| 20 May 2026
Does the biosphere lift nutrients against gravity, or redirect solar energy? A thermodynamic reframing of planetary biogeowork
Abstract. We use the term biogeowork for the thermodynamic work performed by the biosphere in shaping the Earth's energy and entropy budgets – whether supplied by metabolism or mediated through biological structures. Biological work at the Earth's surface is often imagined as mass transport against gravity – from whale pumps to plant transpiration – yet this framing conflates fundamentally different energy pathways. The biosphere shapes these budgets far beyond its own metabolic energy supply, yet biological energy supply and biologically mediated solar fluxes have not been clearly separated in a single quantitative framework. We propose a three-component decomposition of the biosphere's thermodynamic role: (i) active biogeowork, WA, the mechanical work performed using metabolic free energy derived from gross primary production (GPP); (ii) mediated biogeowork, ΔΦM, the solar-driven flux redirected through biological structures – dominated by the biogenic enhancement of latent heat flux, ΔLEbio, relative to an abiotic counterfactual land surface; and (iii) the resulting entropy-export enhancement, ΔṠbio. Using published global datasets, we estimate WA ~ 1020 J yr−1 (≈2 % of GPP), ΔLEbio ~ 5 × 1023 J yr−1 (≈30 % of global latent heat flux of 1.3 × 1024 J yr−1), and ΔṠbio ~ 2 × 1020 J K−1 yr−1 (≈1–2 % of total planetary entropy production). The dimensionless leverage ratio Λbio ≡ ΔΦM / WA ~ 5 × 103 (range: 3–7 × 103) quantifies the amplification by which biological infrastructure redirects solar energy per unit metabolic investment. The decomposition clarifies the distinct pathways through which the biosphere shapes the planetary energy budget, links biosphere degradation to planetary-scale thermodynamic consequences, and yields testable predictions connecting land-cover change, Bowen-ratio variability, and planetary-scale entropy export.
General assessment
The manuscript draws attention to the fact that, while mechanical work performed directly by living organisms is relatively small, the total energy fluxes mediated by living organisms, most notably latent heat fluxes, are several orders of magnitude larger. I agree that these quantities should not be confused, and I find the attempt to distinguish them potentially useful.
At the same time, the manuscript does not yet show that such a confusion is actually present in the existing literature. In particular, lines 35–43 introduce a “naive formalisation” in which the gravitational work associated with transpiration is treated as metabolically powered, but no published example of this accounting is cited. The either/or dichotomy emphasized in the title would therefore benefit from being situated more explicitly in relation to previous work. The distinction could be situated more clearly by identifying studies in which the two forms of energy have been conflated or, alternatively, by presenting it more modestly as a useful clarification and synthesis.
It is nevertheless legitimate to review metabolically powered work and compare it with biologically mediated energy fluxes. What requires further development is the proposed causal interpretation of their ratio. A defining property of natural ecosystems is their capacity to maintain liveable conditions in a dynamic equilibrium by compensating for environmental disturbances. This regulatory capacity depends not only on the magnitude of energy fluxes, but also on their informational organization—that is, on the specific structures, pathways, and feedbacks through which energy and matter are channelled.
Two ecosystems may sustain similar latent heat fluxes while differing profoundly in their capacity to stabilize environmental conditions. A disturbed, heavily exploited ecosystem may therefore be much less resilient than an intact ecosystem despite comparable total energy throughput. In this context, the theory of biotic regulation developed by Gorshkov (1995) may offer a useful complementary perspective, especially as the manuscript already engages with Gorshkov’s work on the biotic pump. It could help distinguish the magnitude of biologically mediated fluxes from the organization of pathways and feedbacks through which these fluxes contribute to environmental stability.
The manuscript invokes a “long-standing intuition” at least three times, and the proposed leverage ratio is repeatedly presented as its quantitative expression. This intuition is therefore central to the manuscript’s motivation and claimed novelty, but it is not formulated consistently. In the Introduction, lines 51–56, it is stated as the proposition that “small biological energy investments drive large planetary effects”; in Sect. 2.5, lines 113–121, it becomes the claim that metabolic work reorganizes much larger solar-driven fluxes and thereby gives life “disproportionate control over planetary conditions”; and in Sect. 4.2, lines 186–190, the control intuition is attributed specifically to Lovelock, while Kleidon is credited more generally with viewing the biosphere as an active planetary participant. These are distinct propositions concerning, respectively, magnitude, causal leverage, and environmental regulation. It would be helpful to define more precisely which of these propositions the framework is intended to quantify.
Because the cited intuition plays such a central role, direct quotations and page references would help clarify what Lovelock (1979) and Kleidon (2010) actually proposed. In particular, it would be useful to establish whether either author formulated the specific idea that a small metabolic-work input controls or amplifies a much larger externally powered energy flux. Kleidon’s framework emphasizes the substantial generation of chemical free energy by photosynthetic life, whereas Lovelock’s argument concerns planetary regulation; neither proposition is obviously equivalent to the leverage concept introduced here. If the specific metabolic-leverage claim cannot be documented in these sources, it would be clearer to present it as the present author’s hypothesis or synthesis rather than as the direct quantification of a pre-existing shared intuition.
At present, the historical “intuition” seems to provide a connection that the analysis has not yet fully established: the connection between a numerical disparity of fluxes and a demonstration of causal environmental control.
The manuscript proposes to decompose “biogeowork” into three quantities associated with biological activity: active biogeowork Wₐ, mediated biogeowork ΔΦₘ, and entropy-export enhancement ΔṠ_bio. These are initially described as physically independent components, although the third is explicitly derived from the second.
In the Introduction, lines 51–53, and at the beginning of Sect. 2, lines 60–62, active biogeowork Wₐ, mediated biogeowork ΔΦₘ, and entropy-export enhancement ΔṠ_bio are described as “physically independent components.” Yet Sect. 2.3 states immediately after Eq. (5) that ΔṠ_bio is “not an independent energy flux but a thermodynamic consequence of ΔΦₘ.” The same ambiguity appears in Sect. 2.5, lines 118–121, where the three components are again called independent while the entropy term is described as a diagnostic derived from ΔΦₘ. Once T_rad and T_surf are prescribed, Eq. (5),
ΔṠ_bio = ΔLE_bio (1/T_rad − 1/T_surf),
simply multiplies ΔLE_bio by a constant. It would therefore be clearer to present the entropy term as a derived diagnostic rather than as a third independent component.
More importantly, Eq. (5) does not establish the claimed planetary consequence. Sect. 2.3 treats latent heat as being transferred from the surface at T_surf = 288 K to an atmospheric radiating temperature T_rad = 255 K and interprets the resulting expression as enhanced planetary entropy export. Latent heat transport, however, is an internal redistribution of energy within the Earth system. It can strongly influence surface temperature by removing heat from the surface and releasing it higher in the atmosphere, but this does not by itself imply an additional entropy flux to space.
At global steady state, outgoing radiative energy is constrained by absorbed solar radiation and hence, to leading order, by planetary albedo. Radiative entropy export can additionally depend on the spatial, spectral, and angular distribution of radiation, but it cannot be inferred directly from a change in the internal partition between latent and sensible heat. Equation (5) might therefore represent, at most, a simplified internal heat-transfer entropy-production term rather than an independently established increase in entropy emitted to space.
For the same reason, the prediction in Sect. 5.5, lines 250–255, does not presently follow from Eq. (5). A genuine change in planetary entropy export should indeed be sought in top-of-atmosphere radiation. However, a change in latent heat transport that leaves planetary albedo unchanged should produce no leading-order change in the global steady-state outgoing radiative flux. A detectable entropy-export signal would have to arise through changes in albedo, a transient planetary energy imbalance, or changes in the spatial, spectral, or angular distribution of radiation. The manuscript would be strengthened by identifying and quantifying one of these pathways.
Recommendation
I regard the distinction between metabolically supplied work and biologically mediated external fluxes as valid and potentially useful. The manuscript also raises an important broader question concerning how living systems influence environmental processes whose energy is supplied externally.
However, the present version does not yet demonstrate that it resolves an existing confusion, that the leverage ratio represents a causal relationship, or that the proposed entropy term has the stated planetary meaning. More generally, the framework would benefit from including the informational and regulatory organization that distinguishes an intact ecosystem from a degraded system with a similar gross energy throughput.
I therefore recommend a major conceptual revision. The required changes go beyond correcting individual numerical estimates, but I believe the underlying question could still support a valuable Perspective.
Further comments for revision
1. Relation to Murota (1985)
Sect. 4.2 should describe Murota’s contribution more precisely. Murota estimated a photosynthesis-related entropy term from the plant-transpired fraction of the gravitational power of the water cycle. This is not the same physical quantity as the counterfactual latent-heat term introduced in Eq. (5).
The numerical comparison also requires caution because the present estimate and Murota’s estimate are normalized by different definitions and magnitudes of total entropy production. The apparent result that both approaches assign an order-one-percent role to biology therefore does not by itself demonstrate quantitative continuity. The present work may be thematically related to Murota’s analysis, but the nature of the proposed continuation or refinement should be explained more carefully.
2. Mechanical power of the hydrological cycle
The value 3 × 10²⁰ J yr⁻¹ cited in the Introduction appears to correspond approximately to the power maintaining continental runoff, rather than to the total mechanical power of hydrological cycling. It is equivalent to only about 0.019 W m⁻² when averaged over Earth’s surface.
The manuscript should therefore identify precisely which mechanical-power component is being compared with the global latent heat flux. Continental runoff power, gravitational precipitation power, atmospheric kinetic-energy generation, and total hydrological mechanical power are different quantities. A narrow component should not be presented as representative of the mechanical power or efficiency of the hydrological cycle as a whole.
3. Causal direction of the leverage argument
It is not self-evident that a small metabolic-work input is used to mediate a much larger latent heat flux. In plants, transpiration and stomatal gas exchange permit the CO₂ uptake that generates photosynthetic free energy and makes subsequent metabolic expenditure possible. The large externally powered flux is therefore not simply a consequence of a prior small metabolic investment; it is also part of the process through which the organism acquires the free energy used for metabolism.
Past photosynthesis has constructed the roots, xylem, and leaves that mediate present transpiration, while present metabolism maintains these structures and regulates stomata. The relationship is therefore a coupled feedback operating over different timescales, rather than a simple sequence in which metabolic work is invested first and subsequently amplified into latent heat transport.
The ratio
Λ_bio = ΔΦₘ/Wₐ
does not by itself establish causal leverage. Demonstrating leverage would require identification of a specific control process, its energetic cost, and the marginal change in the externally powered flux caused by that control process.
4. Definition of the abiotic counterfactual
The central quantity
ΔLE_bio = LE_observed − LE_abiotic,cf
cannot be evaluated without a precise counterfactual. More fundamentally, there may be no unique abiotic latent heat flux until the processes permitted to adjust and the system boundaries are specified.
Removing vegetation reduces transpiration, but it simultaneously changes surface temperature, albedo, aerodynamic roughness, soil moisture, cloudiness, atmospheric circulation, and moisture convergence. Some of these responses may compensate for the direct loss of transpiration. For example, surface warming or altered radiative input may enhance bare-soil evaporation, while coupled circulation changes may alter ocean evaporation and regional precipitation.
The resulting latent heat flux therefore depends on the spatial and temporal scale of the comparison and on which components of the climate system are allowed to respond. Forest-to-cropland conversion, forest-to-bare-soil conversion, and present Earth versus a globally abiotic Earth are fundamentally different counterfactuals and need not produce the same ΔLE_bio.
The manuscript should specify whether the estimate applies over land or over the whole planet; how the abiotic surface’s albedo, roughness, soil hydrology, water-storage capacity, and water availability are defined; and whether coupled changes in ocean evaporation, circulation, clouds, and precipitation are included. Without a common experimental definition and system boundary, ΔLE_bio is a model-dependent difference between alternative climate states, and Λ_bio is correspondingly not unique.
5. Meaning of “mediated biogeowork”
The quantity ΔΦₘ is a difference between energy fluxes. In the principal application, it is a difference in latent heat flux. It is not mechanical work and is not necessarily a rate of free-energy generation. Calling it “mediated biogeowork” may therefore reintroduce some of the category ambiguity that the manuscript is seeking to resolve. The word “work” should be reserved for a physically specified conversion of free energy.
Reference
Gorshkov, V. G.: Physical and Biological Bases of Life Stability: Man, Biota, Environment, Springer, Berlin, 1995.