Embedded Files
Curious about life as an environmental researcher?
Lab Work
Lab Work
Field Research
Field Research
Modeling
Modeling
Research aims to expand our knowledge of the world around us. As an environmental researcher, my work focuses on understanding the impact of natural and engineered systems on our climate. As an engineer, I prioritize work with immediate and real-world applications.
Different research questions require different approaches. Here, I provide an overview of three different types of research based on my experience: lab work, field research, and modeling. I hope this page can help future researchers better understand what life is like as a scientist.
Lab work
Lab work
In laboratory-based research, we typically create a controlled environment that allows us to study the effect of changing a single part, or variable, of this system. We work very hard to isolate the cause of any differences we might observe across our experiments, so often laboratory studies may be highly simplified compared to real-world scenarios.
Methane-consuming bacteria grown in laboratory bottles
The image on the left shows bacteria I grew in a laboratory. These bacteria eat methane and are called methanotrophs. I grew the bacteria in each bottle under the exact same conditions, where the only difference between groups was the mix of gases I injected into the bottle. That way, I could be fairly confident any difference between groups was likely caused by the differences in gas mixture.
However, if I grew these bacteria at an industrial engineering facility, they may be exposed to many other gases and environmental contaminants. These conditions may be difficult to reproduce in a lab, and my lab studies may not perfectly predict what will happen.
Sometimes in laboratory research, we can't use the exact species we are interested in for our experiment: it might not be practical or ethical. So instead we use what is called a model organism.
During my PhD, I was interested in seeing how methane-consuming bacteria might impact the growth of fish and shrimp, providing benefits for aquaculture. However, it would have been impractical to grow a high number of large shrimp in my lab space. So instead, I used brine shrimp - a very small type of shrimp that grows to be only a few millimeters long. With this model organism, I could hatch thousands at a time, and use hundreds in each experiment.
Microscopic image of a brine shrimp
Laboratory bioreactor (4 liters) growing methane-consuming bacteria
Day-to-day life in the lab follows certain patterns. First, plan your experiment: learn how to use any new equipment, order supplies, finalize methods, and develop a data collection plan. Next, run the experiment itself. Depending on the type of science, this could take anywhere from a few minutes to many months! When I grew bacteria in bottles (above), I ran 5 day experiments. When I grew bacteria in a bioreactor, I studied the system for 3 months!
Lab work is exciting because you get to learn so many hands-on skills. Sometimes a research group will specialize in one particular kind of analysis (for example, analyzing samples for certain molecules), and apply it to many different areas. On the other hand, some lab groups will focus on an area, like water treatment, and try to use many different tools to answer their questions.
Field Research
Field Research
While I was a postdoc at Stanford, I led a large field project testing different technologies used for detecting methane, a potent greenhouse gas. We tested how well different sensors mounted on airplanes, satellites, drones, and ground networks could determine if methane was leaking and at what rate.
Setting up the gas release tower (left) and wind tower (right).
Scientists use field research experiments to closely mimic the real world. As a result, there is a huge variety of field research! Because we can't control conditions in the field the way we would in the lab, we document environmental conditions while we're running an experiment. We can then use these measurements to understand the variability we might see in the results.
To test methane sensors, we picked a location in the desert in Arizona to provide a clean background without other methane sources nearby. Each day, we released a known amount of methane, timed with when an airplane or satellite was overhead conducting its measurement. I spent a lot of time designing and troubleshooting a system to precisely measure the gas we released.
We ran this experiment in the field to closely mimic how these sensors are used in the real world. However, performing the experiment in the real world introduced variables we couldn't control - like the weather. This can make field work more difficult, but also incredibly rewarding. You may spend months planning an experiment before you actually carry it out! And you always need back-up plans in case things don't go as planned.
Setting up the methane metering system in the field in Arizona to precisely measure our gas release rate
(Photo Credit: Richard Chen)
Shelter from a sandstorm in our truck after a hard day setting up the field site!
Field research requires thinking on your feet and making use of the limited resources available at your site, all while maintaining the experiment's scientific integrity. In the methane sensor tests, we dealt with delays in equipment delivery, sandstorms (see the photo), unexpected equipment failure, logistical miscommunication, and so much more. However, this is part of what makes field research so exciting and interesting!
Modeling
Modeling
Modeling work focuses on developing new analysis tools, as opposed to collecting original data. Here, we use existing data and fundamental scientific or engineering relationships to make predictions. I started my first modeling project during COVID-19 shelter-in-place, when my lab was temporarily closed. And I've been modeling ever since!
Map of methane emissions and flaring in the United States (El Abbadi et al, 2022)
My modeling work aims to understand patterns on a large geographic scale. The map of the United States shows methane emissions from different industrial sectors. The data was originally reported by facilities or collected using satellite images. I used these thousands of points as inputs to my model.
Modeling is exciting because it allows us to imagine worlds that do net yet exist. We can create different future scenarios and make predictions. However, our model is only as good as the underlying data and assumptions! So it's important to test the limits of our model, and clearly state our approach. This is also why it is so important to make our code available, and to use software that is freely available. Personally, I'm a fan of Python!
Diagram showing the process model I built to estimate cost for industrial production of protein from methanotrophic bacteria (El Abbadi et al, 2022)
Example coding in Python
When I'm modeling, I spend most of my time working with data: finding, organizing, and cleaning it, then incorporating it into the model. Of course, I also need to code the model itself, and I enjoy learning new ways to make my code more efficient and easier to interpret by others.
If you've made it to the bottom of this page, congratulations!
I hope this was helpful, whether you are considering your own future or are just curious about what researchers do every day.
Page updated
Google Sites
Report abuse