-
Notifications
You must be signed in to change notification settings - Fork 5
paper notes: water cycle
Anthropogenic aerosols, greenhouse gases and historical changes in the global freshwater budget.
Ocean freshwater budget changes under different external forcing...
- General characteristics of the water cycle
- The basic "wet get wetter, dry get drier" P-E response in a warming world
- Deviations from WGWDGD (on a large scale - not interested in the myriad of small scale differences)
Moisture from the evaporation-dominated sub-tropical oceans is transported north and south. There's also a net transport from the oceans (mainly Atlantic) to the land.
Atmospheric moisture content increases at 7% per degC: The Clausius-Clapeyron equation is a strong constraint on atmospheric water vapour which is virtually certain to increase on average globally with continued warming. It is very likely that observed near surface specific humidity has increased since the 1970s (Section 2.3.1) and increases of about 7% per degC are expected for constant relative humidity... [AR6, Chapter 8]
Everything else being equal, this means a "wet get wetter, dry get drier" response:
- The rich-get-richer hypothesis (Held and Soden 2006; Chou et al. 2009), associated with thermodynamic increases in moisture transport, would lead to wet regions becoming wetter and dry regions becoming drier in the absence of compensating circulation changes.
- As Held and Soden (2006) point out, a warming atmosphere will cause an increase in atmospheric water vapor. Hence, even if the circulation were to remain fixed, it would be expected that the transports of water vapor would intensify. Consequently, under these assumptions, the pattern of P − E will remain the same but the values will become more extreme, making wet regions wetter and dry regions drier.
This amplification of the P–E pattern is corroborated by an observed “fresh get fresher, salty get saltier” salinity response to warming.
Ma et al, 2018: On very large averaged spatial scales, regional precipitation change has been interpreted as a wet-get-wetter, dry-get-drier pattern, with rainfall increases at the core of existing rainy regions and decreases in current dry areas (Held & Soden 2006) and at the convective margins (Neelin et al. 2003, Chou & Neelin 2004, Chou et al. 2009). However, the dry-get-drier argument has been questioned (Scheff & Frierson 2012), because reduced precipitation appears along the outer flanks of the subtropics due to the poleward expansion of the subtropical dry zones. Nor does the wet-get-wetter interpretation hold at the smaller country-level scales relevant to climate change impacts (Chadwick et al. 2013, Greve et al. 2014, Roderick et al. 2014), where rainfall changes are strongly associated with circulation change. At regional scales, the dominant changes are often shifts in the positions of convective regions, with associated changes in rainfall and water availability. These shifts are associated with a number of mechanisms that differ between land and ocean.
Siler et al, 2019 explain things from an evaporation perspective...
Contributions to changes in P−E can be separated into those caused thermodynamically by changes in specific humidity, dynamically by changes in circulation, and by changes in moisture transports by transient eddies. The thermodynamic and dynamic contributions can be further separated into advective and divergent components (Seager et al, 2010).
Thermodynamic changes:
- Proportionally larger evaporation increases over the subtropical oceans relative to the equatorial convergence zones. The suppressed evaporation increases over low latitudes (∼1%/°C) are partly explained by rapid adjustments to CO2 increases and uptake of heat by the ocean compared with high latitudes. (Siler et al, 2019)
- At higher latitudes, evaporation is further increased by the expansion of open water area as sea and lake ice melts with warming. Allan et al (2020)
Dynamic changes:
- The subtropical dry zones expand poleward. (Seager2010)
- From Hegerl et al (2013): General circulation models (GCMs) generally simulate an expansion of the Hadley cells as the globe warms, with associated poleward migration of subtropical aridity and storm tracks, but the size varies and there is limited agreement on the mechanisms (Yin 2005; Lu et al. 2007; Seidel et al. 2008; Scheff and Frierson 2012a,b).
- Ma et al, 2018: Increases in greenhouse gas (GHG) concentrations warm the troposphere and cool the stratosphere, resulting in a rise of the tropopause. This allows the circulating air of the Hadley cell to travel further, as tropopause height was shown to be important for its poleward limit. The connection between tropopause height and Hadley cell can be explained in the context of the thermal wind relation. An upward extension of the troposphere reduces vertical wind shear since the momentum conservation is hardly influenced. As a consequence of the thermal wind balance, the meridional gradient of potential temperature is reduced, extending poleward the latitude at which the potential temperature matches the equilibrium temperature. This is consistent with the slowdown of the tropical circulation associated with a flattened temperature gradient suggested by MASC (Ma et al. 2012).
- Weakening of the tropical divergent circulation partially opposes the thermodynamic contribution by creating a tendency to decreased P − E in the ITCZ and to increased P − E in the descending branches of the Walker and Hadley cells.
- The changing mean circulation also causes decreased P − E on the poleward flanks of the subtropics because the descending branch of the Hadley Cell expands and the midlatitude meridional circulation cell shifts poleward.
- The tropical mean increase in precipitation is due to the residual between the positive thermodynamic change due to increased specific humidity and the decreased convective mass flux due to the weakening of the circulation (Chadwick et al, 2013)
Transient eddies:
- Subtropical drying and poleward moistening are also contributed to by an increase in poleward moisture transport by transient eddies (Seager et al, 2010)
Land surface:
- Paper on why WGWDGD doesn't hold over land: Byrne & O’Gorman, 2015 (doesn't mention AAs)
- Over land, evaporation is regulated by energy fluxes over wet regions, with atmospheric vapor pressure and aerodynamics playing an important role, but for drier regions evaporation is limited by surface water availability.31, 103 Changes in P–E over drier continental regions are consequently dominated by precipitation changes31 that are strongly determined by alteration in atmospheric circulation.
- While increased moisture transport into wet parts of the atmospheric circulation will amplify P–E patterns globally, the interactions of geography, atmospheric circulation, human activities, and feedbacks involving vegetation and soil moisture lead to a complex regional response over land. However, multiple lines of evidence indicate that the contrast between wet and dry meteorological regimes, seasons, and events will amplify as moisture fluxes increase in a warming climate.
Many of the WGWDGD studies focus on idealised CO2 experiments or future simulations with strong GHG forcing relative to AAs.
We know aerosols have a big impact on precipitation in relation to the global total (e.g. Kvalevåg et al, 2013, Samset et al 2018) and are responsible for a southward shift of the ITCZ and drying of the East Asian monsoon (Wang et al 2016), so it would be useful to put those impacts in the WGWDGD context.
From Hegerl et al (2013): Modeling studies also suggest that aerosols may have contributed to the drying of the Sahel from 1940 to 1980 (Rotstayn and Lohmann 2002; Ackerley et al. 2011; Hwang et al. 2013; Dong et al. 2014) and influence the East Asian monsoon (e.g., Lau et al. 2006; Meehl et al. 2008; Guo et al. 2012; Polson et al. 2014) and midlatitude precipitation (Leibensperger et al. 2012; Rotstayn et al. 2012).
Studies of changes in the global energy budget have done this by looking at single forcing experiments. We'll do that here, using a simplified model of the global water cycle to compare and contrast the separate influence of GHGs and AAs on historical changes in the global water cycle.
Many of the studies discussed above look at models because the observations aren't great (Yu et al, 2020):
- P and E are notoriously difficult to observe.
- E cannot be directly remotely sensed as it cannot be detected by electromagnetic radiation. Estimation of E has large uncertainty because it is computed from bulk parameterizations using near-surface meteorological variables and SST as input.
- P has long been inferred from visible infrared and passive microwave (PMW) retrievals. The launch of the Global Precipitation Measurement mission in 2014 has significantly improved global observations of P.
Hegerl et al 2013: "Challenges in Quantifying Changes in the Global Water Cycle"
Observations (especially of evaporation) are spatially incomplete and don't go back far in time, which is why much of what we know about forced changes to the global freshwater budget are derived from state-of-the-art climate models. Single forcing model experiments, such as those conducted as part of CMIP5 \citep{Taylor2012}, are also required to understand the separate influence of GHGs and AAs. Studies in this area...
Explanation of how to calculate hydrological sensitivity: Flaschner et al (2016)
Hydrological sensitivity in CMIP6 models: Pendergrass (2020)
Observed, CMIP5 and CMIP3 trends in global precipitation: Ren et al, 2013
Global mean precipitation (and evaporation) increase at a lower rate (which implies an increased residence time for atmospheric water vapour).
In contrast to the thermodynamic constraints controlling water vapour changes, global mean precipitation and evaporation are tightly linked to the atmospheric energy budget (Allen and Ingram, 2002; O’Gorman et al., 2012; Pendergrass and Hartmann, 2014b). Latent heat released through precipitation is balanced by the net atmospheric longwave radiative cooling minus the heating from absorbed sunlight and sensible heat flux from the surface (Figure 8.3). Complementary energetic arguments also apply to surface evaporation (Roderick et al., 2014; Siler et al., 2018b).
In a warming climate, the primary mechanism responsible for a global mean increase in precipitation is the so-called "slow" temperature-driven response driven by the increased radiative cooling rate of a warming atmosphere. Physical arguments and idealised GHG forcing experiments (Fläschner et al., 2016; Pendergrass, submitted; Samset et al., 2017) suggest that under this mechanism global mean precipitation increases at 2.1-3.1 % per degC of global mean warming, which is referred to as the planetary hydrological sensitivity.
The actual global mean rate of precipitation/evaporation change per degC of surface warming, referred to as the apparent hydrological sensitivity, is reduced compared to hydrological sensitivity due to a "fast" response caused by near-instantaneous changes in the atmospheric energy budget and atmospheric properties (e.g. temperature, clouds and water vapour) in direct response to the radiative effects of a forcing agent (Sherwood et al., 2015). These rapid atmospheric adjustments to precipitation are primarily caused by GHGs and absorbing aerosols.
The fast and slow responses to GHG forcing oppose one another, while for AAs they reinforce, meaning that the overall influence of AAs on global mean precipitation (per unit forcing) is higher than for GHGs.
The global mean precipitation response to warming was estimated as 1-3% per degC in the AR5 (Collins et al., 2013b).
We haven't seen a detectable increase in global mean precipitation over the historical period: Cooling effects of anthropogenic aerosol and rapid adjustments to increases in GHGs and absorbing aerosols reduce global precipitation, offsetting increases relating to a warming climate. This explains why multi-decadal trends in global precipitation responses in the satellite era are expected to be small and difficult to confirm due to observational uncertainty. [AR6, Chapter 8]
Wu et al 2013: Analysing CMIP5 historical simulations, we find for the first time that there was a detectable weakening of the hydrological cycle (i.e. global mean precipitation) between the 1950s and the 1980s, attributable to increased anthropogenic aerosols, after which the hydrological cycle recovered as a result of increasing greenhouse gas concentrations. The net result of these two counter-acting effects is an insignificant trend in the global hydrological cycle, but the individual influence of each is substantial.
Big reviews:
- Advances in understanding large‐scale responses of the water cycle to climate change (Allan et al, 2020) focuses on the non-ocean components of the water cycle.
- Intensification of the global water cycle and evidence from ocean salinity: a synthesis review (Yu et al 2020) focuses on the ocean component of the water cycle.
- Responses of the Tropical Atmospheric Circulation to Climate Change and Connection to the Hydrological Cycle (Ma et al 2018)
Huge list of water cycle papers
The latest paper on observed global salinity trends and how the CMIP models do at reproducing the trends and climatology is Cheng et al (2020).
Hydrological sensitivity
The Clausius-Clapeyron equation is a strong constraint on atmospheric water vapour which is virtually certain to increase on average globally with continued warming. It is very likely that observed near surface specific humidity has increased since the 1970s (Section 2.3.1) and increases of about 7% per degC are expected for constant relative humidity...
In contrast to the thermodynamic constraints controlling water vapour changes, global mean precipitation and evaporation are tightly linked to the atmospheric energy budget (Allen and Ingram, 2002; O’Gorman et al., 2012; Pendergrass and Hartmann, 2014b). Latent heat released through precipitation is balanced by the net atmospheric longwave radiative cooling minus the heating from absorbed sunlight and sensible heat flux from the surface (Figure 8.3). Complementary energetic arguments also apply to surface evaporation (Roderick et al., 2014; Siler et al., 2018b).
Global mean precipitation response to warming was estimated as 1-3% per degC in the AR5 (Collins et al., 2013b), which includes fast adjustments that scale with atmospheric forcing, and slow temperature-driven responses to radiative forcings (Andrews et al., 2010; Bala et al., 2010; Cao et al., 2012). The fast response is caused by near-instantaneous changes in the atmospheric energy budget and atmospheric properties (e.g. temperature, clouds and water vapour) in direct response to the radiative effects of a forcing agent (Sherwood et al., 2015). A further relatively fast response involves the land-surface temperature which responds more rapidly to radiative forcing than the ocean (Cao et al., 2012; Dong et al., 2014), operates mostly at the regional scale (Chadwick et al., 2019) and depends on the partitioning of the increased net surface radiation between latent and sensible heat and, thereby, on the land hydrology (Berg et al., 2016). The slower temperature-dependent precipitation response (hydrological sensitivity) is driven by the increased atmospheric radiative cooling rate of a warming atmosphere. Since the AR5, the dual fast adjustment and slow response framework has been applied across a range of global climate models, and the global precipitation responses to different forcing agents are physically well understood (Fläschner et al., 2016; MacIntosh et al., 2016; Samset et al., 2016; Myhre et al., 2018a).
Removing fast adjustment effects, there is high confidence based on robust physics and idealised CO2 forcing experiments (Fläschner et al., 2016; Pendergrass, submitted; Samset et al., 2017) that global mean precipitation increases at 2.1-3.1 % per degC of global mean warming, termed hydrological sensitivity (η) (Figure 8.3)... Since physical constraints cause global mean precipitation to increase more slowly than water vapour content per degree of warming, this implies an increased residence time of atmospheric water vapour (Hodnebrog et al., 2019).
The actual global mean rate of precipitation change per degC of surface warming, apparent hydrological sensitivity (ηa), is reduced compared to hydrological sensitivity by the direct influence of radiative forcing agents on the atmospheric energy balance. Rapid atmospheric adjustments to precipitation are primarily caused by GHGs and absorbing aerosols. The precise magnitude of the response depends on forcing type and characteristics, and is understood with medium confidence based on idealised simulations (Fläschner et al., 2016; Samset et al., 2016). A range of fast precipitation adjustments to CO2 between models are attributed to the response of vegetation leading to a repartitioning of surface latent and sensible heat fluxes (DeAngelis et al., 2016). The range obtained from 6 CMIP5 models simulating the last glacial maximum and pre-industrial period (ηa=1.6-3.0 % per oC) is larger than for corresponding 4xCO2 experiment (ηa=1.3–2.6 % per oC) in which larger CO2 forcing suppresses precipitation response due to fast adjustments (Li et al., 2013b). Also contributing to the difference are vegetation and land surface changes and because evaporation is more sensitive to temperature in the colder state due to thermodynamic constraints.
Cooling effects of anthropogenic aerosol and rapid adjustments to increases in GHGs and absorbing aerosols reduce global precipitation, offsetting increases relating to a warming climate. This explains why multi-decadal trends in global precipitation responses in the satellite era are expected to be small and difficult to confirm due to observational uncertainty. The warming influence of continued rises in CO2 concentration, combined with declining aerosol cooling, are expected to increase the importance of the slow temperature-related effects on the energy budget relative to the more rapid direct radiative forcing effects as transient climate change progresses (Shine et al., 2015; Salzmann, 2016; Myhre et al., 2018b).
In summary, it is virtually certain that global mean precipitation and evaporation will increase in response to future warming. Global changes observed in recent decades have been suppressed by rapid adjustments to radiative forcings which are understood with medium confidence based upon a range of responses in climate models. It is likely that increases in multi-annual mean precipitation over land will be smaller than over the ocean due to declining near-surface relative humidity driven by increasing land-ocean thermal contrast and surface feedbacks. There is high confidence that global mean precipitation and evaporation will increase at a slower rate than atmospheric water vapour per oC of global mean warming, driving an increase in water vapour lifetime and changes in precipitation intensity, duration and frequency.
Moisture transport
The discussion on atmospheric moisture transport in AR6 tends to focus on ocean to land transport as opposed to meridional transport.
How aerosols influence precipitation
BOX 8.1: Role of anthopogenic aerosols in water cycle changes
Aerosols affect precipitation in two major pathways, by affecting the shortwave and longwave radiation and cloud microphysical properties.
Aerosol radiative effects on precipitation
Aerosols scatter and absorb solar radiation which reduces the energy available for surface evaporation and subsequent precipitation. In addition, cooling is incurred by the radiation that is reflected back to space directly by the aerosols and indirectly by the aerosol effect on cloud brightening. Northern Hemisphere (NH) station data indicate decreasing precipitation trends during 1950s-1980s which have since partially recovered. These changes are attributable with high confidence to anthropogenic aerosol emissions from North America and Europe causing dimming through reduced surface solar radiation, which peaked during the late-1970s and partially recovered thereafter following improved air quality regulations (Bonfils et al., submitted; Wild, 2012). The dimming over the NH causes a relative cooling there compared to the SH, which induces a southward shift of the northern edge of the tropical rain belt (Allen et al., 2014; Brönnimann et al., 2015a)...
Aerosol cloud microphysical effects
Cloud droplets nucleate on pre-existing aerosols particles which act as cloud condensation nuclei (CCN). Anthropogenic aerosols add CCN that produce clouds with more numerous and smaller droplets, slower to coalesce into raindrops and to freeze into ice hydrometeors at temperatures below 0°C. Adding CCN suppresses light rainfall from shallow and short-lived clouds, but it is compensated by heavier rainfall from deep clouds. Adding aerosols to clouds in extremely clean air invigorates them by more efficient vapour condensation on the added drop surfaces (Koren et al., 2014; Fan et al., 2018a). Clouds forming in more polluted air masses (hence with more numerous and smaller drops) need to grow deeper to initiate rain (Freud and Rosenfeld, 2012; Konwar et al., 2012; Braga et al., 2017). Delaying rain initiation to greater heights transports more cloud water above the 0°C altitude and leads to invigoration of mixed phase precipitation and the resultant hail and cloud electrification (Rosenfeld et al., 2008a; Thornton et al., 2017). This includes the outer convective rainbands of tropical cyclones. There is a medium confidence that air polution enhances the flooding from them at the expense of the inner rain bands (Zhao et 1 al., 2018a). More generally, the microphysical aerosol-related processes often compensate or buffer each other (Stevens and Feingold, 2009). For example, suppressed warm rain enhances mixed phase precipitation. Therefore, despite the potentially large aerosol impacts on the precipitation forming processes, the net outcome of aerosol microphysical effects on precipitation amount has generally low confidence, especially when evaluated with respect to the background of high natural variability in precipitation (Tao et al., 2012).
The strength of the hydrological cycle is set by the net available energy reaching Earth's surface (refs). These papers therefore examine the global mean surface energy budget,
R_Lnet + R_Snet = Q_L + Q_S + G
where (as defined in Wu2010) where R_{Lnet} is the net surface long wave radiation, R_{Snet} the net surface short wave radiation, Q_S the sensible heat flux, Q_L the latent heat flux and G the ocean heat uptake.
Changes in the global hydrological cycle are driven by perturbations to the surface energy budget. At timescales longer than those associated with radiative convective equilibrium (about a month) global mean precipitation must be balanced by evaporation:
L \DeltaP (or \DeltaE) = \DeltaR_{Lnet} + \DeltaR_{Snet} - \DeltaQ_s + \DeltaR_{TOA} (1)
where L denotes the latent heat constant and \DeltaP (or \DeltaE) the change in global mean precipitation (evaporation). They approximate G (the ocean heat uptake) by the TOA radiative imbalance (\DeltaR_{TOA}) as the atmospheric heat capacity (and land heat uptake) is negligible.
Personal note: On a sub-global scale P and E don't have to both have the same change, so the above relationship works for E (which is driven my the local surface energy budget) but not P.
Figure 3a shows the evolution of the above terms in the ALL simulations. A reduced air-sea temperature difference leads to a small decrease in the surface sensible heat flux. Changes in the ocean heat uptake are small. The energy budget is dominated by the two radiative terms. The increase in the net downward long wave radiation at the surface due to GHG emissions is cancelled out by the decrease in the net downward short wave radiation due to aerosol emissions (total cloud has also been decreasing), which results in a surface energy deficit since the mid-twentieth century and a weakening of the hydrological cycle. The long-term decrease in the short wave radiation at the surface is driven by anthropogenic influences, whereas the negative spikes are linked to volcanic eruptions (Fig. 3b). The decrease in the net short wave radiation at the surface is similar to the TOA increase in the outgoing short wave radiation (Fig. 3c).
The water cycle is made up of all transports and fluxes of freshwater, and therefore “intensification” of the water cycle is generally used to describe, in addition to globally increased atmospheric moisture (W), an increase in total precipitation and evaporation, amplification of the existing E − P pattern, and enhanced atmospheric vapor transport Q.
Several caveats should be noted regarding the representation of the water cycle in GCMs. First, although model resolution has improved in recent generations, current grid sizes still require significant topographical smoothing. Therefore, steep mountain ranges such as the Andes are represented with reduced peak heights, and narrow oceanic throughflows like the Bering Strait and Strait of Gibraltar have greatly simplified geometry. A modeling study by Schmittner et al. (2011) demonstrated that alterations to global topography can substantially change water vapor fluxes, ocean circulation, and the climate system as a whole. The highest South American peak in models presented here ranges from 3916 to 4591 m, while the Andes reach an actual height of 6962 m. This discrepancy may result in excess moisture transport across ranges such as the Andes, as discussed by Richter and Xie (2010). To combat these problems, modelers sometimes enhance orography at the expense of increasing the mean height of continents, resulting in a dry bias of atmospheric vapor content over land (Gaffen et al. 1997). In general, however, GCMs display a slightly overactive water cycle, with greater total precipitation and greater rainfall frequency than observations (Stephens et al. 2010; Tian et al. 2013; Demory et al. 2014). Nearly all current GCMs have well-documented issues with simulating tropical precipitation dynamics, particularly in the Pacific Ocean (Lin 2007; Pincus et al. 2008; Collins et al. 2011; Brown et al. 2013; Flato et al. 2013; Li and Xie 2014). A set of connected biases combine to produce an excessive cold tongue in the eastern Pacific, overly strong low-level trade winds, and an unrealistic double ITCZ flanking the equatorial Pacific (Lin 2007; Hwang and Frierson 2013).
General precipitation problems in models:
- too light (constant drizzle)
- not heavy enough
- too much polar snow
If you increase the aerosol size to fix the drizzle problem the aerosol effect is so large that the models don't really warm over the historical period.
Our analysis (of tropical cyclone potential intensity, PI) is closely related to recent studies of the global hydrological cycle. Greenhouse gas warming accelerates Earth’s hydrologic cycle — increasing the rates of surface evaporation and precipitation — and aerosol cooling decelerates it. As in the case of PI, aerosols are about 2–3 times as effective in changing the hydrologic cycle per degree surface temperature change than are greenhouse gases (e.g., Feichter and Roeckner 2004; Liepert and Previdi 2009); this is relevant, for example, to proposed solar radiation management schemes for “geoengineering” (e.g., Bala et al. 2008). Some understanding of this difference has been gained by separating changes in the global energy budget into “fast” or “temperature-independent” and “slow” or “temperature-dependent” components (e.g., Andrews et al. 2009; O’Gorman et al. 2012; Samset et al. 2016). The temperature-independent radiative effect of a given forcing agent at the top of the atmosphere (TOA), or at the surface, is the change in the TOA or surface radiative flux that would occur in the absence of any changes in the global mean surface temperature. In practice, the temperature-independent effect is often estimated as the change that occurs at the very beginning of a simulation in which the radiative forcing agent is switched on abruptly [e.g., using a “Gregory-type” approach (Gregory et al. 2004)], or by running a simulation in which the forcing agent is introduced and SSTs are held fixed. The temperature-dependent effect can be estimated as the change in radiative flux at equilibrium (or some other intermediate state in which there has been a finite temperature change) minus the temperature-independent effect. The temperature-dependent effect depends not only on surface temperature, but also on state variables related to it such as atmospheric temperature and water vapor. These influence TOA and surface radiation through feedbacks that have been extensively defined and documented in the literature, such as the water vapor feedback and lapse rate feedback. Studies with single forcings (e.g., Andrews et al. 2009; Previdi 2010; O’Gorman et al. 2012) show that these temperature-dependent feedbacks are similar for different radiative forcings. The temperature-independent effects of shortwave and longwave forcings, on the other hand, are different, and these differences lead to the differences in the hydrologic cycle response.
The different effects of shortwave and longwave forcings on the global hydrologic cycle can be understood either from the point of view of the tropospheric heat budget or the surface energy budget. In the global mean, over any time scale of interest for climate studies, the tropospheric heat budget requires that the vertically integrated radiative cooling of the atmosphere be balanced by the sum of latent heating due to water condensation and surface sensible heat flux. To the extent that sensible heat flux is small, then, the radiative cooling closely constrains precipitation (Allen and Ingram 2002).
The surface energy budget, on the other hand, requires that the sum of surface latent and sensible heat fluxes balance net surface radiation, after accounting for ocean heat transport and storage. In the global mean, the surface energy budget constrains precipitation as well, since precipitation and surface evaporation must balance on climatically relevant time scales. The relationship between the surface turbulent fluxes and potential intensity, on the other hand, is local. While our analysis bears considerable similarity to those in the hydrologic cycle literature, it differs in our focus on the tropics, and in particular on individual hemispheres of the tropics during the seasons in which TCs are most active. This local focus is facilitated by our use of the surface energy budget rather than the global heat budget. Particularly when considering only a single hemisphere and season at a time within the tropics as we do, we anticipate that the energy transports due to the Hadley cell and extratropical eddies would make it challenging to interpret PI in terms of the atmospheric heat budget. The surface energy balance, though also not without complications (in this case due to ocean heat transport, as will be seen below), appears to us more straightforward. In any case, the local analysis yields results that are quantitatively, and in some respects even qualitatively different from those in the global mean. To illustrate this, we conclude our study with a direct comparison of tropical seasonal results with global annual mean results obtained using an analysis approach that is otherwise—that is, in all ways besides the averaging domain—identical to that used for the tropical results.
Me: Not sure if there are any papers looking at interhemispheric differences in water cycle response??? Given the high hydrological sensitivity to aerosols, you'd think there would be pronounced hemispheric differences...
The horizontal atmospheric water vapor flux will intensify under projected climate change; the high-latitude (Arctic) exhibits the largest percentage increases, especially during DJF (and in general integrated vapor transport increases more as you move towards the poles in either hemisphere).
Bonfils2020: Global precipitation increases as a slow response to surface warming (Allen2002), with regional spatial features that are governed by the intensification and poleward shift of current wet–dry patterns (Marvel2013, Held2006). This GHG-induced warming and precipitation increase is partly offset by AA-induced cooling (Polson2014), which produces a short-lived but larger change in global mean precipitation per degree of temperature change than GHG forcing (Wu2013,Allen2002). The uneven spatial distribution of AA emissions also yields more cooling in the Northern Hemisphere (NH) than in the Southern Hemisphere (SH). This hemispheric asymmetry induces an observed change in cross-equatorial wind19 and ocean temperature12 and a southward intertropical convergence zone (ITCZ) shift7,8,20–23, which in turn impacts rainfall in the Sahel11 and other monsoon regions15,16,24,25. The AA-driven surface cooling is due to the direct scattering of incoming solar radiation by sulfate aerosols and to aerosol indirect effects (AIEs). The latter arise from changes in the number, size and optical properties of liquid water droplets, which affect cloud albedo and lifetime20.
Wang et al (2020) look at the GHG-only and AA-only experiments in CMIP5 band CMIP6. They find that the SH jet moves poleward in the GHG-only experiment. It weakens in the AA-only experiment but stays in the same location.
Venkatachalam Ramaswamy is an authority on this topic
- He mentioned that the aerosol effect on global rainfall rivals that of GHGs, unlike surface temperature (where GHGs have a bigger influence)
- README: Persad et al 2018 looks at why GHG and AA forcing are so similar
Allen2019 talks about the fact that the total aerosol burden actually go up in GHG runs. Aerosol removal from the atmosphere depends on rainfall frequency (which is decreasing), not intensity (which is increasing). Allen (at AMS) estimated that it might offset ~10% of the effect of GHGs (i.e. non-trivial but not huge).
\cite{Levang2015} provides a detailed analysis of water cycle change under the RCP8.5 experiment.
- They make sure the atmospheric budget doesn't drift (by only using models cleared by Liepert2013) but don't consider ocean drfit at all (That's possibly fine - I should check - for such strong forcing but wouldn't be for the historical period.)
- They use 6 hourly atmospheric data to calculate the atmospheric transports
- (The computational expense restricted the size of their ensemble.)
- Atmospheric vapor transport Q can be separated into mean and eddy components via Reynolds decomposition (i.e. Qbar and Qdash) With respect to zonal transport, Qdash is small in the tropics but is pretty much equal in size to Qbar in the mid latitudes. To fully capture the meridional transport then, the data used for this calculation must be sampled at higher resolution than typical time scales of synoptic weather systems. In this study, 6-hourly model outputs have been used.
- They use the wfo data to infer the ocean transports (i.e. they don't use the vo or uo data) for climatologies of different timeseries (i.e. assume no change in so / ocean freshwater redistribution)
- They use a bunch of different obs/reanalysis products to get an observational estimate for P, E and R (and from that infer ocean transport)
- "Integrated over each basin, the Atlantic becomes 0.20 Sv more evaporative, the Pacific 0.16 Sv more precipitative, the Indian Ocean 0.08 Sv more evaporative, and the Arctic Ocean 0.05 Sv more precipitative."
The models listed below have relevant data available for the piControl, historical, GHG-only and AA-only experiments.
- ✔️ - has been processed
- ✅ - might be included in the ensemble (the data exists) but not yet processed
- 🚫 - there's something (prohibitively) wrong with the data
- ❌ - data doesn't exist
| Model | P-E |
wfo & so
|
problems |
|---|---|---|---|
| CCSM4 | ✔️ | ❌ | |
| CSIRO-Mk3-6-0 | ✔️ | ✔️ | |
| CanESM2 | ✔️ | ❌ | |
| FGOALS-g2 | ✔️ | ❌ | |
| GFDL-CM3 | ✔️ | ✔️ | |
| GFDL-ESM2M | ✔️ | ✅ | download historicalMisc so; check budget |
| GISS-E2-H | ✔️ | ❌ | |
| GISS-E2-R | ✔️ | 🚫 | budget doesn't close |
| IPSL-CM5A-LR | ✔️ | 🚫 | budget doesn't close |
| NorESM1-M | ✔️ | ✅ | download historicalMisc wfo and so
|
| Total ✔️ | 10 | 2 | |
| Total ✅ | 0 | 2 |
| Model | P-E |
wfo & so
|
problems |
|---|---|---|---|
| ACCESS-ESM1-5 | ✔️ | ✔️ | P-E: Done all three runs |
| BCC-CSM2-MR | ✔️ | ❌ | P-E: r1i1p1f1 done; need to download evspsbl for r2i1p1f1 and r3i1p1f1 no wfo data |
| CanESM5 | ✔️ | ✔️ | P-E: r1i1p1f1 done, r[1-10]i[1-2]p1f1 available wfo/so: budget doesn't close on short timescales but does on longer? |
| CESM2 | ✔️ | ❌ | P-E: r1i1p1f1 done; need to download pr and evspsbl for r2i1p1f1 and r3i1p1f1no wfo data |
| CNRM-CM6-1 | ❌ | ✔️ | budget closes |
| FGOALS-g3 | ✔️ | ❌ |
thetao missing for historical. Also areacello and volcello mask issues. |
| GFDL-ESM4 | ✔️ | ❌ | no wfo data |
| GISS-E2-1-G | ❌ | 🚫 | budget doesn't close |
| HadGEM3-GC31-LL | ❌ | ✔️ | budget closes |
| IPSL-CM6A-LR | ✔️ | ✔️ | budget closes |
| MIROC6 | ✔️ | ❌ | no wfo data for the hist-GHG experiment |
| MRI-ESM2-0 | ❌ | 🚫 | budget closes; ocean volume problem with gn and gr |
| NorESM2-LM | ❌ | 🚫 | budget closes; wfo on different grid to thetao/so and there's no tos/sos
|
| Total ✔️ | 8 | 5 | |
| Total ✅ | 0 | 0 |
Issues to resolve:
- Crazy salinity values dotted throughout timeseries
This variable represents the net freshwater flux at the ocean surface. In other words, it is ("probably" - direct quote from CMIP5 docs) the sum of:
- pr: rainfall flux
- prsn: snowfall flux
- evs: water evaporation flux
- friver: water flux into seawater from rivers
- ficeberg: water flux into seawater from ice bergs
- fsitherm: water flux into sea water due to sea ice thermodynamics
All of these are second priority variables (as is wfo) except for fsitherm (first priority), which means if wfo isn't archived there's not much chance all the components are. Also, most models that do archive wfo don't archive all the components, but you could probably get a good picture just from pr, evs and friver.
Precipitation and evaporation timeseries are very noisy. Better to look at cumulative anomaly (which is something you can do with models - due to control experiment - but not observations).
What is useful/unique about considering the sum of all the E > P cells and P > E cells as opposed to other approaches?
A key advance from my excess energy paper is that by using the netTOA, ocean surface heat flux and OHC data, you can infer anomalous heat transports rather than having to calculate them directly (which is computationally expensive due to the hourly timescales involved and also many models don't archive all the required variables). The restriction associated with this approach is that you have to span the entire globe because the inference is a cumulative calculation from north to south (for a zonal integral) or east to west (for a meridional integral).
For the freshwater change budget, we have the freshwater flux into sea water (wfo) and can infer the change in ocean freshwater content from the ocean salinity data (so). This allows us to infer the anomalous freshwater transport in the ocean.
The changes in ocean freshwater content at any given latitude look relatively small compared to the cumulative change in surface freshwater flux. It might therefore seem reasonable to simply assume that there is no freshwater reorganisation in the ocean. However, if you do this by setting the zonally integrated change in salinity to zero everywhere (for instance) - which means calculating the cumulative sum of the cumulative change in surface freshwater flux from south to north - then the curve you get deviates from the true value (i.e. the ocean reorganisation matters).
In order to capture the freshwater reorganisation outside of the ocean (i.e. atmosphere, soil, land ice, land snow, sea ice) you'd need a whole bunch of variables:
- prw: atmosphere_water_vapour_content (water vapour path; Amon)
- clwvi: atmosphere_cloud_condensed_water_content (mass of condensed (liquid + ice) water in the column; Amon)
- mrso: soil_moisture_content (total soil moisture content; Lmon)
- lwsnl: liquid_water_content_of_snow_layer (LImon)
- pflw: liquid_water_content_of_permafrost_layer (LImon)
- sim: sea_ice_and_surface_snow_amount (OImon)
- more ???
Not all of these are archived.
I'm not sure how realistic it would be to assume the freshwater reorganisation outside of the ocean is zero everywhere. I could check the change in atmospheric water vapor and if it's negligible I could assume the atmospheric transport is just the inverse of the ocean transport? But that doesn't account for water transported a long way by rivers...
Do I need to look at budget closure as well as drift? (Like Liepert 2012, 2013)
The meridional sum plots didn't work that well - better to look at whole basins like Levang.
We might want to distinguish between changes in P-E over land and ocean? In terms of the background here, Michael Roderick from ANU had a letter published in response to Paul Durack’s Science paper which said “we need to stop the simple extrapolation of results from ocean studies to the land and vice versa.” A couple of years later he published a general framework for understanding the response of the water cycle to global warming over land and ocean (Roderick et al 2014) and there’s also another relatively highly cited paper on the same topic (Byrne & O'Gorman, 2015).
I think we need to relate the ocean surface salinity distribution to the land precipitation distribution (or P-E; but E is so small can probably just to P). I'm not sure that those papers have done that.
Use Sverdrup units (10^6 m3 s-1) for climatology flux plots? (Instead of an average annual accumulation.)
He2019 (in the context of ocean heat transport) look at the transport inferred from surface fluxes alone (OHT_FX), from inhomogenous heat storage alone (OHT_HC) and the true advective flux (OHT_VT) which takes into account both. I could use a similar language. (They make a big deal out of the fact that there's a big difference between OHT_FX and OHT_VT.)
Note that in the historical runs even though the planet is warming in all the models, the water cycle (specifically, global P or E) has intensified in some models and weakened in others (presumably due to the outsized influence of AAs).
Overall comments on single forcing runs:
- The tropical precip region behaves more or less as you'd expect in a warming/cooling world
- Deviations from expected behaviour in the evap dominated regions appear to be explained by area effects
- Deviations drom expected behaviour in the high latitude precip regions don't have a clear explanation yet
Expected warming world response:
- Increases in P and E.
- In P dominated regions you'd expect the increase in P to be stronger (leading to an increase in P-E).
- In E dominated regions you'd expect the increase in E to be stronger (leading to a decrease in P-E).
Broad observations:
- The precipitative and evaporative regions all increase in intensity
- The size of the evaporative regions increase at the expense of the high latitude precipitation regions. This is more pronounced in the SH.
Regional observations:
- Tropical precip: The tropics show the expected warming world response. Both P and E increase, but P moreso and P-E increases.
- SH evap: The increase in E is relatively strong (evap area expansion is stronger in SH) and the increase in P is simply on par with other regions, so the P-E decrease is quite strong.
- NH evap: The increase in E is relatively weak (evap area expansion is weaker in NH) and the increase in P is on par with the other regions, so the P-E decrease is quite weak.
- SH precip: The increase in P is relatively weak (not sure why?) but E actually decreases (perhaps due to decrease in area, but I don't think that completely explains it) so P-E increases as you'd expect in a warming world.
- NH precip: The increase in P is relatively strong (not sure why?) while the increase in E is less remarkable, so P-E increases possibly a little more than you'd expect in a warming world.
Expected cooling world response:
- Declines in P and E.
- In P dominated regions you'd expect the decline in P to be stronger (leading to a decrease in P-E).
- In E dominated regions you'd expect the decline in E to the stronger (leading to an increase in P-E).
Broad observations:
- Precipitation decreases in all five regions. Expressed as a percentage change, the declines get stronger as you move further north.
- Evaporation decreases in all regions except the SH precip region. The decline is especially strong in the NH precip region and marginally more pronounced in the SH evap region versus the NH evap.
- The evap region discrepancy appears to be explained by the fact that the area of the SH evap region decreases markedly, whereas the NH evap region increases in area (presumably due to a southward shift in the ITCZ?).
Regional observations:
- Tropical precip: The tropics show the expected cooling world response. Both P and E decline, but P moreso and thus P-E decreases, as you'd expect in a cooling world.
- SH evap: The decline in P is relatively weak (AAs have more influence on rainfall decline as you go north) and decline in E is relatively strong (due to shrinking area), so P-E increase is quite strong.
- NH evap: The decline in P is relatively strong (AAs have more influence on rainfall decline as you go north) and the decline in E is relatively weak (due to increasing area), so the P-E increase is quite weak (in some models P-E actually decreases).
- SH precip: The decline in P is relatively weak (AAs have more influence on rainfall decline in NH) but the decline in E is non-existent (the ensemble median is close to zero change) so P-E decreases as you'd expect in a cooling world. The unchanging E is a bit of a mystery (it doesn't appear to be related to area effects).
- NH precip: The decline in P is relatively strong (AAs have more influence on rainfall decline in NH) but so is the decline in E (not sure why because not related to area) so P-E decreases as you'd expect in a cooling world (maybe marginally more so than the SH precip region)
Broad observations:
- Precipitation increases for some models/regions and decreases for others. Expressed as a percentage change, more models show an increase in precipitation in the SH and a decrease in the NH (i.e. due to the AA influence)
- The precipitative and evaporative regions essentially increase in intensity in the SH precip, NH evap and NH precip regions but show mixed change in SH evap and tropics
- There is very little ensemble median change in the area of the SH evap region
- The relatively large increase in area associated with GHG forcing is cancelled by the relatively large decrease by AA forcing
- The area of the SH precip region decreases
- So the SH evap region expands to the south (GHG forcing) and contracts to the north (AA forcing)
- The area of the NH evap region increases a little (both AA and GHG forcing)
Regional observations:
- Tropical precip: The models are fairly evenly spread between increases and decreases in P, E and P-E.
- SH evap: The models are fairly evenly spread between increases and decreases in P, E and P-E. There's probably no change in E because the area remains unchanged (due to competing influence of AAs and GHGs). P doesn't decrease like it does in the NH evap region because AA induced P decline is less pronounced in SH.
- NH evap: All models show a decrease in P-E (i.e. amplification of the evaporation dominance). Individually, the ensemble P and E changes are a mix of positive and negative values, but in general there's a skew towards decreased P (consistent with AAs reducing P more in the NH) and increased E (consistent with the increase in area that both AAs and GHGs cause).
- SH precip: The models skew towards an increase in P and a decrease in E (due to GHG forced decrease in area), resulting in an increase in P-E (i.e. amplication of the P dominance).
- NH precip: The models very slightly skew towards a decrease in P but show a decrease in E (due to strong AA forced decline - not sure why it's strong), so overall P-E increases.
The increase in Atlantic‐to‐Pacific moisture export may be attributed to either of two mechanisms: (1) the Atlantic‐to‐Pacific moisture export scales with increased evaporation over the Atlantic basin, which is exported from the Atlantic by the climatological moisture transport (the increased evaporation hypothesis), or (2) the length and/or direction that water parcels originating within the Atlantic travel changes with warming, independent of the increase in evaporation (the altered transport hypothesis). Singh2016 find that in a pure GHG forcing experiment, altered transport is by far the dominant mechanism. They propose that increased Atlantic‐to‐Pacific moisture export is due to an increase in the advective moisture transport length scale with warming, which is a result of a decrease in the large‐scale precipitation efficiency (the ratio of precipitation to atmospheric specific humidity - we know that specific humidity is increasing faster than precipitation). Decreasing precipitation efficiency corresponds to increased moisture residence times and, therefore, increased advective length scales.
So we know that warming decreases precipitation efficiency, while cooling increases it. Independent of their cooling effect, aerosols also decrease precipitation efficiency ("the precipitation efficiency of clouds in polluted air is found to be several times lower than that of clouds forming in clean air." Khain2008). We can therefore explain the single forcing experiment results as follows:
- hist-GHG: warming decreases precipitation efficiency, so we get a modest (time integrated anomaly) transport from Atlantic to Pacific
- hist-aer: the cooling and aerosol impacts on precipitation efficiency cancel, so not much change in Atlantic to Pacific transport
There's now a better reference for size of aerosol forcing: Bellouin2019
In Chris Bladwell's analysis of the climatological water cycle, essentially all the ITCZ freshwater is transported south. (I think my accumulated freshwater flux analysis might show something similar. We get used to looking at the zonal mean P-E - which is hemispherically very symmetrical - but don't account for the fact that the surface area of the southern mid-latitudes is much larger than the northern.)