Origin of the data for Monitor and Forecast

Solargis Monitor and Solargis Forecast solar data comes from three origins, depending on how far the requested time is from the present moment: the Solargis satellite-based model for everything up to the current moment, the Solargis Cloud Motion Vector (CMV) model for approximately the next three hours, and post-processed numerical weather prediction (NWP) models out to DAY+14. Meteorological parameters come from the ERA5 reanalysis and from NWP models. This article sets out which origin applies to which time frame, where satellite data is available, how quickly it arrives, and how often it is updated. The data is delivered through the Solargis Monitor API and the Solargis Forecast API.

Origin of the solar data in Monitor and Forecast

The Solargis Monitor and Solargis Forecast services deliver essential solar data parameters, including GHI (Global Horizontal Irradiance), DNI (Direct Normal Irradiance), and DIF (Diffuse Horizontal Irradiance), which form the foundation for solar resource analysis.

They also provide calculated parameters like GTI (Global Tilted Irradiance) and PVOUT (Photovoltaic Power Output) to support advanced solar energy modeling and project optimization.

Solargis Monitor and Solargis Forecast solar data passes through five time frames, each with its own data origin:

• Historical data – Solargis satellite-based model – from the beginning of the archive to the last completed calendar month (MONTH-1)

• Operational data – Solargis satellite-based model – from the start of the current calendar month to the last completed calendar day (DAY-1)

• Real-time data – Solargis satellite-based model – from the start of the current calendar day (DAY+0) to the current time

• Nowcasting data – Solargis Cloud Motion Vector (CMV) model – from the current time to approximately 3 hours ahead

• NWP forecast data – post-processed NWP model outputs (IFS, ICON, GFS, HRRR) – from approximately 3 hours ahead to DAY+14

The tabs below give the full description of each time frame; Figure 1 shows them on a single timeline.

Figure 1: Data coverage timeline for Solargis Monitor and Solargis Forecast.

Tracking solar data origin with the Cloud Identification Flag (CI_FLAG)

Within the Solargis Monitor and Solargis Forecast services, you can determine the origin of solar data by requesting the CI_FLAG parameter (Cloud Identification Quality Flag – i.e. from which source Solargis obtained the cloud situation) in your XML request. The CI_FLAG parameter labels each data record with one of the following categories:

This system ensures transparency in data provenance, critical for validating solar analyses in your applications.

Tip: For accurate interpretation, use CI_FLAG values with the highest solar data resolution (15 or 10 minutes) to ensure granular insights into data quality and source.

Reanalysis of solar data in Monitor and Forecast

Solargis ensures high data accuracy in Solargis Monitor and Solargis Forecast through systematic reprocessing of historical and operational solar data. Here’s how the timeline works:

Daily reanalysis of DAY-2

• Every day, solar irradiance data is actually calculated for two days: DAY-1 and DAY-2, so that the DAY-2 gets its first reanalysis.

Monthly reanalysis of MONTH-1

At the start of each calendar month (not before 3rd calendar day of the month):

• Solar irradiance data for the previous month is recalculated using finalized atmospheric inputs.

• Atmospheric data is homogenized with historical records to maintain consistency and prevent abrupt changes caused by updates to atmospheric models.

Note: Daily and monthly reanalysis ensures long-term reliability of the historical and operational stages of Solargis Monitor and Solargis Forecast solar data.

For a deeper dive into atmospheric data homogenization and reanalysis methods, see Solargis satellite-based model.

Where Monitor and Forecast satellite data is available

The availability and update frequency of Solargis Monitor and Solargis Forecast solar data depend on the region, as shown in Figure 2:

• Orange regions: Data is updated every few minutes, providing access to satellite-based real-time and nowcasting models for the current day.

• Yellow regions: Data is updated daily, ensuring DAY-1 data (the previous day’s data) is available.

Figure 2: Geographical availability of Solargis Monitor data.

Note: For the timing and satellite characteristics of each region, see Satellite data regions used by Monitor and Forecast below.

Satellite data regions used by Monitor and Forecast

This section gives an overview of the satellite data regions used in the Solargis Monitor and Solargis Forecast services, detailing their historical data availability, satellite characteristics, update timing for DAY-1 data, and real-time or nowcasting capabilities. Use Figure 3 to locate a specific region, then find its row in the table below.

Figure 3: Solargis Monitor and Solargis Forecast satellite coverage regions.

Tip: To request historical data in the areas out of coverage, please use Solargis Evaluate or contact us.

Nowcasting data availability and delay in Monitor and Forecast

The total delay of near real-time Solargis Monitor and Solargis Forecast data combines three factors: satellite data availability delay, data processing delay, and user request delay after processing is complete.

• Satellite data availability delay: Varies by latitude and satellite region:

  • PRIME, IODC: 2–16 minutes (increases from north to south)

  • HIMAWARI: 5–15 minutes (increases from south to north)

  • GOES EAST & WEST (excluding Contiguous US): 2–12 minutes (increases from south to north)

  • Contiguous US: 2–6 minutes (increases from south to north)

• Data processing delay: Also includes retrieval, preprocessing, and running the nowcasting model, takes 5–15 minutes. Data becomes accessible immediately after processing is finished. Data processing frequency depends on the satellite region:

  • HIMAWARI, GOES EAST & WEST: Every 10 minutes

  • PRIME, IODC: Every 15 minutes

  • Contiguous US: Every 5 minutes

• User request delay: This is the time between when data becomes available and when the user retrieves it. The delay depends on how often the user’s system (API or SFTP) checks for new data — more frequent polling results in shorter delays.

Each new satellite image triggers another nowcasting run. The processing frequency determines the window for delivering a given nowcast run before it is replaced by the next. The timing of a user’s request with respect to the processing window introduces the user request delay, ranging from 0–10 minutes, 0–15 minutes, and 0–5 minutes for the respective nowcasting region shown in Figure 5.

Figure 4: Satellite data availability, data processing and user request delays across three consecutive nowcasting runs.

Nowcasting and forecasting update frequency

Figure 5 illustrates the geographical availability of Solargis Forecast data, highlighting the update frequencies for short-term nowcasting and forecasting data across different regions. It shows areas with 5-, 10-, and 15-minute updates using Solargis CMV models, alongside regions covered by NWP models with 6-hourly or 1-hourly updates.

Figure 5: Geographical availability and update frequency of Solargis Forecast data.

Origin of the meteorological data in Monitor and Forecast

The Solargis Monitor and Solargis Forecast services provide key meteorological data parameters, including:

• TEMP (Temperature)

• WS (Wind Speed)

• WG (Wind Gusts)

• WD (Wind Direction)

• PREC (Precipitation)

• RH (Relative Humidity)

• PWAT (Precipitable Water)

• AP (Air Pressure)

These parameters are derived from the post-processed outputs of global Numerical Weather Prediction (NWP) models and the ERA5 reanalysis meteorological model to ensure high accuracy and reliability. For how Solargis sources and post-processes these models, see Meteorological model data.

The table below outlines how meteorological data is integrated into Solargis Monitor and Solargis Forecast:

Tip: For the meteorological parameters included in each subscription tier, see Forecast subscription parameters.