The Certify API allows you to upload time series data for your carbon removal projects. Time series data consists of sensor measurements collected over time, such as temperature, pressure, flow rates, and chemical concentrations. This data is critical for monitoring and verifying carbon removal operations.
Time series data is different from measurement samples. While measurement samples represent discrete laboratory measurements at specific points in time, time series data represents continuous or regularly sampled sensor readings over extended periods.
Time series data can currently be associated with either a Direct Air Capture (DAC) capture facility or a DAC storage location (saline aquifer).

​
Upload Process Overview

Uploading time series data involves several steps:
1

Create sensors

First call the POST Sensor endpoint to create sensor records that define the metadata about the sensor, including what physical parameter it measures, its units, and the facility or storage location it is linked to.
2

Get upload URL

Get a signed upload URL by creating a file upload resource using the POST File Uploads endpoint.
3

Upload file to cloud storage

Use the signed URL to upload your parquet file containing time series data to cloud storage. The content-type header in the PUT request to cloud storage must match the file type supplied to the file upload resource.
4

Submit for processing

Create a data upload submission with the POST Data upload submission. Isometric will process the files asynchronously.
5

Monitor processing

The GET Data upload submissions endpoint allows you to monitor the status of your data upload submissions. Any errors in processing will be reported here.

​
Measurement Properties

A measurement property consists of:
  • Quantity Kind: The physical parameter being measured (e.g., temperature, mass flow rate)
  • Qualifier (optional): Additional specification for the measurement (e.g., CO2 mass fraction)
The allowed measurement properties depend on the type of facility or storage location your sensor is linked to.

​
DAC Capture Facility

​
DAC Storage Location (Saline Aquifer)

​
Data Format Requirements

​
Parquet file structure

We currently support upload of time series as Parquet files. The Parquet file must contain the following columns:
  • aggregation_period_start_timestamp: A timestamp[ns] representing the start timestamp of the aggregation period.
  • aggregation_period_end_timestamp: A timestamp[ns] representing the end timestamp of the aggregation period.
  • sensor_reference: A string containing the reference of the sensor that recorded the measurement.
  • minimum: A double containing the minimum value recorded by the sensor during the measurement period.
  • maximum: A double containing the maximum value recorded by the sensor during the measurement period.
  • mean: A double containing the mean value recorded by the sensor during the measurement period.
  • median: A double containing the median value recorded by the sensor during the measurement period.
  • count: An integer indicating how many measurements were taken during this period.
  • standard_deviation: A double indicating the sample standard deviation (using Bessel’s correction) of the measurements taken during this period.
  • first_timestamp: A timestamp[ns] representing the timestamp of the first actual recorded value in the aggregation period.
  • last_timestamp: A timestamp[ns] representing the timestamp of the last actual recorded value in the aggregation period.
We will ignore any extra columns in the Parquet file that are not listed above.

​
File Size and Performance

  • Maximum file size: 100 MB per upload.
  • Processing time: Processing a file upload will usually complete within a few minutes.