|Availability of PV, solar and meteorological data||Technical features|
|historical||operational||real-time & nowcast||NWP forecast||long-term average||protocol||type of communication||content type|
|Data Delivery Web Service|
|pvPlanner Web Service||NO||NO||NO||NO||YES||HTTP||synchronous||XML|
|FTP data delivery||YES||YES||YES||YES||NO||FTP||asynchronous||CSV|
Table: Description of data available through the Solargis API.
Solargis API consists of two different endpoints:
- Data Delivery Web Service - the main service for accessing Solargis time series data. Both request and response are XML documentsis an XML document. The request parameters (XML elements and attributes) are formally described by XML Schema Definition documents (XSD). By using the the schema, request or response can be verified programmatically. For this service, we provide two architectural styles, the REST-like endpoint, and SOAP endpoint. Look for more technical information here. Authentication and billing is based on API key registered with the user. Please contact us to discuss details, set up trial or ask for a quotation.
- pvPlanner Web Service - this simple web service provides monthly long-term averaged data (including yearly value) of PV, solar and meteorological data with global coverage.The The service is targeted for prospection and prefeasibility. Sending pre-feasibility. Sending an XML request mimics the click on the Calculate button in the interactive Solargis pvPlanner application. Request and response for the service is not described in this user guide. More information can be found here.
Additionally, we provide an FTP data delivery service service where the request (CSV file) is stored in the user's FTP directory. The service is then scheduled to deliver CSV response files regularly. Request processing is asynchronous - client puts the client creates the CSV request, server the server processes the request according to schedule (e.g., 4x per day, every hour), the client then checks for the response files. The CSV request allows for multiple locations in one file. For pricing and setting up a trial FTP user account, please contact us.
Description of data available through the Solargis API
In In the case of solar and PV time series, we use satellite data since available history up to the present moment plus additional 4-5 hours ahead (in the regions where the real-time & nowcasting satellite data is available). Satellite data includes historical (archived) data, operational data, real-time and nowcasting data. Historical data ranges up to the last completed calendar month and can be considered as "definitive". Data in the current calendar month up to DAY-1 is so-called "operational", and will be re-analysed analyzed in the next month using final versions of required data inputs (e.g. atmospheric data parameters). Important to note is that differences introduced with the re-analysis is are typically small. Solar data in the current day is from the "real-time" satellite model and will be updated when the day is finished. Then, based on the latest satellite images we predict cloud motion vectors (CMV) in the range of next 4-5 hours ("nowcasting"). The present moment and short period before is covered by the nowcasting model data as the recent satellite scene is still in progress. This delay can take up to 30 minutes (depends on the satellite scanning frequency). Later on, after the nowcasting time range, we use post-processed outputs from Numerical Weather Prediction models (NWP). Satellite-based data is seamlessly integrated with NWP forecasting data within one response. In the case of locations where real-time & nowcasting data is not available, NWP data is used for the course of the current day. Also, not every location on the globe is supplied by more accurate ECMWF IFS data. In such case, NOAA GFS data is used for all forecasted values. Meteorological data (TEMP, WS, AP, RH...) is comprised of NWP (NOAA GFS) modeled data.
Schema below shows how the data sources are integrated on an example of the the Data Delivery Web Service response having 9 days of data (generated at 12:00 of a given day).
Satellite based solar and PV data - from history up to real-time
Orange regions on the map are accessible via the API and data is updated everyday every day (DAY-1 is available). In the subset of these regions, the real-time/nowcasting data is available (within the current day DAY+0, updated every 30 minuteminutes). Main data parameters include GHI, DNI, DIF, GTI, PVOUT.
The following table will help users to schedule a time for sending requests to Data Delivery Web Service:
Each daily update of the data re-calculates values for two days backwards backward (DAY-1 and DAY-2). Monthly update (on the 3rd day of each calendar month) re-calculates the whole previous month as soon as it's completed. The purpose of these updates is described in this article. We gradually expand spatial coverage of satellite data accessible via API. To request operational and historical data in the grey areas on the map, please use Solargis climData online shop.
The data from orange zones in the map is also available by using interactive application pvSpot (daily operational data) and the data is accessible within minutes after purchase via climData online shop (as historical multi-year archived datatime series).
Meteorological data from numerical weather models - from history up to current day
Main data parameters include air temperature (TEMP), wind speed (WS), wind direction (WD), relative humidity (RH) and many others. Historical meteorological data comes from post-processed numerical weather models and it is available globally. The DAY-1 and DAY-2 values are taken from NWP models - NOAA GFS (resp. ECMWF IFS) data sources (they are forecasted values). The preliminary meteorological data from the GFS model is later updated with data from the NOAA CFS v2 data source (re-analysed analyzed archive data). Meteorological data for period DAY-3 and older can be considered as definitive.
Solargis forecast is based on the post-processing of outputs from NWP models. The forecast time series include the following data parameters:
- violet regions: high resolution, higher reliability forecast data is available in the violet regions marked on the map. Upon request, we can start this kind of forecasting service for any other area. Source: IFS model from ECMWF, UK. Frequency The frequency of the update is at UTC hours 00, 06, 12 and 18 (4 forecasts runs per day, every 6 hours). Forecast The forecast range is from DAY+0 up to DAY+3. Original temporal resolution for the first 48 hours is 1 hour, hours 48-84 are received in 3 hourly original resolution, however, in the final response, this can be interpolated into higher resolution.
- the rest of the map (in white color) is covered by lower resolution global forecasting data from GFS from the GFS model (NOAA, USA). Forecast The forecast range is from DAY+0 up to DAY+10. Frequency The frequency of the update is once in 6 hours.
The client (most often a computer) will send the XML request and waits for the XML response. Users can test web service services directly from the web browser by using e.g. REST Client for Firefox or via a native application like Postman. Before sending requests user must set the HTTP Method to "POST", define endpoint URL to: to https://solargis.info/ws/rest/datadelivery/request?key=demo and also set a header to "Content-Type: application/xml". Then use the XML request examples below and send them in the body of the HTTP request and explore XML responses. Typically, developers will create client code to send requests and handle responses scheduled in time. For all technicalities visit this link. In the next section, there is a description of XML request requests with examples. They can serve as starter templates for most typical scenarios.