Observing Ecosystems from Orbit
Karl Fred Huemmrich, ecologist at the University of Maryland, Baltimore County
Media Credit: Karl Fred Huemmrich
April 23, 2018 • By Amelia Williamson Smith, Staff Writer
Ecosystem productivity—the ability of an ecosystem to convert carbon dioxide into biomass—is the process that supports all life in an ecosystem. But understanding ecosystem productivity and how it varies daily, seasonally, and across vegetation types is important for more than just ecologists.
Because ecosystems are one of the largest carbon “sinks” (that is, they remove carbon dioxide from the atmosphere), looking at variations in ecosystem productivity is crucial to understanding changes in the carbon cycle and in Earth’s climate. Farmers, foresters, and land managers can use information about the productivity of their fields to improve crop and forest growth and optimize utilization of resources such as water and fertilizer.
Small improvements can go a long way toward helping global issues such as hunger and sustainability; for example, agriculture accounts for approximately 80 percent of the water consumed in the U.S.
Scientists study ecosystem productivity on the ground at specific locations where there are towers that use a technique called “eddy covariance” to measure gas and energy movement between Earth’s surface and atmosphere. Sensors on the eddy covariance towers measure the movement of carbon dioxide between the atmosphere and the ecosystem, but this information only tells scientists about the immediate area around the tower.
To better understand ecosystem productivity down on the ground, Karl Fred Huemmrich, an ecologist at the University of Maryland, Baltimore County, looked upward—to the International Space Station (ISSInternational Space Station).
Huemmrich and his team used imagery from the Hyperspectral Imager for the Coastal Ocean (HICO), which operated on the ISS from 2009 until 2014. They combined the HICO imagery with data from eddy covariance towers on the ground to develop algorithms for measuring ecosystem productivity.
“The ground data gives you a continuous measurement of the carbon exchange, but it does this at only one spot, and you don’t know what’s happening around it,” Huemmrich said. “Hyperspectral data from the space station gives you an image of the whole area, but you only get it at specific points in time, when the ISS passes over—so the key to this study was to combine the hyperspectral data with ground measurements.”
Correlating Data from the Ground and Space
Ecologists use the flux, or movement, of carbon dioxide from the atmosphere into an ecosystem to estimate production because plants take in carbon dioxide for photosynthesis. A plant’s productivity varies throughout the day and over a growing season, and these variations can be seen in data from eddy covariance towers.
“In the ground data, you can see the carbon dioxide getting drawn out of the atmosphere by the ecosystem as the sun comes up, and at the end of the day the plants’ photosynthesis shuts down, and you can see the respiration of the plants and the soil go into the air,” Huemmrich said. “The question was, can we see that happen in data from space?”
Huemmrich and his team used data from four sites with different types of vegetation and different seasonal patterns: a cold grassland in Canada, a Douglas fir site in Oregon, a Mediterranean grassland in southern California, and a shrub land in southern California.
Media Credit: Karl Fred Huemmrich
Huemmrich and his team wanted to see whether they could develop algorithms to correlate HICO imagery from the ISS with data from eddy covariance towers. They wondered whether the algorithms could be used to detect variations in productivity over the course of a day and over the growing season for different types of vegetation and landscapes.
To do this, Huemmrich and his team found four sites across the U.S. and Canada that had both eddy covariance towers and HICO imagery. The four sites had different types of vegetation and seasonal patterns (forest land, shrub land, and warm- and cool-season grasslands).
The team extracted pixels from the HICO images at the locations of the eddy covariance towers. This gave the researchers the two pieces of information they needed—hyperspectral reflectance data from HICO and gross ecosystem productivity data from the eddy covariance towers. Having these two sources of information for the same location allowed the researchers to directly relate the two.
Plant Pigments and Productivity
Researchers can use hyperspectral reflectance to look at variations in ecosystem productivity due to changes in chemicals called “pigments” in the leaves of plants. These pigments, such as chlorophyll, play an important role in photosynthesis. Concentrations of pigments in a plant’s leaves vary depending on the physical conditions of the plant. These variations can signal periods of stress and reduced photosynthesis, which tells researchers about the plant’s productivity.
Changes in leaf pigment concentrations also result in subtle changes in the hyperspectral reflectance of plants. Huemmrich and his team used several approaches to develop algorithms to detect these subtle changes in the hyperspectral data. The research team found not just one but multiple robust algorithms that were successful in detecting these changes in different types of vegetation at different times of day and multiple times during the growing season.
Map of gross ecosystem production derived from HICO imagery from the ISS for the study area around Lethbridge, Alberta, Canada. Image collected July 2, 2013. The triangle indicates the location of the flux tower. The area is a mixture of grassland and crops, resulting in a complex spatial pattern of productivity.
Media Credit: Karl Fred Huemmrich
“The best of the algorithms described more than 80 percent of the variance in photosynthesis at the time the space station went over the location, which was very good,” Huemmrich said. “If you set up additional flux towers right next to the one that’s there, you wouldn’t do a whole lot better than that.”
These findings suggest that robust algorithms could be used with hyperspectral data from space to describe ecosystem productivity, potentially on a global level. Such data would enable scientists to model how changes in carbon dioxide levels in the atmosphere will impact future agricultural production and to predict ecosystem stability.
The world’s increasing population has put higher demands on agricultural crops and forests, while increasing environmental stress and Earth’s changing climate have resulted in additional pressure on these ecosystems. It is important to monitor changes in crop, rangeland, and forest production, but it is currently difficult to obtain timely information on variations in productivity. Low Earth orbit(Abbreviation: LEO) The orbit around the Earth that extends up to an altitude of 2,000 km (1,200 miles) from Earth’s surface. The International Space Station’s orbit is in LEO, at an altitude of approximately 250 miles. provides a valuable opportunity to collect such data; for example, the space station’s orbit covers most of the Earth’s populated areas and agricultural regions.
A Valuable Toolbox to Study Plants from Orbit
The next step is figuring out how to get information about ecosystem productivity to people like farmers, foresters, and land managers who can use it. Huemmrich said he and his team can begin working toward this goal once hyperspectral reflectance data from the space station is available again.
Although HICO is no longer operating on the ISS, several instruments planned for installation on the space station over next few years will provide a valuable toolbox for researchers like Huemmrich to study the physiology of plants on the ground from space—and who knows what future imaging platforms in low Earth orbit might allow.
“What’s really cool to me,” Huemmrich said, “is that these instruments are on the space station because they each have their own project, but because the ISS exists as a place where instruments can accumulate, you end up with a suite of instruments in which the total is greater than the sum of the parts—and that is powerful.”