Forest fires have the grim ability to burn millions of acres of land rapidly, consuming everything in its path. The frequency of forest fires has been rapidly increasing due to human activity and climate change. Forest fires cause huge economical and ecological damage, where carbon emissions can severely affect human health, wildlife, and climate on an enormous scale. Forest fires are common in many parts of the world, with no particular location or country, or continent to pinpoint its recurrence to. Many countries are modeling these smoke patterns and are trying to correlate these models with their community health.  With forest fires, the emission of smoke causes horrible air quality. There is a need for reliable tools to predict population smoke exposure. With EOfactory.ai, we have made it easy to capture the extent of smoke automatically and accurately. 

EOfactory is a one-stop platform for AI/ML model building and creation that allows you to understand earth observation data like never before. The first global EO platform, connecting EO content creators to consumers using pre-trained and custom models in the SkyMap Global EOFactory. Take a look at a case study we conducted. 

A Case Study – Sonoma County, California, USA November 2019 – Done on EOfactory.ai 

Sonoma County is in Northern California. It’s known for the Sonoma Valley wine region, as well as other notable winemaking areas such as the Dry Creek and Alexander valleys. The 2019 forest fire of Sonoma County was named “Kincade Fire”. The fire started northeast of Geyserville in the Geysers at 9:24 p.m. on October 23, 2019, and subsequently burned 77,758 acres (31,468 ha) until the fire was fully contained on November 6, 2019. The fire threatened over 90,000 structures and caused widespread evacuations throughout Sonoma County

With EOFactory.ai, one can download satellite images and investigate. Accordingly, Sentinel-2 satellite images were downloaded for the period of forest fire and analyzed.

EOFactory.ai allows users to select training datasets from the images. The above area (blue box) was selected as a training area. Smoke was classified as smoke and other features such as vegetation, water, and city were classified as others. Accordingly, the entire 1,800 sq km of AOI was analyzed

As seen above, the smoke was extracted accurately from EOFactory (In Yellow) and non-smoke features as others (green)

It can be seen that every smoke area has been accurately traced and vectorized. The shadow areas are left from the smoke area. One of the first in the world, EOFactory allows users to create AI models by visual interpretation and get the required output accurately but in no time. The above analysis covering nearly 2,000 sq km took less than 15 min of manual effort. Come try our EOfactory platform for free! https://eofactory.ai/