MALIBU Project Taps NEON Data for Space-borne Data Validation

November 14, 2018

Ecological data collected in the field gives researchers a window into how ecosystems and climate are changing in a particular area. But to see the bigger picture, you have to take a step back—way back. All the way to space.

Satellite data provide an important perspective on Earth’s systems—including land, atmosphere, oceans and cryosphere—and how they are changing as a result of natural and human-induced stresses. MALIBU (a joint project of NASA’s Terrestrial Ecology Program, NOAA/NESDIS/STAR, and BlackSwift technologies) is using NEON data to verify that the satellite-derived time series data we are getting from above are accurate and reliable over time. 

Watching Earth Systems from Space

Satellites have played an essential role in monitoring the earth systems for decades, including capturing data to track changes in vegetation patterns, global temperatures, albedo (the amount of energy reflected by the earth), sea ice extent and other critical climate indicators. The Global Climate Observing System (GCOS) has developed a list of Essential Climate Variables (ECVs) that are considered to be key indicators for observing global climate change. These ECVs are both currently feasible for global implementation and meet the requirements of the United Nations Framework Convention on Climate Change. Terrestrial ECVs include river discharge, water use, soil moisture, above ground biomass, albedo, fire disturbance and soil carbon, among others.

NASA and NOAA are two of 60 space agencies around the globe using satellites to collect data for climate and ecosystem researchers. These efforts are coordinated through the Committee on Earth Observation Satellites (CEOS), an international group dedicated to establishing common methods for data collection, formatting, validation and system calibration for satellite-based observations. These efforts ensure that data collected by different agencies will be directly comparable and are of consistent quality.

CEOS space agencies operate 153 satellites, each of which has its own unique orbital path and set of data products that it collects. Together, they provide a complete picture of the earth and all of its systems, with data that now goes back nearly 50 years.

MALIBU: Validating and Calibrating Satellite Data

Having an extended record of data from space is an invaluable resource for climate change researchers. However, data accuracy can be hard to maintain over time. Many of the satellites observing earth systems have aging instrumentation. Over time, this can lead to "data drift" caused by changes in the satellite platform and the instrumentation itself rather than actual changes in the systems it is observing.

The MALIBU project (for Multi-AngLe Imaging BRDF Unmanned aerial system) is part of a CEOS-led effort to validate and calibrate satellite data. The MALIBU drone is fitted with an instrumentation payload designed to mirror the data collected by NASA/NOAA satellites that collect land-based data products. The project falls under the CEOS Working Group on Calibration and Validation – Land Product Validation Subgroup.

Dr. Miguel Román, a Physical Research Scientist at NASA's Goddard Space Flight Center and Chair of the CEOS Land Product Validation working group, explains, "When we're trying to document something like climate change, small errors in our data can lead to big problems with our models and forecasts. If an instrument starts returning inaccurate data, that bad data will result in faulty models and conclusions that lead to bad policy decisions." For example, in one case what first looked like an authentic change in vegetation patterns turned out to be caused by an error in the instrument collecting spectral data in the visible channels.

The MALIBU drone collects a reference data set (traceable to rigorous SI standards) that can be compared to the data gathered by the satellite. Land products being verified by MALIBU include albedo, snow cover, plant phenology metrics and a variety of other vegetation indices, measures of reflectivity and spectral data. In order to mimic the angle of data collection from the satellites, it uses a multi-angular instrument system. Comparing these different data sets will allow researchers to detect any anomalies in the satellite data that may require recalibration.

Dr. Zhuosen Wang, an Assistant Research Scientist at the University of Maryland and co-lead of the CEOS Land Product Validation working group surface radiation/albedo focus area, is acting as the Science Principal Investigator in charge of developing data collection methods and algorithms for processing and interpreting the data.

NEON and MALIBU: Collecting the Reference Data Set

Before the MALIBU data can be used for calibration purposes, its data and instruments must first be validated against known reference data. MALIBU is using data from some NEON sites as an anchor to test the validity of the MALIBU data products.

MALIBU and the NASA/NOAA satellites collect data products that are also collected at the NEON field sites, including atmospheric measurements, vegetation indices and canopy chemistry. NEON data provide a good anchor point for the MALIBU reference data because of the way NEON instruments are calibrated and validated. All of the instruments used for NEON data collection are recalibrated on a regular schedule to ensure that there is no instrument drift that could create anomalies in the data. "The people working on the NEON infrastructure are experts in the instrumentation and in ecological data collection. They follow internationally recognized protocols to ensure that the data collected by their instruments is highly accurate and traceable," explains Dr. Román. "If there is a discrepancy between the NEON data and our model, we know our model has a problem, because the NEON data are correct."

Visitors take a tour of the flux tower at SRER. Photo by Abe Karam.

So far, the MALIBU science team has run one flight over the Santa Rita Experimental Range (SRER) NEON terrestrial site in Arizona. The calibration project will incorporate data from the NEON flux towers as well as from NEON's Airborne Observation Platforms (AOPs). “NEON data are so critical that CEOS satellite experts included 11 of NEON’s core sites as part of their global list of super sites for space-borne data validation,” said Dr. Wang. Additional validation flights are planned over NEON field sites at Harvard Forest (HARV) and San Joaquin Experimental Range (SJER). MALIBU has also completed a mission over an AmeriFlux site near Boulder, Colorado.

Everything must be carefully timed for proper instrument calibration. The MALIBU drone flight was timed to be coincident with the NEON AOP flight and with overpasses from the satellites being calibrated (in this case, LANDSAT, MODIS and VIIRS instruments). Looking at data taken under clear-sunny skies during the same time period from all four sources—the satellites, MALIBU, the NEON flux tower and the AOP—will enable researchers to determine whether the MALIBU drone dataset is accurate enough to use as a reference set to monitor satellite instrument drift over time.

"MALIBU will help us maintain continuity of our global satellite datasets over very long periods of time, which is essential if we want to really understand how the climate is changing and how earth’s biosphere and atmosphere work together," says Dr. Román. "NEON data provides another long-term record that we can reference against. We know this data will be collected in the same way and the instruments will be maintained to avoid any drift over the next three decades. That will help us ensure that our satellite data is valid and accurate so we can continue to improve our models and forecasts."

Dialog content.