Mapping carbon pollution globally with satellites

Опубликовано 10 марта 2022, 18:00

Blog post → goo.gle/3hbKLSA
Notebook used to host the codelab → goo.gle/3LAVJyW
Climate TRACE → goo.gle/3uSKL1T

Welcome back to People and Planet AI, where we explore how machine learning helps us understand the biological variety and variability on Earth. In this episode, open_eco_source@ discusses how to generate a geo-spatial classification model that uses images of power plants, collected from a satellite called Sentinel-2, in order to determine whether the power plant is on or off. Watch to discover the power of machine learning and how we can harness this technology to prevent the planet from reaching that crucial 1.5 degrees Celsius of warming.

Chapters:
0:00 - 1:07 What is Climate TRACE
1:08 - 1:19 Model type for this episode
1:19 - 1:42 Series intro
1:42 - 1:56 Meet Gavin from Wattime
1:56 - 2:36 Challenges with carbon pollution measuring today
2:36 - 3:13 Data and computing for Climate TRACE
3:13 - 4:05 Google’s involvement in Climate TRACE
4:05 - 5:21 Review inputs & outputs for building a geo-classification model of coal plants
5:21 - 6:22 Datasets for building the model
6:22 - 7:08 Training & validating the model
7:08 - 7:30 Host model on Cloud Run to run as a prediction service
7:30 - 7:41 View of notebook with code
7:41 - 7:57 Closing resources
7:57 - 8:34 Gavin’s fun fact

Climate:
• Report on 1.5C climate change by the IPCC → goo.gle/3LBFtOj
• Sentinel 2 satellite information → goo.gle/3GRpk3I

Google resources:
• How to start a Google cloud project $300 free trial link → goo.gle/3HTWNvK
• Earth Engine → goo.gle/3gMOtBV
• Colab (interactive notebooks) → goo.gle/3uSJc3W
• Free course by Google on Data Preparation and Feature Engineering in ML → goo.gle/3HWiXxw
• Data normalization → goo.gle/3oQiWUd
• Folium (open source Python library for interactive maps) → goo.gle/3HWYZTa
• Vertex AI → goo.gle/33mlTEn
• Autopackage feature in Vertex AI → goo.gle/3LxEP4x
• TensorflowKeras → goo.gle/3GPDeTV
• Cloud Run (web app hosting) → goo.gle/3HRCA9K
• Hosting a website on Google Cloud using Cloud Run → goo.gle/3sKgNus

People and Planet AI playlist→ goo.gle/PeopleAndPlanetAI
Subscribe to Google Cloud Tech → goo.gle/GoogleCloudTech

product: Cloud - General; fullname: Alexandrina Garcia-Verdin;