These past three weeks were a whirlwind of meetings, trainings, and making progress on the Google Earth Engine script (my main deliverable). I love the diversity of the work that I am performing. I enjoy meeting new people and learning what they do, attending trainings, and learning about the amazing work other people are doing through webinars. While Google Earth Engine is frustrating (when it’s not working) there is a sense of excitement when everything goes right.
Never stop learning…
I am reaching the end of my third week with the DNR. The first week contained setting up and exploring Google Earth Engine scripts. This was necessary to get a handle on how to tackle my project. Most of my research and learning how to use Google Earth Engine was through https://developers.google.com/earth-engine/, research papers, and with the help of my supportive academic adviser, Mutlu Ozdogan. The website is a huge help because it goes through sample scripts. Mutlu is a fantastic resource due to his extensive knowledge with remote sensing and using Google Earth Engine.
In the past two weeks, I’ve attended several trainings and meetings. The meetings are interesting to meet new people and learn about what the Water Quality personnel are doing across Wisconsin. Many of my co-workers are performing field work with lakes, streams, and wetlands. Some of this work includes taking samples of lake water (for water chemistry), monitoring wetlands, managing/monitoring Aquatic Invasive Species (AIS), and collecting optical data with back-scattering light sensors (something I hope to help with) .
An important training held annually for the Water Quality Bureau is the Water Resources Field Training. This year the training was held at the Central Wisconsin Environmental Station near Steven’s Point over two days. This was quite an experience for me. I got to learn about AIS species, was reminded about being safe in the field (watch out for those ticks!), learned more about harmful algal blooms, and I was exposed to some field training that went over some of the field work that the Water Quality department performs. This training was important to me because I saw what other people are doing in the mission to measure/analyze/improve water quality in Wisconsin.
Trip to Central Wisconsin Environmental Station for Water Resource Field Training:

Smells like team spirit or maybe that’s bug spray? 

Beautiful red pines and some black cherries leading to Sunset Lake 
How many bull frogs do you see?
Google Earth Engine script

When not in meetings or trainings, I worked on my script to replicate the current water quality data products produced from satellite imagery using ENVI. Unfortunately, scripting never goes as planned. Some of my time was spent working out (and still working out) errors and manipulating the script to do what I want. I’m pretty sure I’ll need some magic and unicorn sprinkles to get it to work successfully.

Despite the typical setbacks, I am happy about the progress of my work. Currently, the script pulls in raw satellite imagery of Wisconsin. The raw imagery is then transformed to Top of the Atmosphere reflectance to get better water quality data from the satellite bands. A mask from the current protocol is applied to remove land, shallow water, water vegetation, lake edges, and clouds. At this point, only water remains. Another part of the script brings in points of Secchi data to cross reference ground truth data with the satellite images. This information is then exported as a CSV. I am continuing to improve and work toward replicating the current protocol process.
Below is a link to a Google Earth Engine script I created from some of the things I’ve learned. The script runs a supervised classification with training sites selected as points in the map area.
https://code.earthengine.google.com/bccf49265896fd8e618dfdc45011098d
Note: a Google Earth Engine account is required to run scripts.
The pictures show the results of the added map layers. I selected 5 classes: forest (dark green), bare soil (brown), water (blue), urban area (light pink), and farmland (light green).

Classification with training sites 
Classification of land cover 
Landsat 8 TOA Satellite Image with training sites 
Landsat 8 TOA satellite image

Zoomed view of classification 
Zoomed view of TOA satellite image
The Big Picture
These past few weeks I’ve met a lot of wonderful people who are doing incredible things. I’ve learned a large amount in three weeks and value the connections I am continuing to make. I’m grateful for my supervisors, Tim and Daniela, support and encouragement in this project. Connecting to the work I do is important to me. Knowing how much people care about the water in Wisconsin helps drive my motivation for this project. Attending the meetings and trainings opens my vision to the bigger picture which I can connect to my work as a piece of that picture.
In this project, I am actively applying what I learned in my EOI program. I am tapping many sources to work toward my goal. Some of those are websites, research papers, and most importantly people. The EOI program has helped prepare me to work in different platforms, work through various problems, and manage/explore large quantities of data (in this case, many satellite images and Secchi measurements). I hope to continue to learn and grow from this experience.
