Water feedback does not follow the theory of the IPCC science

Summary

Positive water feedback has been a cornerstone in any Global Climate Model (GCM). In the AR5, the IPCC writes that “water vapour is a strong and fast feedback that amplifies any initial forcing by a typical factor between two and three". My latest research study shows that, according to the empirical observations, the maximum warming feedback amplifies the original radiative forcing by a factor of 1.14 (14 %). In the AR6, the Climate Sensitivity Parameter (CSP or λ) is 0.47 K/(W/m2), which means water feedback amplification by a factor of 1.8 (80%) since the CSP is 0.265 K/(W/m2)  without water feedback.

Dr. Ollila has calculated the first time the radiative forcing (RF) equation for different water vapour concentrations in the same way as for other greenhouse gases by applying the spectral analysis method. Ollila says that “this is a kind of 'silver bullet' solution, since in this way, the question of the magnitude of water vapour feedback can be solved simply by calculating the RF values of water vapour using the observed global humidity observations.”

A greatly oversized water vapour feedback can be found in the AR6 warming calculations, in which the observed and the  GCM-calculated temperature increase from 1750 to 2019 is practically the same, 1.27 ºC - 1.29 ºC. A closer analysis reveals that the RF value of cloud-aerosol-radiation driver has decreased by -0.12 Wm^-2 from 2011 to 2019, but during the same period, the absorbed solar radiation (ASR) due to decreased cloudiness has increased 1.29 W/m2, which obviously has not been included in this RF driver.  The reason may be that, according to the IPCC science, the warming impact of the ASR increase is 0.6 ºC, which would make the GCM-calculated temperature increase to be about 1.9 ºC – 2.0 ºC.

The water feedback theory according to the IPCC

The IPCC proposed a positive feedback of water as part of computer models in its first assessment report (FIR), but without any physical basis. In the Second Assessment Report (SAR) in 1995, the IPCC justified positive water feedback in this way: "The earliest radiation-convection models (Manabe and Wetherald, 1967) simulating global warming assumed a positive feedback of water vapor. The feedback is generated by the strong dependence of the vapor pressure saturated with water vapor near the surface on temperature, as shown by the Clausius-Clapeyron equation. The increase in temperature is thus expected to lead to an increase in the mixing ratio of atmospheric water vapor."

A critical reader observes that the C-C equation works only in saturated humidity conditions, which do not prevail in the atmosphere. In the Fourth Assessment Report AR4 from 2007, the IPCC had found a more precise numerical value for the positive feedback of water, stating it this way: "However, a well-established law of physics (the Clausius-Clapeyron ratio) dictates that the atmospheric water retention capacity increases by about 7% for every 1°C increase in temperature." 

The radiative forcing of water vapour

The author has calculated the first time the radiative forcing (RF) dependency for the water vapour concentration by varying the water vapour concentration H_TPW from 4 mm to 41mm, and the CO₂ concentration from 330 ppm to 490 ppm. In the LBL (line-by-line) calculations, the RF effects of water vapour are calculated based on the temperature, pressure, and water vapour concentration profiles of different climate zones, which are combined into one average climate atmospheric (AGA) profile. The H_TPW value is a measure of the total water vapour amount in the atmosphere, which is available in atmospheric data sets (NOAA 2025a).

The calculations show that the impact of CO₂ concentration was minimal. The RF effect between the 330 ppm and 490 ppm was only 0.04 W/m2 on the RF value of water vapour. Since this is smaller than the estimated calculation accuracy, this effect was neglected. The RF curve of absolute humidity H_TPW variation from 4 to 41 mm has been depicted in Fig. 1.

Figure 1. The RF dependency RF according to the TPW values for the range from 4 mm to 41 mm. The dotted curve is the fitted curve.

The fitting according to the second-order equation is

RF = -5.3526 + 1.5733 H_TPW - 0.0156  H_TPW ² [Wm^-2].       (1)

The coefficient of determination R² is 0.9959, and the standard error of the fitting is 0.89 W/m2, but for the narrower TPW-range of 20 mm - 30 mm it is only 0.034 W/m2.

Empirical evidence of water feedback

Reliable empirical conclusions about the water feedback can be drawn from the behaviour of the climate since 1979, after the worldwide use of the new humidity semiconductor technology Humicap® of Vaisala.

Figure 2. The temperature trend (MetOffice, 2025) and Total Precipitable Water (NOAA, 2025a) trends according to two humidity measurements from 1980 to 2024. ERA5 stands for the fifth generation of the European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis for the global climate and weather. The NCEP/NCAR reanalysis is a joint project between the National Centers for Environmental Prediction (NCEP) and the National Center for Atmospheric Research (NCAR) in the United States.

These data sets have been depicted in Fig. 2 as yearly and 7-year running mean values. It can be noticed that the long-term value of temperature has increased by about 0.8 ⁰C from 1979 to 1994, but both TPW graph values show a negative trend (a 7-year running mean). These empirical trends of TPW versus temperature conflict with the positive water feedback theory. 

The second empirical evidence can be found in the global seasonal temperature variations. Since the dynamic delays and time constant differ between the hemispheres, the global temperature simulations have been carried out separately for both hemispheres, and the global simulation is the sum of these simulations (Fig. 3). The humidity changes, which should cause the positive water feedback, are fast changes happening essentially at the same speed as the temperature changes with a few days delay

Figure 3. The graphs of the observed (NOAA) and simulated temperature anomalies of the globe. The temperature impacts of ASR and water vapour have been depicted.

In Fig. 3, it can be noticed the fact that the ASR is the dominating climate driver of the Earth. The yearly temperature effect of GH gases according to IPCC science is only about 0.02 °C, and that is why it has not been depicted. The major finding of these simulations is that the temperature effect of water vapour variations is only from 12.8 % to 14.5% in addition to the ASR warming effect. This result is practically the same as that found by Harde (2017), that the water vapour feedback increases the climate sensitivity of the CO₂ impact by about 14 %. According to the positive water feedback theory, it should be about 100 %.

Water vapour warming impacts during the 2000s

The water feedback theory can be tested between 2001 and 2024, when the most accurate observations are available. The positive water feedback theory can also be expressed that any surface temperature increase should include a water vapour impact corresponding to about 50 % of the total change. The temperature and humidity observations have been depicted in Fig. 4 together with major variables. During this short simulation period, the ENSO warming impacts must be included. The warming impact of ENSO originates from the absorbed solar energy, which is released in the El Niño phase, and then during the cooling period of La Niña, this energy is paid back.

Figure 4.  The temperature effects of CO₂, CH₄, and N₂O according to Ollila-2 (green solid curve) and IPCC models (dotted turquoise curve), water vapour (blue solid curve), and ASR+ENSO (brownish curve) have been depicted. The temperature anomaly (red curve) is according to the GISS (2025) data set calculated as a 5-month running mean. The lilac dotted curve illustrates the water vapour feedback effect caused by GH gases according to the C-C theory, which has doubled the original radiative forcings, and it is 50 % of the temperature curve. The warming impacts of ENSO have been calculated by Eq.(5).  All variables have been normalised to zero in the period 2003-2008.

It is easy to notice that the 50 % temperature anomaly (dotted lilac curve) does not vary according to the temperature effect of GH gases as implied by the positive water feedback theory by the IPCC. It should be noticed that according to AR6, CO₂ corresponds to about 80 % of the temperature increase from 1750 to 2019 (IPCC 2021). By judging with the eye, the ASR & ENSO effect has had the dominant role in the temperature increase after the year 2014. 

Water feedback dependency on primary energy

One of the hypotheses in this study has been that water feedback impact depends on the primary energy changes, and this was tested during the period from 2010 to 2025. The most important energy input is the ASR, which has increased by 2.01 W/m2 from 2000 to the year 2023, which can be compared to the RF impact of 2.16 W/m2 by CO₂ from 1750 to 2019 (IPCC 2021).  The temperature effects of ASR and absolute humidity H_TPW have been illustrated in Fig. 5.

Figure 5: The trend curves of UAH temperature, the temperature simulations of the Ollila-2 model, and the temperature effects of  TPW absolute humidity, ASR & ENSO, and GH gases by the Ollila-2 model from 2011 to 2025.

The H_TPW curve seems to correlate quite well with the ASR&ENSO curve. By judging with the eye, the H_TPW changes do not correlate with the impacts of GH gases. The multicorrelation coefficient of regression R² for the period 2005-2024 is 0.756 between the water vapour (H_TPW) and two variables, which are the temperature impacts of ASR&ENSO and the GH gases, according to the Ollila-2 model. The coefficient R² of the model with only ASR&ENSO is only slightly smaller, 0.688. It means that the H_TPW values depend mainly on ASR and ENSO, which are the primary energy inputs.

Temperature simulations of the 2000s

The temperature simulations during the 2000s have been carried out by applying the Ollila-2 model and the IPCC simple model (Fig. 6).

Figure 6. The graphs of the Ollila-2 model and GISS temperature from 2001 to 2025. The warming impacts of the three major climate drivers of ASR, GHGs, and TPW have been depicted for the same period according to the Ollila-2 model, as well as the ENSO according to the empirical equation. The dashed black curve is the simulated temperature response applying the λ-value of 0.47 K/(W/m2) according to the IPCC (2021).

The overall response of the Ollila-2 model is very good in comparison to observed temperature changes. By comparing the water vapour trend changes to the ENSO temperature changes, it is obvious that the major part of the ENSO effect happens through the changes in the atmospheric humidity and ASR changes. It can be noticed that the warming impacts of GH gases are very low. The ASR flux changes have had a major role in the temperature increase after the very strong El Niño in 2015-2016. The H_TPW values have stayed at a record level after the El Niño of 2023-2024, and it seems to be the main reason, besides the ASR, for the very high temperatures of 2023 and 2024.

It should be noticed that there is no water vapour feedback in the Ollila-2 model simulations, but only the warming impacts of greenhouse gases, water vapour, and absorbed solar radiation. In the IPCC model simulations, water vapour feedback has been applied according to the climate sensitivity parameter of 0.47 K/(W/m2) in the AR6.

Validation

The magnitude of positive water feedback of 12.8% to 14.5% obtained in my study is empirical in nature, and in that sense, it does not need validation.

References

1.      Ollila A (2025). Radiative forcing of water vapour and its use in climate models. SCC, 5.3, pp. 186-206. https://scienceofclimatechange.org/wp-content/uploads/SCC-Vol.5.3-Ollila.pdf