Schrödinger's cat lives, in a forest flux

Results from globally distributed eddy-covariance observations show that estimates of ecosystem photosynthesis and respiration, which are widely used to test models of global ecosystem function, may have been biased high.
Schrödinger's cat lives, in a forest flux
Cat in a forest

Schrödinger's cat exists in a box, or so the proposition goes. In what state his cat exists is unknown, however, as it is not possible to observe it. Schrödinger, in his wisdom, proposed the straw man argument that the unobserved cat actually exists in two opposing states: both alive and well, and very much inhibited (a.k.a. dead). Where the box is exactly does not really matter, but let’s say it is somewhere in a forest.

Ecosystem scientists are faced with a somewhat similar conundrum. Both photosynthesis and respiration are by far the largest annual fluxes of carbon dioxide between the land surface and the atmosphere, but neither is directly measurable on an ecosystem scale. Scientists have devised different methods to infer both, primarily from eddy-covariance measurements of observed net fluxes of CO2 (the balance between photosynthesis and respiration) and satellite-based proxies. Each approach to estimate global photosynthesis and respiration requires assumptions, however, and much like Schrödinger's cat, the processes underlying these assumptions are often impossible to observe at the ecosystem scale.

Eddy-covariance measurements above the University of Michigan Biological Station (UMBS)
Eddy-covariance instruments, measuring the net exchange of carbon dioxide between the ecosystem and the atmosphere, above the forest at the University of Michigan Biological Station (UMBS) (Credit: Gil Bohrer & Chris Vogel)

One such process is the inhibition of respiration of leaves in the canopy by light. Plant physiologists have long proposed that leaf respiration is inhibited during the day, compared to at night. If true, this could greatly change current estimates of global ecosystem respiration. But although leaf level measurements suggest varying degrees of inhibition, ecosystem scientists lacked the observations and tools to assess whether there was any inhibition of respiration at the ecosystem scale. With a lack of ecosystem-scale evidence, the majority of methods assume no inhibition occurs. I doubt Schrödinger would have been impressed with this Neo-inspired third option: there is no cat!

In our recent paper (Keenan et al., 2019) we devise a method to peer inside the box and test whether the cat, in this case the inhibition of ecosystem-scale leaf respiration during the day, is alive and well. To do so, we use more than 5 million observations of CO2 fluxes between ecosystems and the atmosphere, gathered over the past 30 years by a grass-roots community of scientists all over the world, and distributed though the FLUXNET database ( We found evidence for a strong inhibition of ecosystem respiration, which varies in dependence of ecosystem type. The results imply a potentially large bias in global estimates of photosynthesis and respiration, and a needed revision of how these processes are incorporated into global ecosystem models.

Perhaps most importantly, the results highlight the utility of globally networked science to answer questions that were previously unaddressable. Such discoveries would not be possible without the efforts of 100's of scientists worldwide who make the data they collect openly available, and the efforts of the community to enable synthesis through collaborating and harmonizing observations across 10's of regional networks. And of course the generosity of program managers and funding agencies (though FLUXNET is currently unfunded). The observational records are getting longer (some are nearing three decades, at 20Hz!), providing new opportunities to better understand ecosystem function and responses to environmental change. To learn more about our work and the global FLUXNET database, visit and

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in