Using SolarAnywhere® Data

  • Q. What is included with a SolarAnywhere historical download?

    All SolarAnywhere historical time-series downloads include a file prefixed “SolarAnywhere Time-series” that contains hourly, half-hourly or 1-minute averages over the time range selected, along with a license file.

    Typical GHI and DNI Year downloads also come with two files prefixed “SolarAnywhere Typical GHI (or DNI) Year.” Typical GHI or DNI year files contain hourly data for a combination of the twelve months whose summed GHI or DNI insolation is closest to the average for that month over the period 1998 through the last complete year.

    The .MEF file specifies the protocol for PVsyst version 5.31 compatibility and is included in each download, but only needs to be used once for each computer you use per the directions below.

  • Q. What is the difference between Standard and Enhanced resolution data?

    Standard SolarAnywhere spatial resolution is approximately one-tenth of a degree, which equates to an area that is approximately 10 by 10 km, and is available with 60-minute time resolution. Enhanced resolution data is one-hundredth of a degree, which equates to about a 1 by 1 km area, and is available with 30 or 60-minute time resolution. Currently, enhanced resolution data is available on a limited basis.

    SolarAnywhere Data Type Spatial Resolution Temporal Resolution Availability
    Standard 10 x 10 km 60-minute Historical and forecast data
    Enhanced 1 x 1 km 30 or 60-minute Historical data
  • Q. How do I use SolarAnywhere data with PVsyst 5.31?

    SolarAnywhere data can easily be imported into PVsyst 5.31 by following a few simple steps:

    1. Download data from SolarAnywhere in TMY3 format. We find it easiest to select the 60-minute time resolution even when using Enhanced Resolution data when working with PVsyst.
    2. In PVsyst, select Tools and then Import Meteo Data. Select a U.S. TMY2/3 data type and import the GHI SolarAnywhere ‘typical GHI/DNI year’ file. This will establish a ‘site’ with the SiteName, Elevation and Timezone, as well as monthly averages for GHI, wind and temperature.

    If you’re only interested in a typical year analysis, you can now do your analysis. If you want to study a particular year, you need to do the following:

    1. Install the SolarAnywhere TMY3 .MEF file in PVsyst’s “Data\Meteo” directory. The .MEF file specifies the protocol for PVsyst and only needs to be done once for each computer you use. SolarAnywhere provides you with the .MEF file every time you get new data for your convenience.

    For a default install of PVsyst 5.31, use “Data\Meteo” paths from the table below.

    Operating System “Data\Meteo” Path
    Windows XP C:\Documents and Settings\All Users\Application Data\PVsyst\Data\Meteo
    Windows 7 C:\ProgramData\PVsyst\Data\Meteo
    C:\Program Files\PVsyst5\Data\Meteo
    C:\Program Files\PVsyst6 (x86)\DataRO\PVsyst6_Data\Meteo
    1. Go to Tools and select Import Ascii Meteo File, then select the SolarAnywhere Time-series TMY3 file.
    2. Select the SolarAnywhere TMY3 protocol file (it should show up automatically) in the drop-down list, and click Start Conversion.

    The time-series data will be imported and can be used for analysis. Note that PVsyst can only work with a single year of data at a time.

  • Q. How do I use SolarAnywhere data with the NREL System Advisor Model (SAM)?

    To import data to SAM you will need to:

    1. Download data from SolarAnywhere in TMY3 format, selecting the 60-minute time resolution even when using Enhanced Resolution data.
    2. Create a project in SAM. n the Climate tab, click Add/Remove. Add the folder containing the SolarAnywhere data you have downloaded. The files should now appear in the location drop-down.
    3. Select either the typical year or time-series file, and you will be able to use the data with your SAM project.
    4. In certain versions of SAM, a SolarAnywhere file conversion script is required to import data. This is available directly from NREL.
  • Q. Are there options to fill periods of missing data in SolarAnywhere?

    Yes, SolarAnywhere offers ways to replace periods of missing data. The following options will be available to all SolarAnywhere Data users, under their existing license.

    • Average Values – This option replaces data gaps with the climatological average for the specific day of the year. For instance, if April 22nd, 2012 presents a data gap, the software will replace that day with the average measurements from all other intact April 22nd days in the database (i.e., 1998, 1999, 2000…). This option is the preselected, default value and is available in TMY3 and SolarAnywhere file formats.
    • Blanks – This option returns data gaps with blank cells. This option is available in TMY3 and SolarAnywhere file formats.
    • -999 – This option replaces all data gaps with the integer -999. This option is often preferred when looking to sift for data gap measurement periods. This option is only available in SolarAnywhere file format.
    • NaN – This option replaces all data gaps with the text string NaN. Also preferred for sifting periods with data gaps, this option gives additional compatibility with database software used to further manipulate SolarAnywhere Data (i.e., Microsoft Excel). This option is only available in SolarAnywhere file format.
  • Q. What do the column headers in the first row of SolarAnywhere time-series output files mean?
    The first row in SolarAnywhere time-series data files contains standard TMY3 format header information, plus a summary comment. From left to right, the columns correspond to:

    1. Site identifier code – Because SolarAnywhere data covers areas where no TMY3 site exists, we report all site identifier codes as ‘0.’
    2. Station name – The Site Name specified on the ‘Select Delivery Method’ page (see image below) plus ‘SA’ plus the type of data in the current file (Time-series, Monthly Totals, Annual Totals, Typical GHI/DNI Year, or Clear Sky). For example: Site Name_SA_MonthlyTotals.solaranywhere-sitename-field
    3. Station state – Always listed as “NA” or Not Applicable, since our data locations may include areas of multiple states.
    4. Site time-zone – Listed in hours offset from UTC; time zones west of Greenwich are negative (e.g., PST is 8 hours behind UTC, therefore PST is listed as -8).
    5. Site latitude – Decimal degrees.
    6. Site longitude – Decimal degrees.
    7. Site elevation – Meters.
    8. Summary comment:
      1. Latitude/Longitude resolution – Width and height of the selected tile in decimal degrees.
      2. Time resolution – Period between data points.
      3. Averaging method – Middle-of-period or end-of-period.
      4. Copyright information.

    More information about the TMY3 file format and data headers can be found in Section 1.4 of the User’s Manual for TMY3 Datasets at NREL.

  • Q. What is the difference in the averaging method used for the different output formats?

    There are two period-averaging conventions available through SolarAnywhere: ‘centered on satellite measurement time,’ and ‘standard top-of-hour, end-of-period integration.’ Both conventions reflect the average irradiance that occurred for the applicable time period.

    • Centered on Satellite Measurement Time (i.e., 11:15 represents 10:45 to 11:45) – Satellite images are retrieved by SolarAnywhere on a 30-minute basis. Depending on the exact satellite network, the time of image capture may vary. For instance, the GOES-E device takes images at 0:15 and 0:45 past each hour. Hourly data with time stamps of 0:15 past each hour will represent the average irradiance from 0:45 of the proceeding hour to 0:45 past the current hour (i.e., 11:15 represents 10:45 to 11:45).
    • Standard Top-of-Hour, End-of-Period Integration (i.e., 11:00 represents 10:00 to 11:00) – Cloud cover at times when satellite images are collected are projected to top-of-hour, end-of-period using techniques to normalize irradiation against the time period. Hourly data with time stamps of 0:00 will represent the normalized irradiation from the proceeding hour (i.e., 11:00 represents 10:00 to 11:00).

How SolarAnywhere Data compares with other data sources

  • Q. Is the Perez model used to generate SolarAnywhere Data new or different from the one used by the NREL NSRDB?
    SolarAnywhere Data is generated using a more recent version of the Perez model than that used to generate the NSRDB 2005 and 2010 updated datasets. SolarAnywhere also provides more recent data through the last hour, forecasts and other features that are not available from the NSRDB.
  • Q. How does SolarAnywhere Data compare with NREL TMY data?
    Typical Meteorological Year irradiance data available from NREL is synthesized from historical sources to represent a “typical” year for a fixed number of sites in the U.S. SolarAnywhere represents actual hourly estimates of irradiance for each specific location based on satellite imagery and atmospheric conditions at the site.
  • Q. How do satellite-derived irradiation sources compare with output from ground-based measurement instruments?
    If properly calibrated and maintained, ground-based instruments can be very accurate for the immediate area around the equipment. As the area of interest is located further from the unit, the accuracy declines, increasing the usefulness of satellite-based observations. The cost of equipment, inaccuracies from calibration and long setup times can favor satellite over ground-based measurements. Satellite irradiance is also often used in conjunction with ground-based instruments, and is particularly useful for detecting ground device calibration drift and for filling in missing data.

SolarAnywhere Methodology

  • Q. What data sources are used when generating SolarAnywhere Data?
    Cloud, albedo, elevation, temperature and wind speed data is used in conjunction with satellite imagery collected from geosynchronous satellite networks.
  • Q. How often is SolarAnywhere Data updated?
    SolarAnywhere Data is collected, processed and available for use within approximately one hour. When the forecast option is licensed, a continuous dataset extending from January 1, 1998 through the present, and up to 168 hours into the future is available.
  • Q. Why are different models used when forecasting days-ahead and hours-ahead data periods?
    Short-term SolarAnywhere forecasts utilize a vector based cloud model. Longer-term forecasts rely on numerical analysis. The bifurcated approach utilizes the method expected to give the best results for the time interval requested.
  • Q. What happens when the SolarAnywhere model changes?

    SolarAnywhere algorithms are occasionally updated to improve accuracy. See the release notes for more information. After a model update, historical data is often reprocessed to take advantage of the new model’s improved accuracy.

    To provide continuity for users of older SolarAnywhere datasets, prior versions may be provided upon user request. Presently, only support back to SolarAnywhere version 2.2 is available via data.solaranywhere.com. Contact us for information on prior datasets.

  • Q. There are periods of missing measurements in my SolarAnywhere file. Why?
    Missing data occur in the SolarAnywhere irradiance database due to missing satellite images. Missing images are normal and occur due to rare unplanned outages and regular maintenance performed by the National Oceanic and Atmospheric Administration (NOAA). Missing data also occur in the ancillary surface air temperature and wind speed data due to periods of missing measurement from ground-based sensor networks.
  • Q. How does SolarAnywhere handle missing data?

    SolarAnywhere strives to generate uninterrupted solar irradiance on an hourly basis; however, occasionally it is not possible, usually due to satellite image interruptions.

    Neighboring values may be used to generate estimates in place of the missing values. When this occurs, the observationType column will contain a suffix of “E” indicating that a value was generated from the surrounding observations.

    If too many samples are missing, the ObservationType column will contain a suffix of “M,” indicating that no data is available for that hour. Note that a prefix value of “M” in the ObservationType column is different from a suffix value of “M.” A prefix value of “M” indicates that the values in the row were generated for the current month.

Accessing Licensed Data

  • Q. How are geographic areas defined when licensing data?
    The basic geographic unit is a satellite visible “tile.” Each tile represents one degree of coverage from the satellite, which equates to approximately a 10 x 10 km area. In some areas including all of California, we can provide granularity up to 1km x 1km. Custom areas can also be licensed. For more information, see pricing or contact us.
  • Q. I don’t see a grid when I zoom into the SolarAnywhere map. What’s wrong?
    The grid of satellite tiles becomes visible after zooming in on a location that is accessible as defined in a user’s license agreement. Public users have access to all areas, with limited capabilities. The guest user has access to an area centered on the Ozarks National Forest. Users who purchase licenses have grid access to the specific areas defined in the license agreement.

SolarAnywhere Versions

  • Q. What is the difference between v2.3 and v2.4 SolarAnywhere Data?
    SolarAnywhere v2.4 incorporates an updated algorithm that better integrates irradiance data to improve the hourly accuracy of the standard top-of-hour, end-of-period integration format. Irradiance measurements centered on satellite measurement time will not be affected by changes in v2.4. For more information, see the release notes.
  • Q. What is SolarAnywhere v3.0 and what are IR images?

    SolarAnywhere v3.0 differs from previous models in that it incorporates infrared (IR) image processing into the measurement of cloud location and resulting irradiance. Under previous models, only visible wavelength light images were used to determine cloud location; however, regions with heavy snow cover often confounded the model by detecting snow not cloud cover. The use of IR image channels allows SolarAnywhere to differentiate between snow and cloud, and improve model accuracy when snow cover exists.

    IR images are similar to visible images in that they are captured by the same geostationary satellite networks. They are different, though, because they capture images at infrared (3.8 – 13.3 microns) wavelengths. For more information, see the release notes.