We're Making Climate Data Accessible To The Rest of Us
Nobody should need a PhD in Geophysics to understand how climate change will affect their interests
We're using big data expertise developed running adtech data engineering at Adobe to make climate data more accessible for non-scientists and scientists alike, via a Climate Data Workbench. Instead of having to download terabytes of data from disparate sources, then figure out how to reformat out of NetCDF into a SQL friendly format, then figure out how to store it alongside their own proprietary data, and then finally cross-reference the two, analysts in policy, the media, financial services or insurance will instead have a workbench where they can bring their own data and the climate data will already be there to cross-reference with, accessible to use out of the box.
Imagine having a workspace running in the cloud where you bring your own data and can run queries against it. Now imagine all the climate data you might want to join to is already parked there, also, ready for your analysis. Things like historical temperatures, RCP pathways, SSP pathways, and sea level rise projections, all at the tip of a SQL query like this:
e.sku,
e.failure_rate as current_failure_rate,
e.failure_rate + (tp.mean_temperature_increase * e.failure_rate_per_degree_temp_increase) as future_failure_rate
from
my_equipment_inventory e
join
global.projected_global_temperatures tp on tp.zipcode = e.zipcode
where
tp.years_from_now = 30
You probably haven't found a tool quite like ours. There are some other near-analogs, but most of them are built for hardcore climate data scientists, and not so much for everybody else:
- Pangeo, which is an open source framework for hardcore climate scientists that you have to host yourself
- Descartes Labs: similar, but more hardcore sciencey and focused on satellites and visual imagery
- 2i2c: an avidly open source platform for hosting Pangeo-related tools founded by some of the maintainers of Pangeo-related tools
- Google Earth Engine, which already hosts many of the climate data sets for querying in their own proprietary format.
- Qubole: a general (i.e. non-climate) data workbench on top of open source tools like Hadoop, Hive and Spark
We're still in the extremely early stages of building, which makes this the perfect time to get involved.
Interested? We'd love your feedback: info@pollen.io.
2023 © Pollen Analytics LLC. ALL Rights Reserved.