simply saying “it’s correct because it was calculated with PVSyst” is insufficient. Being able to explain which MET data were used, how they were imported under what assumptions, and the degree of uncertainty involved leads to practical trust.

This article outlines six fundamental perspectives for those reviewing MET data in the PVSyst manual. Rather than rote memorization of operational steps, understanding the meaning of the input data, the points to check, and the impact on results will improve the accuracy and explanatory power of your simulations.

MET Data Basics 1: Check Solar Radiation and Temperature Separately

When looking at MET data, the first things you want to check are solar irradiance and air temperature. Both of these affect the output of photovoltaic power generation, but their roles are very different. Solar irradiance is an input condition that works to increase generation, while air temperature affects generation efficiency through module temperature. Regions with higher irradiance tend to produce more power, whereas regions with higher temperatures are more prone to output decreases caused by rises in module temperature.

When reviewing the MET data screen in the PVSyst manual, it is important not to be reassured by the annual solar irradiation alone. Even if the annual irradiation appears sufficiently high, the generation curve will change depending on whether that irradiation is extremely concentrated in summer or remains stable in winter. Especially for self-consumption projects aimed at end users, monthly and hourly generation patterns can be more important than the annual total. You should check not only the annual value but also the month-to-month variability.

For temperature, we consider not just whether the average temperature is high or low, but also what temperatures are like during periods of strong solar irradiance. Solar modules experience reduced output at high temperatures, so if months with high solar irradiance also have high temperatures, the expected energy yield may not increase as much as anticipated. Conversely, in cool regions with ample sunlight, module temperatures can be advantageous.

In practice, solar irradiance and air temperature are often provided as a single dataset, but when checking it’s easier to detect problems if you examine them separately. For example, if power output is lower than expected you can determine whether it’s due to low irradiance or large temperature-related losses. Understanding the contents of the MET data before inspecting the loss chart makes interpreting the results much easier.

Also, there are multiple ways to express solar irradiance, such as horizontal-plane irradiance, direct irradiance, and diffuse irradiance. The calculations used in PVSyst involve the process of deriving tilted-surface irradiance from horizontal-plane meteorological data according to the installation’s tilt and azimuth. Therefore, if you load data without checking the irradiance field names and units, you may end up running simulations under unintended assumptions.

When checking MET data, first look at the types of solar irradiance, units, annual values, and monthly values. Next, look at the annual average temperature, monthly trends, and the relationship with periods of strong solar irradiance. Checking in this order makes it easier to organize the factors that affect power generation.

MET Data Basics 2: Understanding the Difference Between Monthly Data and Hourly Data

What you need to understand when handling MET data in the PVSyst manual is the difference between monthly data and hourly data. In solar power generation simulations, it is important to deal not only with annual or monthly representative values but also with changes on an hourly basis. Because generation varies from moment to moment depending on solar altitude, weather, temperature, and shading conditions, the time resolution affects the results.

Monthly data summarize each month’s solar radiation and temperature as representative values. They make it easy to grasp the overall picture and are useful for observing regional climate trends. However, they cannot capture details such as when during the day solar radiation is highest, how much morning and evening shadows affect conditions, or in which time period peak output occurs. Relying only on monthly data can cause you to overlook time-of-day power generation characteristics.

On the other hand, time-based data are data that include meteorological conditions at intervals such as one hour. Because they can more finely reflect changes in solar irradiance and temperature, they are suitable for time-series calculations of power generation. In particular, when examining self-consumption solar power, battery integration, comparisons with demand curves, peak shaving, or the relationship with time-of-use tariffs, time-based data become especially important.

However, time-series data are not necessarily highly accurate. If the quality of the original data is poor, higher temporal resolution will not increase reliability. Conversely, when time-series data are generated from monthly values, some modeling and estimation are involved. Therefore, when reading the PVSyst manual you should be aware whether the data are time-series data based on actual measurements or data generated from monthly values.

In practice, there is also a workflow in which monthly data are used in the initial study stage, and hourly data are checked for detailed design and final evaluation. In the initial stage the purpose is to obtain rough estimates of power generation and to compare design proposals, with speed prioritized over detail. On the other hand, for business decisions and pre-contract explanations, it is necessary to look more carefully at the accuracy, sources, and generation methods of the meteorological data that serve as the basis.

When handling time-series data, check seasonal solar irradiance patterns, whether there are extremely high or low values, and whether any unnatural irradiance appears at night. Even if the data columns are complete, if the values themselves are anomalous, the simulation results become difficult to trust. A successful load in PVSyst’s interface is not the same as the data being valid for practical use.

Understanding the difference between monthly data and hourly data can also be helpful when explaining results. For example, saying "These results are a rough estimate based only on annual values" versus "We have verified them using hourly weather data, including shading and temperature losses" will convey different levels of confidence to the listener.

MET Data Basics 3: Check for discrepancies between observation sites and the planned site

One aspect that is easy to overlook when handling MET data is the mismatch between the observation point and the planned site. If the planned site for a solar power installation is in mountainous areas, coastal areas, basins, urban areas, snowy regions, or high-temperature regions, simply using meteorological data from a nearby representative location may not adequately reflect the local conditions.

When checking how to load MET data in the PVSyst manual, be sure to verify not only which data to select but also the location information for that data. Latitude, longitude, elevation, and distance from the planned site are important. Even within the same prefecture, coastal and inland areas, and plains and mountainous areas differ in solar irradiance, temperature, cloud cover, and snow conditions. If you simply decide "it's fine because it's a nearby city," you may later have difficulty explaining the power generation.

Elevation differences, in particular, tend to affect temperature. Locations at higher elevations are likely to have lower temperatures, which can be advantageous for module temperature. Conversely, sites that are prone to clouds, fog, or snowfall may receive less solar radiation than expected. When the elevation of the observation point differs substantially from that of the planned site, the impact of that difference on the results needs to be considered.

Also, potential sites for solar power generation are not necessarily located near meteorological observation stations. In large-scale ground-mounted projects, areas such as forest development sites or unused suburban land may be targeted. In such locations, the nearest meteorological data alone may not adequately represent the on-site solar radiation conditions. It is important to compare multiple candidate datasets and assess which data are reasonable to adopt.

When checking the offset between the observation point and the planned site, look not only at distances on the map but also at differences in terrain. In adjacent areas separated by mountains, the climate can change even over short distances. Coastal areas may be affected by sea breezes and cloud formation, while inland areas may experience temperature swings and fog. Do not confine your assessment to the PVSyst input screen; you need a perspective that compares the planned site against its geographic characteristics.

In practice, recording the station names and coordinates of the MET data used and the distances to the project site makes it easier to explain later. When asked during a power generation review, "Why did you use this meteorological data?", being able to explain reasons such as distance, elevation, climate classification, and data quality will increase confidence in the simulation.

The purpose of reading the PVSyst manual is not simply to learn how to operate the software. It is to become able to explain the assumptions behind the calculation results. The discrepancy between the observation site and the planned site is precisely where that explanatory ability is tested.

MET Data Basics 4: Don't Overlook Missing or Anomalous Values

When loading MET data, it is important to check for missing and anomalous values. Even if the data format is correct and it loads successfully in PVSyst, the dataset may contain gaps or implausible numbers. If you proceed with the simulation as is, the annual energy output and monthly generation results can be unexpectedly affected.

Missing values are situations where values that should be present—such as solar radiation or temperature—are missing. They occur due to instrument outages, communication errors, faults during data conversion, mistakes when creating files, and the like. For hourly data, a year's worth contains many rows, so it is difficult to fully verify by visual inspection alone. However, if missing values are concentrated in specific months or times of day, the power generation for that period will appear anomalous.

Anomalous values are values that are physically impossible or that deviate extremely from surrounding values. For example, cases include a large amount of solar radiation recorded at night, temperatures that are far outside what is expected for the local climate, solar radiation being continuously zero, or values being off by orders of magnitude due to unit conversion errors. Because such anomalous values directly affect calculation results, it is essential to check for them after loading.

When working through the PVSyst manual, it's easy to focus on file formats and import methods. However, what matters in practice is checking the trends in the data after importing. Use monthly summaries and graph displays to look for months that suddenly dip or are unnaturally high. Also check whether the seasonal variations in solar irradiance and temperature are natural for the project site’s climate.

Be careful about differences in units. If you mistake whether the unit for solar radiation is per hour, per day, or a monthly total, the results can change dramatically. You also need to check whether temperature is being handled in Celsius (°C) or has been converted from another unit. It is important to verify not only the file column names and descriptions but also the plausibility of the values based on their magnitudes.

When correcting missing or anomalous values, record how you made the corrections. Whether you simply interpolated using the surrounding values, replaced them with other data, or excluded the affected period changes the meaning of the results. It is desirable to keep a history of the data-processing steps so you can later explain the basis for the reported power generation.

Simulations proceed on the assumption that the input data are correct. Just because PVSyst does not report an error does not mean the quality of the data is guaranteed. By making it a habit to check for missing or anomalous values, you can further improve the reliability of the results.

MET Data Basics 5: Be aware of conversion to solar radiation on tilted surfaces

In photovoltaic power generation, modules are often installed with a tilt rather than horizontally. Therefore, instead of using the horizontal-plane irradiance included in MET data directly for generation calculations, it is necessary to consider the irradiance incident on the module surface according to the installation surface's azimuth and tilt. This is an important point for understanding MET data in the PVSyst manual.

In meteorological data, solar irradiance on the horizontal plane is sometimes used as basic information. However, actual solar modules are deployed in a variety of conditions—south-facing, east- or west-facing, low-tilt, steep-tilt, or installed to match roof pitch. Even with the same horizontal-plane irradiance, the amount of irradiance reaching the module surface changes when orientation and tilt vary.

In PVSyst, the irradiance received on an inclined surface is calculated according to the installation conditions. Therefore, the way MET data’s horizontal plane irradiance, direct component, and diffuse component are handled affects the final energy yield. In particular, in regions with a high amount of diffuse irradiance or frequent cloudy conditions, it is necessary to be aware of the effects of conversion to the inclined surface. It is insufficient to simply assume “because a region has high irradiance, the energy production will also be high.”

When comparing design options, the way irradiance on inclined surfaces is considered is important. For example, slightly changing the tilt angle affects not only annual power generation but also the balance between summer and winter output. An east–west layout tends to spread peak production, while a south-facing layout tends to have higher output during specific times of day, so differences due to layout also emerge. These differences arise from the combination of the underlying MET data and the installation conditions.

Also, in roof-mounted projects the roof pitch and orientation are often already fixed, so it may not be possible to achieve the ideal angle. In such cases, it is important not only to set the tilt and azimuth in PVSyst but also to check how much solar irradiance will be received under those conditions. For ground-mounted projects, because the tilt angle is determined by balancing power output, site development conditions, racking costs, wind loads, snow loads, and maintainability, it is necessary to examine the relationship with meteorological data.

Being mindful of the conversion to irradiance on an inclined surface makes MET data more practically meaningful. The data does not simply show "regional irradiance." When tied to design conditions, it becomes the input that determines the actual energy a module receives. When reading the PVSyst manual, it's important not to treat MET data and azimuth and tilt settings as separate items, but to view them together as a single, integrated set of calculation conditions.

MET Data Basics 6: Examining the Connection with Simulation Results

Checking MET data does not end when it is loaded. Ultimately, you need to judge its validity by comparing it with the simulation results. When reviewing annual energy production, monthly production, PR, loss diagrams, and so on in PVSyst, being aware of how the MET data is reflected in the results makes it easier to understand the meaning of the numbers.

For example, if a monthly generation graph shows a particular month with lower output, you need to distinguish whether the cause is low solar irradiance, temperature-related losses, shading effects, or system constraints. In that case, if you first check the monthly solar irradiance and temperature in the MET data, it becomes easier to track down the cause. It is natural for generation to be low in months with low solar irradiance, but if irradiance is high and generation does not increase, you can suspect other loss factors.

When examining PR, it is important to understand its relationship with MET data. PR is used as an indicator to evaluate system performance, but it cannot be completely separated from the influence of meteorological conditions. In high-temperature regions, thermal losses tend to be larger, and in environments with frequent shading or low-irradiance conditions, the expected performance may be difficult to achieve. When PR appears low, you should not suspect only the equipment or design; you need to consider the meteorological conditions as well.

In the loss diagram, you can check the flow of losses from solar irradiance to the final AC output. The first input condition here is the solar irradiance based on meteorological data. Before looking at items such as temperature loss, IAM loss, shading loss, wiring loss, and conversion loss, you should first verify whether the input solar irradiance is reasonable. If the input is too large, the generated energy will appear too high; if the input is too small, the generated energy will appear low even if the design is not poor.

Also, when comparing multiple cases, confirm that the same MET data is being used. If you compare design proposal A and design proposal B and the meteorological data differ, you will not be able to tell whether the differences are due to equipment conditions or weather conditions. If the purpose of the comparison is layout or equipment selection, it is fundamental to standardize the MET data to the same conditions. Conversely, if you want to examine sensitivity to differences in meteorological data, make clear which data were changed.

MET data is important when comparing to actual generation performance. If actual performance after start of operation is lower than the simulation, check how the solar radiation conditions in the actual year compared with the long-term average before suspecting equipment malfunction. If the MET data used in the simulation represents the long-term average and the actual year was a year with below-average solar radiation, it is not unusual for the generated output to be lower. Being able to explain this difference is also important in post-operation evaluation.

By linking MET data with simulation results, the numbers from PVSyst become not simply output results but interpretable evaluation documentation. When reading the manual, it is important not to memorize the input screen, calculation conditions, and result screen separately, but to understand them as a single flow.

Workflow for Using MET Data in Practice While Reading the PVSyst Manual

When handling PVSyst MET data in practice, start by organizing the planned site's conditions. Confirm the site location, latitude and longitude, elevation, surrounding terrain, installation type (roof-mounted or ground-mounted), possibility of snow or high temperatures, shading factors, and so on. Meteorological data should not be selected in isolation, but chosen together with the planned site's conditions.

Next, review the candidate MET data to be used. The options vary by project, such as standard datasets that can be used routinely, externally sourced datasets, and datasets based on on-site observations. In the initial stage it is realistic to make rough estimates using representative data, and in the detailed stage to check data that reflect conditions closer to the project site.

Before loading data, verify the file format, units, time range, and the meaning of each column. In particular, when using external data, mistakes in column order or unit conversion can lead to files that appear to load correctly but contain incorrect values. Refer to the relevant sections of the PVSyst manual to understand what each item represents.

After loading, we check the annual values, monthly values, graphs, and whether there are any anomalies. If we find any unusual trends at this point, we do not immediately proceed with the simulation; instead, we investigate the cause. We determine whether the problem lies with the data itself, with units or loading settings, or with the special characteristics of the planned site.

Then enter the azimuth, tilt, system configuration, and loss settings, and run the simulation. When reviewing the results, look not only at the annual energy yield but also at the monthly energy yield, PR, loss diagram, temperature losses, shading losses, and so on. By checking whether the trends in the MET data naturally align with the trends in the results, it becomes easier to spot configuration errors or data inconsistencies.

Finally, record the information about the MET data used. Leaving the data name, site, period, units, whether corrections were applied, reasons for adoption, and any notes or caveats will be useful for internal reviews, explanations to clients, and future re-evaluations. Rather than relying solely on the PVSyst project file, it is important to organize the assumptions in documentation.

If you make this workflow a habit, you can progress from operating while consulting the PVSyst manual to producing simulations you can explain in professional practice. MET data is not something you load once and forget; it is important information that links the project site, design conditions, results, and presentation materials.

Common mistakes when checking MET data and how to prevent them

One common mistake with MET data is using data from a seemingly nearby site without thoroughly checking it. Even if a site is close to the project location, differences in elevation, terrain, or distance from the sea can alter solar radiation and temperature. To prevent this, it is important to verify the coordinates and elevation of the data site and record any differences from the project location.

Another common mistake is judging based only on annual solar radiation. Even if the annual value is high, there may be little solar radiation during the periods when you want to generate power or when demand is high. For projects considering self-consumption or battery integration, you need to examine monthly and time-of-day trends. Check not only the annual generation but also the relationship between monthly generation and solar radiation.

Confusing units are also a major problem. When importing external data, mistaking the units of solar radiation, the time unit, or whether values are cumulative or averages can greatly change the results. By checking the column definitions before importing and confirming that the magnitudes of the values look natural after importing, you can more easily prevent mistakes.

Another common mistake is failing to verify the period covered by the data. Whether the data are for a single specific year or a long-term average changes the meaning of the results. In business viability assessments, long-term trends are often emphasized, while in comparisons of actual performance the weather conditions of the year in question are important. It is necessary to clearly state which period the data cover.

When duplicating simulation cases for comparison, MET data can sometimes change unintentionally. When comparing design proposals, keeping meteorological conditions consistent is fundamental. If differences appear in the comparison results, take care not to confuse whether they are caused by design differences or by differences in the MET data.

Another mistake is looking only at the results and not returning to the MET data. When you see results indicating high or low power generation, always check for consistency with the input meteorological conditions. By going back and forth between the results screen and the MET data screen, it becomes easier to spot configuration errors or implausible assumptions.

These mistakes are not limited to people who are inexperienced with PVSyst. Even practitioners who are accustomed to the work can, when the number of projects grows, let checks become a rote process and overlook basic verifications. That is why it is important to establish the MET data check items and review them from the same perspectives each time.

Summary

In the PVSyst manual, when working with MET data it is important not simply to learn the operation of importing meteorological data, but to understand the meaning of the data and their impact on the results. MET data are the starting point for photovoltaic power generation simulations and affect a wide range of outputs, including annual energy yield, monthly generation, PR, loss diagrams, financial viability assessments, and comparisons with actual performance.

As basic points to keep in mind: check solar irradiance and temperature separately, understand the difference between monthly and hourly data, look for discrepancies between the observation site and the planned site, check for missing or anomalous values, be aware of conversion to solar irradiance on tilted surfaces, and examine the connection to the simulation results. Simply mastering these six points will significantly change your ability to interpret PVSyst results.

In the design and evaluation of solar power systems, being able to explain the numbers is more important than the numbers themselves. If you can explain which MET data were used, why that data was selected, how much it differs from the project site, and how it affects power generation, you will be more persuasive in internal reviews and when explaining to customers.

When reading the PVSyst manual, don’t just follow the on‑screen steps in sequence; be mindful of the flow of inputs, calculations, results, and explanations. Learning to handle MET data correctly will not only improve the accuracy of energy yield simulations but also enhance the quality of design decisions and proposal materials. Understanding MET data is an essential foundation you cannot avoid when using PVSyst in professional practice.