The aim of this thesis was to compare quantum and classical machine learning methods to classify and predict crop parcels from time-series data obtained from SENTINEL-2 satellites. The dataset used is the Eurocrops dataset which consists of labeled crop data from various EU countries along with their reflectance data across 13 bands. The first step was looking at existing classical machine learning models and their performance on the Eurocrops dataset. For the next part, we look at how various quantum machine learning methods perform on the same dataset. We find a way to encode the data for a quantum computer using a quantum kernel. The work of this thesis was purely theoretical and invaluable insights were gained as to how in the future time-series data can be processed and predicted by a quantum computer.