The objective of this lab is to use spectral indices to better understand the irrigated landscape. By the end of this lab, you will be able to use spectral indices to explore the vegetated landscape and use these indices and ancillary data to recognize when an area is irrigated.

In this module, I will discuss the following concepts:

How to rename bands of an image in GEE.
How to work with premade indices.
How to generate your own indices using band math.

Acknowledgement: This lab assignment is adapted from a lab assignment by Dan Carver and Peder Engelstad, Ecology and Natural Resource Ecology Laboratory at CSU.

Using Vegetation Indices: Calling ancillary climate and cropland data and combining these datasets with Landsat spectral indices.

Background

The majority of the most productive agricultural land in the world has been extensively developed for cultivation. Efforts to increase overall agricultural production often rely on increasing the viability of marginal lands by importing water through water diversion projects or pulling ground water from subsurface reservoirs called aquifers. The Ogallala aquifer, which is found below 10 western states in the United States, provides water for approximately 30% of all irrigated agriculture in the country. You can read more about the Ogallala aquifer and see the origins of the figure below in this link.

The extensive use of the aquifer combined with a slow recharge rate means that many regions of the aquifer have seen significant decreases in water levels. The extraction of water from these aquifers can be so dramatic that NASA GRACE satellites have been able to detect changes in the gravitational pull of the earth over these aquifers over time. The decrease in the Ogallala is minor compared to other major aquifers around the world. Yet, there is a large amount of data available that shows agricultural land production over the Ogallala. In this module, we are going to see what remote sensing data are well suited for identifying irrigated crop lands within the Ogallala. Our goal is to apply these same methods to other regions of the world that have less publicly available agricultural data but rely on ground water extraction to promote marginal land agriculture.

Trends in the changes of ground water storage in the Earth’s 37 largest aquifers. Read more about this image here. Image credit: UC Irvine/NASA/JPL-Caltech.

Identifying Irrigated Lands with Remotely Sensed Imagery

You will need to open a new Google Earth Engine script for this module. Do so at code.earthengine.google.com/

Defining an Area of Interest

Our first step is going to be defining the area of interest.
Use the image of the Ogallala aquifer below to draw a bounding box within your GEE script using the draw a rectangle tool in the upper left-hand corner of the map.
The bounding box does not need to be perfect. We just want to narrow our focus to the general area, which could be pulling from this water source. That is all the land that is above the aquifer.

Note that this figure comes from the National Climate Assessment from 2018. You can find a full description here.