Google Earth Engine: The Complete Guide

In this lecture, part of Chapter 2, we will learn how to work with Earth Engine (EE) vector data, including Features and Feature Collections such as points, lines, and polygons. You will get familiar with geometries and perform basic spatial analyses in GEE, including buffering and creating grids.

In this lecture, you will learn how to create geometries in Google Earth Engine (GEE) using the Drawing Tool. We will explore how to input latitude and longitude coordinates, draw points, lines, and polygons, and even store non-spatial data along with your geometries.

In this lecture, you will learn about the FAO GAUL (Global Administrative Unit Layers) dataset, including Level 0, 1, and 2 administrative boundaries. We will explore how to filter the GAUL dataset to load specific state or county boundaries and define your Area of Interest (AOI) for further coding in upcoming lectures. Finally, we will visualize these features on the map using a color palette.

In this lecture, you will learn how to create buffer zones around any feature in Google Earth Engine (GEE) using the .buffer() function. We will also explore how to calculate the area of these features directly within GEE, which is essential for spatial analysis and mapping.

In this lecture, you will learn how to calculate the Euclidean distance from any feature in Google Earth Engine (GEE). This technique is widely used for proximity analysis, helping you measure distances between features and analyze spatial relationships.

In this lecture, you will learn how to create a grid to divide your Area of Interest (AOI) into smaller, manageable parts in Google Earth Engine (GEE). This technique is useful for systematic spatial analysis and sampling within your AOI.

In this lecture, you will learn how to import data from your local disk into Google Earth Engine (GEE) assets. We will cover both vector data (points, lines, polygons) and raster data, enabling you to use your own datasets for analysis and mapping within GEE.

In this lecture, you will explore the Google Earth Engine (GEE) data catalog, view metadata and bands of Sentinel-2 imagery, and learn how to load Sentinel-2 Level 1 images. We will use code snippets from the catalog, paste them into the Code Editor, run the code, and visualize the imagery directly in GEE.

In this lecture, you will work step by step with Sentinel-2 imagery in Google Earth Engine (GEE). We will load an image collection and apply filters based on date, location (AOI), and cloud cover using the metadata, helping you focus on the most relevant images for your analysis.

In this lecture, you will learn how to create a single image from an image collection in Google Earth Engine (GEE) using two methods: Mosaic and Composite.

• Mosaic: Displays the top-most pixels as visible.

• Composite: Calculates statistics across all images (mean, median, mode) to generate a final image for analysis.

We will then load the Sentinel-2 images on the map to visualize and analyze the results.

In this lecture, you will learn how to remove unwanted areas or boundaries from your analysis in Google Earth Engine (GEE) using the JavaScript API. We will explore the .clip() function, which can be applied while displaying images on the map or before exporting data, allowing you to focus only on your Area of Interest (AOI) without unnecessary data.

In this lecture, you will learn how to export satellite images or data from Google Earth Engine (GEE) to Google Drive. This allows you to perform further analysis, create maps, or use the data in other GIS and remote sensing applications.

In this lecture, you will learn how to calculate various remote sensing indices in Google Earth Engine (GEE) to analyze different features on Earth's surface. We will use the normalizedDifference() function to calculate NDVI and extract high-vegetation areas. For more complex indices, we will explore the ee.Image.expression() method.

Some of the spectral indices covered include:

• AVI (Advanced Vegetation Index)

• NDVI (Normalized Difference Vegetation Index)

• MNDWI (Modified Normalized Difference Water Index) – Bands B3 & B11

• NDBI (Normalized Difference Built-up Index) – Bands B11 & B8

• NDSI (Normalized Difference Snow Index) – Bands B3 & B11

• BSI (Bare Soil Index)

• EVI (Enhanced Vegetation Index)

• NBRI (Normalized Burned Ratio Index)

In this lecture, you will learn how to use the .map() function in Google Earth Engine (GEE) for parallel processing. This allows you to process hundreds of images simultaneously, significantly speeding up your workflow by leveraging GEE’s cloud computing power.

In this lecture, you will learn about Reducers in Google Earth Engine (GEE) and how they are used to simplify and summarize large datasets. Reducers combine multiple values into a single meaningful number, such as sum, mean, minimum, maximum, median, or standard deviation, making it easier to analyze and interpret geospatial data.

In this lecture, you will learn how to visualize geospatial data in Google Earth Engine (GEE) using different types of charts, including ScatterChart, LineChart, BarChart, and PieChart. GEE integrates with Google Charts, allowing for customization using .setOptions(). You will also learn how to export chart data for further analysis or reporting.

In Chapter 5, you will learn about machine learning for image classification in Google Earth Engine (GEE), focusing on supervised classification and its workflow. We will also discuss the differences between supervised and unsupervised classification, helping you choose the right approach for your remote sensing analysis.

In this lecture, you will learn how to collect training Ground Control Points (GCPs) for each class to prepare an accurate LULC map in Google Earth Engine (GEE). We will explore which feature type—point, line, or polygon—is best for collecting GCPs using different satellite band composites. Finally, we will merge the datasets according to class to create a complete training dataset for classification.

In this lecture, you will learn how to create a Land Use/Land Cover (LULC) map in Google Earth Engine (GEE) using your collected Ground Control Points (GCPs). We will explore three supervised classification methods in GEE: Random Forest, CART, and SVM and focus on the Random Forest classifier, the most widely used method, to generate an accurate LULC map.

In this lecture, you will learn how to assess the accuracy of your LULC map in Google Earth Engine (GEE) to ensure it is reliable for further use. We will split the GCP dataset into 70% for training and 30% for validation, and calculate key accuracy metrics including:

• Confusion Matrix

• Overall Accuracy

• Kappa Coefficient

• Producer's Accuracy

• Consumer's Accuracy

In this lecture, we will focus on single-band change detection using two time-period satellite images. We will calculate relevant indices, subtract the before and after values, and apply a threshold to extract areas of change caused by an incident.

In this lecture, you will learn how to perform change detection between LULC classes using two different time-period LULC maps in Google Earth Engine (GEE). We will also calculate the area of change for each class, helping you quantify land cover transformations over time.

In this lecture, you will learn how to create NDVI animations over different time periods in Google Earth Engine (GEE) to visualize changes in vegetation. You can also export the animation as a GIF and include dates or timestamps for better visualization of changes over time.

In this lecture of chapter 7, you will explore how to create a user interface (UI) in Google Earth Engine (GEE) to build interactive web apps. You will practice publishing your app for free and discover existing Earth Engine apps for inspiration.

In this lecture, you will practice adding interactive elements such as titles, dropdowns, and buttons to your Google Earth Engine (GEE) app. These UI widgets help make your app dynamic, user-friendly and interactive.

In this lecture, you will practice building a complete interactive Earth Engine app by adding panels and layouts to your map. Inside panels, you can include UI elements for interactivity, such as images, links, charts, and other widgets, making your app fully dynamic and user-friendly.

In this lecture, you will practice publishing your Earth Engine app step by step without any cost. You will learn how to host your app in Google Cloud or your project and share your analysis with the public via unique URLs, making your results easily accessible.

In this lecture, you will practice creating two panels with a divider in your Earth Engine app. Panels can be dragged and resized, and by default, they are divided equally. You will also learn how to link the left and right panels so that zooming or moving one panel updates the other, enhancing interactivity.