A recent project aimed to enhance public understanding of ecosystems by developing a tool that classifies climate categories from user-taken landscape photos using machine learning. The project leveraged a Convolutional Neural Network (CNN), specifically a ResNet-18 architecture, trained on a dataset of 320,000 geotagged Flickr images, filtered for natural landscapes and tagged with a modified version of the Köppen climate classification system. The CNN outperformed logistic regression and support vector machine models, achieving more accurate climate predictions by focusing on relevant image features. Despite challenges such as dataset noise and computational limitations, the approach showed promise, with potential improvements suggested for future iterations, including incorporating seasonal data to reduce misclassifications.