Visual Crossing

Visual Crossing provides historical weather observations and weather forecasts for any point on Earth. With Weld you can sync daily and hourly weather history and a 15-day forecast for the locations you choose into your data warehouse, and join it with sales, footfall or delivery data to see how the weather affects your business.

Features

Feature nameSupported
Column HashingTrueColumn level
BlockingTrueColumn level
IncrementalTrue
Custom dataFalse
HistoryTrue
ReSyncTrueTable level
TemplatesFalse

Data Schema

Setup Guide - ELT

Prerequisites

You need a Visual Crossing account and its API key. Weld uses the key on your own Visual Crossing plan, so the records Weld fetches count towards your plan's usage. You also need the latitude and longitude of each location you want weather for.

Step 1 - Copy your Visual Crossing API key

  1. Sign in to Visual Crossing.
  2. Open your account page and copy your API key.

Step 2 - Create the connection

  1. In the Weld app, click Data Sources in the sidebar, then Add Data Source.
  2. Search for Visual Crossing, select it and choose Create new connection.
  3. Give the connection a Name.
  4. Paste your API key, then click Connect.

Step 3 - Add your locations

In the sync settings, add each place you want weather for under Locations:

  • Latitude and Longitude of the location, in decimal degrees (for example 55.6761 and 12.5683 for Copenhagen).
  • ID (optional) identifies the location in your warehouse, in the location_id column of every table. Use something you can join on, such as a store number. If you leave it empty, Weld generates an ID when you save, and it stays the same even if you edit the coordinates later.

Don't change a location's ID after the first sync. Data already synced keeps the old ID, so the location would end up split across two IDs. Each location needs its own ID.

You can add, edit or remove locations later. A location added later is backfilled on its own from the sync's start date.

Set the Daily record limit to the most Visual Crossing records this connection may use per day (UTC). It defaults to 1,000, the limit of Visual Crossing's free plan. Each day of history costs 24 records per location, and each forecast costs 1 record per location. Set it to 0 if your Visual Crossing plan has no daily cap; otherwise the lowest allowed value is 24.

When the daily record limit is reached, history pauses for the affected locations and continues after midnight UTC, and the forecast is skipped for the locations it can't afford that day. A long backfill on a small limit can therefore take several days to complete.

Step 4 - Data To Sync

Select the tables you wish to include in the sync. You can view the schema, remove columns or hash sensitive information.

  • History — history holds daily weather, one row per location per day, from the sync's start date up to yesterday. The tables history_hour (hourly weather), history_preciptype and history_hour_preciptype (the day's and the hour's precipitation types) come along with it.
  • Forecast — forecast holds the 15-day forecast for each location. Every forecast is stored with the date it was issued (forecast_issued_date), so earlier forecasts are kept next to later ones and you can compare them with what actually happened. The tables forecast_hour, forecast_preciptype and forecast_hour_preciptype come along with it.

All measurements are in metric units (°C, mm, km/h, km). Hourly rows carry both the local hour and a UTC timestamp (datetime_epoch).

Step 5 - Configure sync

  1. Select how often you would like the data to sync.
  2. Choose a start date for the history backfill. If you don't set one, history starts one year back. Every day of history costs 24 records per location, so pick a start date that fits your daily record limit.
  3. Check the Destination schema name, which is where the tables land in your warehouse, and click Start Initial Sync.

Was this page helpful?