A Dynamic Theory for Explaining the Evaporation Paradox and Global Energy Transpiration

Kejing Liu

REVIEW 4 major objections 5 minor 4 references

The evaporation paradox is one oscillator, not separate regional mysteries, this paper argues.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · deepseek-v4-flash

2026-08-04 01:29 UTC pith:PTLCWNKN

load-bearing objection Solid regional trend analysis, but the dynamic theory is asserted, circular, and never validated—reject. the 4 major comments →

arxiv 2608.00049 v1 pith:PTLCWNKN submitted 2026-07-25 physics.geo-ph

A Dynamic Theory for Explaining the Evaporation Paradox and Global Energy Transpiration

Kejing Liu This is my paper

classification physics.geo-ph

keywords evaporation paradoxpan evaporationcloud quantityneutral-delay dynamic systemlogistic mapcomplementary relationshipHuaihe River Basinenergy transpiration

verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

Pan evaporation has widely declined over the past decades even as temperatures rose—the 'evaporation paradox.' Previous explanations treat the paradox as either a radiation effect (more clouds mean less evaporation) or a complementary effect (actual evaporation rises when pan evaporation falls). This paper argues that cloud quantity and pan evaporation together form a single nonlinear, time-delayed dynamic system, like a damped oscillator, and that different observed regional patterns are just different phases of that one system. The paper constructs a model, called the 'steamer,' from a nonlinear second-order neutral-delay equation, shows how the paradox can form, persist, wane, and recur as clouds and evaporation chase each other with a lag, and uses the same relation to discuss global energy exchange and atmosphere–surface stability.

Core claim

The central claim is that the relationship between cloud quantity (Q) and pan evaporation (Epan) is not a fixed correlation but a delayed, oscillatory dynamic system. Analyzing 14 stations in the Huaihe River Basin from 1954–2005, the paper finds that at the annual scale Epan and Q mostly move in opposite directions, but their long-term trends and abrupt-change points are sequential, not simultaneous. This leads to a nonlinear second-order neutral-delay equation (NSNDE) describing how changes in Q drive changes in Epan with a time lag. Reducing this to a logistic-map-like iterative form, the paper derives a mapping expression indicating that a change in Q is a driver of changes in E, with Ep

What carries the argument

The nonlinear second-order neutral-delay dynamic equation (NSNDE), a form of Duffing-type oscillator with a delayed restoring force, models the long-term oscillations of Epan and Q. A Poincaré-map reduction converts it to a 1-D discrete logistic-map-like iteration (eqs. 15, 23, 25), which yields the explicit lagged mapping Q_{t-1} → E_{t+1}. This mapping is the load-bearing bridge that turns a qualitative delay equation into the quantitative iterative relations (eqs. 23, 25, 28, 29).

Load-bearing premise

That the nonlinear delay equation can be legitimately reduced to a one-dimensional logistic map by treating 2π/ω as a discrete time unit and ignoring the delay dynamics beyond that step.

What would settle it

A direct test would be to take the NSNDE with fitted delay τ and compare its forecast of Epan against the logistic-map iteration (eq. 23) on a withheld portion of the Huaihe station data: if the logistic map does not track the oscillations, the central mapping is falsified. Alternatively, search for the predicted phase-lead pattern (Q leading Epan by roughly ωπ−2 time units, eq. 30) in other long-term pairs of cloud and pan records; if no region shows that lag, the model's core claim fails.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

Share X Bluesky LinkedIn Reddit HN

If this is right

If the dynamic-system view is correct, the evaporation paradox is not a single trend to be explained but a multistage process; regional contradictions (Epan decreasing with cloud increase vs. decrease) are compatible phases of one cycle.
The 'complementary relationship' between actual evaporation Ea and Epan is not universal: it holds in some stages of the paradox but fails in others, so regional assessments of water availability from pan data need to specify which stage applies.
Because Q and Epan are phase-lagged, predicting one from the other requires accounting for the delay; annual or decadal correlations alone will be misleading.
The alternating stability of the land surface and atmosphere, tied to the stages of the paradox, could help explain the observed increase in extreme weather intensity during paradox-forming and recurring stages.
The same mechanism of coupled oscillation may apply beyond pans: any surface evapotranspiration metric tied to cloud feedbacks should show the predicted lag patterns.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The paper's logistic-map reduction (taking 2π/ω as a 'minimum unit' and asserting a Poincaré map) is an idealization; if tested, the predicted ε-values and λ-golden-ratio connection would be a way to check the theory, not a result the paper itself claims.
A natural extension would be to fit the NSNDE parameters to long-term station data with an explicit delay estimation, rather than using the conceptual 'steamer' stages qualitatively; a robust fit would be a strong validation.
The stability/entropy argument suggests a testable hypothesis: the chaos parameter µ in the logistic map should correlate with observed extreme-weather frequency in a region, a consequence not directly tested here.
If the model is right, traditional hemispheric or global averages of pan evaporation may obscure the phase structure; regional phase-resolved analyses would show leads and lags that global means do not reveal.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Desk Editor's Note

Solid regional trend analysis, but the dynamic theory is asserted, circular, and never validated—reject.

read the letter

Quick take: the empirical part of this paper is a competent regional case study, but the 'dynamic theory' it claims to build is not derived. The central model is asserted, circular, and unvalidated. If you're short on time, read Section 3 for the Huaihe River Basin analysis and skip the theory.

What's actually here: The paper documents that over 1954–2005, at 14 stations in the Huaihe River Basin, both total cloud amount and pan evaporation mostly declined, while annual changes were more often opposite-signed than same-signed. The Mann-Kendall trends and B-G change points are standard tools, and Table 3's inverse-change counts do support a short-term negative association. The observation that the evaporation paradox isn't a single monotonic relation—that it varies by region and over time—is a fair and useful reminder. The three station types based on change-point ordering are a reasonable descriptive typology.

The problems are in the theory. The paper starts by writing down a second-order neutral delay differential equation (Eq. 11) as if it were the obvious model, then asserts that because the solution is periodic, a Poincare map reduces the system to a 1-D logistic map (Eq. 15). No derivation is given. A second-order DDE has infinite-dimensional state space; even under strong reduction you'd need to show the observed series actually lie on a 1-D attractor. None of that is shown. Eqs. (18)–(19) are supposed to come from combining the logistic fixed point with Priestley-Taylor, but they do not follow dimensionally or algebraically. Then Eq. (22) simply assumes Rn = k/Q, which is precisely the inverse cloud–evaporation relationship the paper claims to predict. So Eq. (23) and everything after is the model telling itself what it already assumed. The four-stage 'steamer' narrative is post hoc; no parameters are estimated, no model is fitted to data, and there is no out-of-sample or independent validation. The claims about global energy transpiration and atmosphere–surface stability rest entirely on this unvalidated machinery.

The empirical section is worth a look, but the theoretical contribution is not. This paper would not survive genuine peer review as written; the derivation gap is load-bearing and the central relation is built in. I would not send it to referees. If the author wants to rehabilitate it, they would need to estimate the delay model from data, show the return map is genuinely 1-D, and test the predicted lagged E–Q relation out of sample. As it stands, it's a case study with an unsupported theory attached.

Referee Report

4 major / 5 minor

Summary. The paper analyzes 1954–2005 cloud quantity (Q) and pan evaporation (Epan) records from 14 stations in the Huaihe River Basin using Mann–Kendall trend tests and Bernaola-Galván change-point detection. It reports that Epan and Q generally decrease in the long term but vary inversely at annual scales, with change-point lags. The author constructs a 'steamer' conceptual model and a nonlinear second-order neutral-delay dynamic equation (NSNDE), claims a reduction to a one-dimensional logistic map, and derives iterative equations (Eqs. 23–29) intended to explain the evaporation paradox as four stages (formation, duration, waning, recurring) and to support conclusions about global energy transpiration and atmosphere–surface stability. The empirical trend analysis is standard and internally coherent, but the theoretical core is not rigorously derived or validated.

Significance. If the dynamic system claim were established, the paper would offer a unifying framework in which regional differences in the evaporation paradox are phases of one delayed oscillator, and it would link cloud–evaporation dynamics to global energy exchange. However, the significance is conditional on a derivation that is currently unsupported. The manuscript's positive contributions are the careful compilation and basic statistical analysis of a regional dataset and the clear framing of the paradox as a multi-stage phenomenon. No machine-checked proofs, reproducible code, parameter estimation, or quantitative model validation are provided, so the theoretical claims remain qualitative. The paper's value is thus limited to an empirical descriptive study until the model is properly derived and tested.

major comments (4)

[Section 4b, Eqs. (11)–(15)] The central reduction from the NSNDE (Eq. 11) to the one-dimensional logistic map (Eq. 15) is not demonstrated. Eq. (11) is a second-order neutral delay differential equation; its solution space is infinite-dimensional, and a Poincaré section generically produces a map of dimension equal to the state-space dimension, not necessarily one. The text asserts that because y is periodic, 'this formula yields a Poincare map' and 'the problem can be simplified to a 1-D discrete dynamic system,' but no derivation or conditions are given. Moreover, no phase-space reconstruction, return-map analysis, or any empirical evidence is offered to show that the observed Q/Epan series lie on a one-dimensional logistic attractor. Equations (23), (25), (28), and (29) all depend on Eq. (15); if this reduction is invalid, the quantitative iterative model does not follow from the NSNDE.
[Section 4b, Eqs. (16)–(19)] Equations (18) and (19) are not algebraic consequences of Eqs. (16) and (17). Eq. (16) sets E_t = (α−1)/β for the fixed point, and Eq. (17) is the Priestley–Taylor relation. Equating these expressions does not yield α = 1.26 s R_n (with no denominator) or β = (⋯) as stated; the displayed formulas are dimensionally inconsistent (α and β should be dimensionless, while R_n has units of W m⁻²). The derivation also silently introduces E0 and defines it as maximum potential evaporation but does not use it consistently. These inconsistencies undermine the parameter identifications used in later equations.
[Section 4b, Eq. (22)] The inverse relation R_n = k/Q is assumed ad hoc, with no empirical support or physical justification beyond the desired conclusion that E_{t+1} decreases as Q_t increases. This makes the 'prediction' in Eq. (23) circular: the model is constructed to yield an inverse E–Q relationship, so the subsequent interpretation that this is a novel dynamic prediction is not justified. The reader's report correctly notes that this step, combined with the free parameters in the NSNDE and logistic map, allows the model to accommodate any observed sign of the E–Q correlation, thereby reducing its falsifiability.
[Section 4a and Section 5] The three empirically classified station types (type I, II, III) are post hoc assigned to stages of the 'steamer' narrative. The manuscript offers no quantitative criterion for matching a station to a stage, no test of the model's stage predictions against independent data, and no uncertainly quantification. The discussion of actual evaporation and the complementary relationship (Section 5) is qualitative; the statements such as 'Ea is likely to decrease' or 'may not satisfy the complementary relationship' are not derived from the preceding equations. The entropy/stability analysis in Section 5c also uses sign conventions (e.g., dR_n/dt and dE/dt) without a formal Lyapunov argument, so the conclusion that the atmosphere and land surface alternately become unstable is not supported.

minor comments (5)

[Abstract] There are typographical errors: 'exits in worldwide' should be 'exists worldwide'; 'spatio – temporal' should be 'spatio-temporally.'
[Data and Figure 4] The text states records span 1954–2005 and later says 'during this 52-year period'; the number of years should be checked (1954–2005 is 52 years inclusive, but earlier text says 51 years in Tables 1–3). Consistency in the number of years used in the statistics is needed.
[Figure 4 and Tables 1–3] The figure captions and table notes are dense; the meaning of 'MKTE*' and significance codes (e.g., '××' for p>0.25) should be clarified. Also, the station 58314 is described as 'type III' but the criteria for type III are not stated as precisely as for types I and II.
[Section 4b, Eq. (29)] Equation (29) is not a correct, well-formed expression: the variable ω is used both as a parameter in the original NSNDE and as an index in the equation, and the notation 'ω=1, 2, …' is inconsistent with the earlier definition of ω as a frequency. This makes the periodicity statement ambiguous.
[References] Some references are incomplete or inconsistently formatted (e.g., 'Hirsch, M. W., S. Smale, R. L. Devaney, 2008' lacks a publisher city; 'Kendall, M., and J. D. Gibbons, 1990' is listed but the in-text citation is 'Kendall 1975'; 'Brutsaer, W., and M. B. Parlanger, 1998' is a typo for Brutsaert and Parlange).

Circularity Check

3 steps flagged

The inverse E-Q relation is inserted at Eq. (22) and the logistic maps are adopted by assertion, so the central 'prediction' and stage classification reduce to the model's construction.

specific steps

ansatz smuggled in via citation [Section 4b, between Eqs. (13)-(15)]
"This formula yields a Poincare map (Hirsch et al. 2008) with a periodicity of y; on this basis, the problem can be simplified to a 1-D discrete dynamic system, and a simple iterative function can be used to present it in ideal form. The most popular and effective iterative function for dynamic systems is the logistic map (Hirsch et al. 2008; Liu et al. 2003; Lorenz 1963)"

Eq. (15) is not obtained from the NSNDE (Eq. 11); periodic solutions of a second-order neutral delay equation do not imply a one-dimensional logistic map. The paper simply asserts the Poincare map 'can be simplified' and then adopts the logistic map by citation. Because Eqs. (23), (25), (28), and (29) all substitute this logistic form for E and Q, the subsequent 'dynamic theory' equations rest on an imported ansatz rather than a derivation from Eq. (11).
self definitional [Section 4b, Eqs. (22)-(23)]
"Given that Rn decreases as Q increases, we establish the following relationship: Rn=k/Q (22) ... This equation reflects the reverse trend relationship between Q and E. Et+1 decreases as Qt increases."

The inverse E-Q relation is inserted by Eq. (22) before it is 'predicted' by Eq. (23). Priestley-Taylor (Eq. 17) already makes E proportional to Rn; setting Rn=k/Q therefore forces E to decrease as Q increases. Eq. (23) is an algebraic restatement of the input assumption, not an independent prediction. The same assumption is then used to classify observed station types as stages.
self definitional [Section 4b, Eqs. (24)-(27)]
"In addition, Qt can be expressed in a logistic mapping-like form as Q_t = μ Q_{t-1}(1 - Q_{t-1}) (24). This expression can be substituted into eq. (23) to obtain ... The mapping expression indicates that a change in Q is a driver of changes in E and that a change in Epan lags behind changes in Q."

Q is declared to obey a logistic map that depends on Q_{t-1}; E_{t+1} is made to depend on Q_t in Eq. (23). Substituting the two logistic forms into Eq. (25) yields the mapping Q_{t-1} -> E_{t+1} (Eq. 27). The claimed causal/driver statement and lag are therefore built into the definitions of the iterative equations, not inferred from the NSNDE or from data.

full rationale

The paper's empirical M-K and B-G analyses are legitimate descriptions of the Huaihe River Basin data, but the theoretical derivation chain is constructed so that the conclusions are already contained in its assumptions. The Poincare/logistic reduction is asserted and cited, not derived: a periodic solution of Eq. (11) does not imply the one-dimensional logistic map (15), and no phase-space reconstruction, return-map computation, or independent parameter estimation is given. The inverse E-Q law is inserted at Eq. (22) as 'Rn=k/Q' and then 'predicted' in Eq. (23); since the Priestley-Taylor relation already makes E proportional to Rn, this is an algebraic restatement of the input. The lagged driver relation Q_{t-1} -> E_{t+1} is likewise obtained by assuming both E and Q obey logistic maps. The observed types I-III are then named as stages of this constructed model, so the model 'explains' phenomena that were placed into its equations. This is partial circularity: the central dynamic prediction reduces to construction, though there is no self-citation chain that would make the score 8-10. Score 7.

Axiom & Free-Parameter Ledger

4 free parameters · 7 axioms · 1 invented entities

The central model rests on (i) the assertion that Epan and cloud quantity obey an NSNDE, (ii) the reduction of a periodic solution to a logistic map, (iii) identification of logistic coefficients with Priestley-Taylor variables, (iv) a monotonic inverse cloud-radiation relation, and (v) a logistic map for cloud quantity. None of these is derived from first principles or fitted against the presented time series; E0 and k are unmeasured constants.

free parameters (4)

NSNDE coefficients a0, a1, A, omega, tau = not estimated
Introduced ad hoc in Eq. (11); never fitted to or validated against the observed Q/Epan series.
alpha and beta in logistic map = not estimated; E0 unspecified
Eqs. (18)-(19) claim alpha = 1.26 R_n s and beta = (E0+s)/(lambda+gamma); the derivation is algebraically unjustified and E0 is an unmeasured maximum-potential-evaporation constant.
k in R_n = k/Q = not estimated
Proportionality constant in Eq. (22) is introduced ad hoc and never estimated; it hardwires the inverse cloud-radiation relation.
B-G test threshold P0 = not specified
The Bernaola-Galvan segmentation depends on P0; the paper never states its value, so change-point sets are not fully reproducible.

axioms (7)

ad hoc to paper The NSNDE (Eq. 11) with sinusoidal delayed forcing is an appropriate model for the coupled Epan-Q system.
The equation is asserted, not derived from physics or fit to data; solution and stages are read off from it.
ad hoc to paper Periodicity of the system permits reduction to a Poincare map and the logistic map (Eqs. 13-15).
The Poincare-map/logistic-map reduction is assumed rather than shown; the 'minimum unit 2*pi/omega' is not defined quantitatively.
domain assumption Priestley-Taylor relation (Eq. 17) is valid for Epan in the Huaihe River Basin.
The Priestley-Taylor equation is an equilibrium evaporation estimate; its direct equality to pan evaporation and use to fix alpha and beta is assumed.
domain assumption R_n decreases as Q increases, so R_n = k/Q (Eq. 22).
Cloud-radiative effect is treated as a monotonic inverse proportionality with constant k; no coefficient is estimated or tested here.
ad hoc to paper Q_t itself evolves by a logistic map (Eq. 24).
No data analysis establishes a logistic map for cloud quantity; it is chosen to make the system closed.
domain assumption Entropy generation/stability can be inferred from signs of dRn/dt and dE/dt in G_g = G_T + G_r + R_n - E and G_A = -G_T - G_r - R_n + E.
The link between these energy-balance derivatives and Lyapunov/entropy stability is asserted, not derived (Section 5c).
standard math Mann-Kendall and Bernaola-Galvan statistics are valid for these series despite possible serial dependence.
Standard nonparametric tests, but no handling of autocorrelation is described.

invented entities (1)

Steamer (Zheng Long) cloud-vapor structure no independent evidence
purpose: Explains the formation/duration/waning/recurring stages of the evaporation paradox by analogy to a bamboo steamer retaining vapor.
No quantitative prediction or measurement of the 'steamer' is provided; it is a narrative device rather than a falsifiable physical entity.

pith-pipeline@v1.3.0-alltime-deepseek · 18557 in / 15292 out tokens · 151361 ms · 2026-08-04T01:29:35.134868+00:00 · methodology

0 comments

Cite this review

Pith. "Pith review of A Dynamic Theory for Explaining the Evaporation Paradox and Global Energy Transpiration." pith.science (2026). https://pith.science/paper/PTLCWNKN

@misc{pith2026260800049,
  author       = {Pith},
  title        = {Pith review of: A Dynamic Theory for Explaining the Evaporation Paradox and Global Energy Transpiration},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/PTLCWNKN}},
  note         = {Machine review of arXiv:2608.00049}
}

Share X Bluesky LinkedIn Reddit HN

read the original abstract

Evaporation is a part of water cycle and a process of energy exchange between atmosphere and land surface, its variation reflects global change. Pan evaporation decreases with global warming is the phenomena named evaporation paradox, which exits in worldwide and spatio-temporal. Broad-scale observations between the 1950s and 2000s revealed that global pan evaporation (Epan) decreases with increasing quantity of clouds. However, in the Huaihe River Basin, both the total cloud quantity and Epan decreased during this period, and similar phenomena were observed in some other regions of the globe. A nonlinear second-order neutral-delay dynamic equation (NSNDE) of the change in the cloud quantity and Epan with time was constructed, encompassing different stages (formation, duration, waning, recurring) of the evaporation paradox. On the basis of this equation, a new model named "steamer" was proposed, encompassing a set of dynamic equations to explore the evaporation paradox. The effects of the total cloud quantity on factors that affect the sensible heat flux are investigated, revealing that actual evaporation (Ea) displays similar oscillation properties as Epan and the total cloud quantity, and their relationship is complimentary in some stages of the evaporation paradox. On the basis of the relation between the total cloud quantity and evaporation, an expression for global energy transpiration was established, and the time delay plays an important role in energy exchange between global spheres. This relation indicates the stability of atmosphere and surface.

Figures

Figures reproduced from arXiv: 2608.00049 by Kejing Liu.

**Figure 1.** Figure 1: (a) 1952—1997 surface observations of total cloud cover from the Extended Edited Cloud Report Archive (EECRA), created by S. Warren and C. Hahn. (b) 1983—2004 total cloud quantity from the ISCCP, created by the W. Rossow group (a) (b) (c) [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗

**Figure 2.** Figure 2: 1983—2001 cloud quantity variation distributions; created by Ding et al. (2004). The positive values indicate where cloud quantity increase and the negative values indicate where it decrease, the unit is %/10 year. On the basis of the temporal and spatial distributions of the cloud quantity and the Epan trend distributions, in the observed period, in most parts of Europe, the USA and Australia, Epan and cl… view at source ↗

**Figure 4.** Figure 4: Annual changes of Epan and Q. On the top of every picture: The light gray dotted line represents the annual Epan observation data, the dark gray dashed line represents the annual Q observation data, and the accompanied black polylines represent the average abrupt changes detected via the B‒G tests, the turning points represents the most obvious abrupt changes. On the bottom of every picture: The black dott… view at source ↗

discussion (0)

Reference graph

Works this paper leans on

4 extracted references · 3 canonical work pages

[1]

Longtin, M

an der Heiden, U., A. Longtin, M. C. Mackey, J. G. Milton, and R. Scholl, 1990: Oscillatory modes in a nonlinear second-order differential equation with delay. Journal Dyn Diff Equat ., 2, 423-449, https://doi.org/10.1007/BF01054042. Arking, A., 1991: The radiative effects of clouds and their impact on climate. Bull. Amer. Meteor. Soc ., 72, 795-813, http...

work page doi:10.1007/bf01054042 1990
[34]

Physics and Control

2001, St Petersburg, Russia, 1549-1554, https://doi.org/10.1016/S1474-6670(17)35410-1. Pykh, Y. A., 2003: Energy lyapunov function for generalized replicator equations. Proc. Int. Conf. “Physics and Control” , Petersburg, Russia, 270-275, https://doi.org/10.1109/PHYCON.2003.1236830. Qian, Y., D. P. Kaiser, R. Leung, and M. Xu, 2006: More frequent cloud-fr...

Pith/arXiv arXiv 2001
[2000]

Geophys. Res. Lett., 33, L01812, https://doi.org/10.1029/2005GL024586. Qian, Y. P., and Y. Y. Huang, 1990: Analysis of affecting factors for soil and ground surface temperatures. Sci. Metrol. Sin., 10, 237-247. Qiu, X. F., C. M. Liu, and Y. Zeng, 2003: Changes of pan evaporation in the recent 40 years over the Yellow River basin. J. Nat. Resour. , 18, 437...

work page doi:10.1029/2005gl024586 1990
[2002]

Int. J. Climatol., 24, 1077-1090, https://doi.org/10.1002/joc.1061. 30 Rossow, W.B., and E. Duenas, 2004: The International Satellite Cloud Climatology Project (ISCCP) web site: An online resource for research. Bull. Amer. Meteorol. Soc., 85, 167- 172, doi:10.1175/BAMS-85-2-167. Senior, C. A., and J. F. B. Mitchell, 1993: Carbon dioxide and climate: The i...

work page doi:10.1002/joc.1061 2004

[1] [1]

Longtin, M

an der Heiden, U., A. Longtin, M. C. Mackey, J. G. Milton, and R. Scholl, 1990: Oscillatory modes in a nonlinear second-order differential equation with delay. Journal Dyn Diff Equat ., 2, 423-449, https://doi.org/10.1007/BF01054042. Arking, A., 1991: The radiative effects of clouds and their impact on climate. Bull. Amer. Meteor. Soc ., 72, 795-813, http...

work page doi:10.1007/bf01054042 1990

[2] [34]

Physics and Control

2001, St Petersburg, Russia, 1549-1554, https://doi.org/10.1016/S1474-6670(17)35410-1. Pykh, Y. A., 2003: Energy lyapunov function for generalized replicator equations. Proc. Int. Conf. “Physics and Control” , Petersburg, Russia, 270-275, https://doi.org/10.1109/PHYCON.2003.1236830. Qian, Y., D. P. Kaiser, R. Leung, and M. Xu, 2006: More frequent cloud-fr...

Pith/arXiv arXiv 2001

[3] [2000]

Geophys. Res. Lett., 33, L01812, https://doi.org/10.1029/2005GL024586. Qian, Y. P., and Y. Y. Huang, 1990: Analysis of affecting factors for soil and ground surface temperatures. Sci. Metrol. Sin., 10, 237-247. Qiu, X. F., C. M. Liu, and Y. Zeng, 2003: Changes of pan evaporation in the recent 40 years over the Yellow River basin. J. Nat. Resour. , 18, 437...

work page doi:10.1029/2005gl024586 1990

[4] [2002]

Int. J. Climatol., 24, 1077-1090, https://doi.org/10.1002/joc.1061. 30 Rossow, W.B., and E. Duenas, 2004: The International Satellite Cloud Climatology Project (ISCCP) web site: An online resource for research. Bull. Amer. Meteorol. Soc., 85, 167- 172, doi:10.1175/BAMS-85-2-167. Senior, C. A., and J. F. B. Mitchell, 1993: Carbon dioxide and climate: The i...

work page doi:10.1002/joc.1061 2004

A Dynamic Theory for Explaining the Evaporation Paradox and Global Energy Transpiration

Core claim §

What carries the argument

Load-bearing premise §

What would settle it §

If this is right

Where Pith is reading between the lines

discussion (0)

Core claim

Load-bearing premise

What would settle it