> For the complete documentation index, see [llms.txt](https://doc.re-cast.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://doc.re-cast.ai/readme.md).
# Encored Weather API for Renewable Energy
***
## Table of Contents
1. [Welcome to the Encored Renewable Weather API](#welcome-to-the-encored-renewable-weather-api)
2. [Authentication](#authentication)
3. [Rate Limits and Quotas](#rate-limits-and-quotas)
4. [Data Source Comparison](#data-source-comparison)
5. [Temporal Resolution](#temporal-resolution)
6. [Data Retrieval Methods](#data-retrieval-methods)
7. [Protocols (NWP Forecast)](#protocols)
***
## Welcome to the Encored Renewable Weather API
Encored Renewable API (RE:Cast) is a cloud-based weather data service specialized for renewable energy generation forecasting. It uses global forecast data as initial conditions to generate high-resolution weather predictions tailored to the Korean Peninsula. Through a RESTful API, it provides real-time Forecast Data updated 2–4 times a day, as well as Historical Data covering past forecasts and downscaled analysis fields since 2020.
※ The availability period of historical forecasts may vary by location.
### Why high resolution
Existing weather forecasting services typically operate at low spatial resolutions of 5 km or coarser, making it difficult to deliver accurate forecasts at the exact location of a power plant. Reducing the spatial resolution to below 1 km enables accurate weather predictions for specific power plant locations and improves the accuracy of generation forecasts.
Rapidly changing variables such as wind speed cannot be captured accurately with hourly forecasts alone. Forecasts at 10-minute intervals can particularly improve the precision of wind generation forecasting.
### Weather model
RE:Cast uses Encored's weather forecasting model to produce high-resolution weather predictions for the Korean Peninsula. Compared to global models, it more accurately captures regional characteristics (topography, terrain type, topographic resolution, etc.), improving forecast accuracy. Its strengths span very short-range forecasts (within 6 hours) through short-range forecasts looking several hours to several days ahead.
### Workflow
RE:Cast can be integrated into renewable energy generation forecasting systems in a phased manner through the following workflow:
* Collecting past forecasts and analysis data
* Training generation forecasting models
* Collecting real-time forecast data and running generation inference
***
## Authentication
All API requests require an API key. Contact the [administrator](mailto:esolution@encoredtech.com) to request one.
## Rate Limits and Quotas
Depending on the usage plan, API request rate and total usage volume over a given period may be limited.
***
## Data Source Comparison
| Category | KMA Very Short-Range Forecast (VSRF) | KMA Short-Range Forecast (SRF) | NWP Forecast |
| ------------------------ | ------------------------------------ | ------------------------------ | ------------------------- |
| Forecast range | Up to 6 hours | 72 hours (3 days) | 48–168 hours |
| Update interval | 1 hour | 3 hours | 6 hours |
| Coverage | South Korea | South Korea | South Korea |
| Coordinate system | KMA NX/NY grid | KMA NX/NY grid | NX/NY grid |
| Spatial resolution | 5 km × 5 km | 5 km × 5 km | 1 km × 1 km – 0.1° × 0.1° |
| Data availability period | 2025.09– | 2025.09– | 2020.01– |
| Response format | JSON | JSON | JSON |
***
## Temporal Resolution
| Data source | Issue time format | Forecast interval | Forecast range | Update interval |
| ----------------------- | ----------------------------------------------------- | -------------------- | -------------- | --------------- |
| VSRF (Very Short-Range) | HH30 (0030, 0130, ... 2330) | 1 hour | 6 hours | 1 hour |
| SRF (Short-Range) | HH00 (0200, 0500, 0800, 1100, 1400, 1700, 2000, 2300) | 1 hour | 72 hours | 3 hours |
| NWP Forecast | HH00 (0300, 0900, 1500, 2100) | 10 minutes to 1 hour | 48–168 hours | 6–12 hours |
***
## Data Retrieval Methods
KMA forecast data (SRF / VSRF) and NWP forecast data differ in their data source and characteristics, and use separate query type systems.
| Category | KMA SRF / VSRF | NWP Forecast |
| ------------- | --------------------------------- | ----------------------------------------------------------------------- |
| Data source | Provided after receiving from KMA | Generated by Encored's weather model |
| Model variant | None | Distinguished by the `modelComb` parameter (protocol) |
| Query types | `default` / `recent` / `latest` | `most-recent` / `recent` / `recent-latest` / `recent-dayahead` / `past` |
***
### KMA SRF / VSRF
SRF and VSRF differ only in their endpoint paths and share the same query type options.
**`default`**
* Returns SRF/VSRF data for the target grid at the specified issue date and time.
* Requires the issue date and time (`YYYYMMDD`, `HH`) to be specified.
**`recent`**
* Returns data for the most recent issue time relative to the request time.
**`latest`**
* Stitches together forecasts from multiple issue times and returns only the data for the latest basetime.
* Represents the most up-to-date forecast for each basetime.
### NWP Forecast Data
Forecast data generated by Encored's weather model. The set of weather parameters returned depends on the `modelComb` parameter (protocol). → [See protocols](#protocols)
**`most-recent`**
* Returns the latest available cycle forecast.
* Represents the most up-to-date forecast at the current moment.
* Equivalent to KMA's `recent`. NWP operates on a cycle basis, so the term is distinguished.
**`recent`**
* Returns the cycle forecast for the requested issue date and time.
* Requires the issue date and time (`YYYYMMDDHH`) to be specified.
**`recent-latest`**
* Stitches together multiple cycle forecasts and returns the latest forecast for each basetime (valid time).
* Represents the most up-to-date forecast for each basetime.
**`recent-dayahead`**
* Stitches together cycle forecasts across multiple dates to construct a day-ahead forecast.
* Requires only the hour (`HH`) of the issue date and time to be specified.
**`past`**
* Returns results from past forecast data filtered by the specified basetime range and filter conditions.
### Protocols (NWP Forecast)
Protocols are divided into two groups based on the generation type.
* **PV (Solar)**: Includes the parameters needed for solar generation forecasting.
* **WT (Wind)**: Includes the parameters needed for wind generation forecasting.
| Protocol | Type | Always-present count | Conditional count |
| -------------------- | ---- | -------------------- | ----------------- |
| EchoLimaWhiskeySolar | PV | 902 | 4 |
| EclipseSolar | PV | 278 | 4 |
| EclipseWhiskeySolar | PV | 598 | 4 |
| EchoLimaWhiskeyPlus | WT | 715 | 4 |
| EclipseWinds | WT | 278 | 4 |
| EclipseWhiskeyWinds | WT | 411 | 4 |
※ The `EchoLimaWhiskeySolar` and `EchoLimaWhiskeyPlus` protocols will be available only until March 31, 2026.
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter:
```
GET https://doc.re-cast.ai/readme.md?ask=
```
The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.