1. Introduction
Water availability is vital to our existence and is a key component in Earth’s ecosystems (e.g., Trenberth and Asrar 2014). Climate change has a profound effect on the global water cycle (Trenberth 2011), and ongoing changes in the water cycle have already had serious consequences on the environment, economy, and society (Vinogradova et al. 2016; Zhang et al. 2007; Willett et al. 2007; Huntington 2006). Predicting and understanding the changes that are taking place in Earth’s water budget is timely and is identified as one of the Science Grand Challenges by the World Climate Research Programme (WCRP). Because the ocean contains 97% of Earth’s water supply and it is the ultimate source of all terrestrial water (Rodell et al. 2015; Gimeno et al. 2015), understanding the global water cycle is impossible without a grasp on changes that are happening with its largest component—the flux of freshwater over the oceans, its transport and storage, and exchange with the atmospheric, terrestrial, and cryospheric elements of the water cycle.
While recent progress in space-based measurements allowed one to produce global, high-resolution products of the components of the ocean water cycle [i.e., evaporation, precipitation, and terrestrial runoff; see, e.g., Schanze et al. (2010) for review], significant uncertainties remain in the flux estimates and long time scales needed to assess possible trends related to anthropogenic climate change (e.g., Schanze et al. 2010; Yu and Weller 2007; Beranger et al. 2006; Stammer et al. 2004; Grist and Josey 2003). One way around this issue is to use variations in ocean salinity as an indicator of the hydrological cycle, given the sensitivity of salinity to freshwater transport in and out of the ocean and availability of accurate salinity measurements over the globe (Durack et al. 2016; Reul et al. 2014; Lagerloef 2012; Roemmich et al. 2009).
This can be done under the assumption that the salt content in the ocean is nearly constant and that any changes in salinity are due to the addition or subtraction of freshwater. The use of salinity as a “rain gauge” representing changes in the hydrological cycle has been successfully demonstrated on sufficiently long, multidecadal time scales (Lago et al. 2015; Skliris et al. 2014; Pierce et al. 2012; Terray et al. 2012; Durack et al. 2012; Durack and Wijffels 2010). In particular, the patterns of 50-yr changes in surface salinity can be used as a direct fingerprint of the amplified water cycle (Durack et al. 2012) with salty areas getting saltier and fresh areas getting fresher (Lago et al. 2015; Skliris et al. 2014; IPCC 2013; Giorgi et al. 2011; Trenberth 2011; Dai 2011; Durack and Wijffels 2010). Apart from multidecadal changes, this simplified view of surface salinity as a rain gauge has been explored but with limited success on subannual-to-interannual times scales and from local to global spatial scales by Vinogradova and Ponte (2013), who concluded that, unless ocean salt fluxes can be parameterized as a function of freshwater fluxes, the use of salinity as a direct proxy of freshwater flux is unlikely. The question remains, however, whether the rain-gauge notion can be extended to time scales from decadal to bidecadal. More generally our goal here is to examine trends in surface salinity and the oceanic component of the global hydrological cycle during the contemporary, relatively well-observed period of the last two decades. Unlike our previous focus on controlling mechanisms of salinity variations and its steady state (Vinogradova and Ponte 2013; Ponte and Vinogradova 2016), here we are assessing whether amplification of the ocean water cycle can be seen over this shorter period. Is the simplifying concept of the amplified global water cycle applicable over the oceans during the last two decades, and if so, to what extent?
Herein the ocean water cycle refers to the resulting flux of freshwater entering the ocean via precipitation P and river runoff and sea ice melt R and leaving the ocean via evaporation E, also commonly referred to as E − P − R. Our analysis indicates that, for the period studied (1993–2010), there is an overall amplification of the ocean water cycle seen in surface flux estimates (section 3) but little evidence of such global amplification in the surface salinity record (section 4). These findings imply a time of emergence of anthropogenic hydrological signals shorter in surface freshwater fluxes than in surface salinity and point to the importance of ocean circulation and salt transports in shaping patterns of decadal change in surface salinity. Budget analysis based on a recent global ocean state estimate (Forget et al. 2015) shows that the processes controlling the observed decadal salinity involve ocean circulation changes, which act to reset the influences of surface freshwater fluxes in many regions (section 5), with implications for the interpretation of the rain-gauge concept and salinity patterns in a changing climate (section 6).
2. Approach
a. Basic tools
A direct link between surface salinity and the amount of freshwater that enters the ocean is often obscured by upper-ocean dynamics by virtue of the salt conservation equation (Vinogradova and Ponte 2013; Hasson et al. 2013; Yu 2011). Even in regions that are dominated by extreme surface freshwater fluxes, such as the tropical convergence zones featuring heavy rainfall or subtropical evaporative regions in the North Atlantic and North Pacific, oceanic contributions to salinity variations complicate the use of salinity observations as a direct proxy of freshwater flux (Yu 2011; Vinogradova and Ponte 2013). Other complications in assessing whether the direct link is even possible are limitations of the methods used to address this problem, which either involve the use of available observations (Bingham et al. 2012; Skliris et al. 2014) or output of an ocean general circulation model (Yu 2011; Helm et al. 2010; Durack et al. 2012). While model realism is often a concern (Durack et al. 2012), observational analyses generally imply significant imbalances in the salinity budget (Yu 2011), partly due to large uncertainties in the estimates of freshwater fluxes (Yu and Weller 2007; Schanze et al. 2010; Josey et al. 2013; Skliris et al. 2014). Here we present an alternative approach that combines observational and modeling frameworks and provides dynamically consistent estimates of salinity and associated atmospheric and oceanic freshwater fluxes, without introducing arbitrary sources or sinks of freshwater (Forget et al. 2015). Another advantage of the proposed method is the closeness of the analyzed estimates to existing ocean observations, including salinity from Argo, WOCE, and hydrographic climatologies (Vinogradova and Ponte 2013).
Our monthly estimates of surface salinity and surface freshwater fluxes are from the most recent configuration of an ocean state estimate developed as part of the Estimating the Circulation and Climate of the Ocean (ECCO) project in the presence of an active sea ice model on a global grid (Forget et al. 2015). In the ECCO framework, the Massachusetts Institute of Technology General Circulation Model (MITgcm) is fit in a least squares sense to several hundred million (satellite and in situ) ocean observations (Wunsch and Heimbach 2007; Wunsch et al. 2009; Forget et al. 2015). Horizontal resolution is 1°, except that it is refined to about ⅓° near the equator in the meridional direction. In the vertical the model has 50 levels, and the grid spacing increases from 10 m near the surface to 450 m in the abyss. Subgrid mixing includes diapycnal and isopycnal components following Gaspar et al. (1990), while geostrophic eddies are parameterized using Redi (1982) and Gent and McWilliams (1990). ECCO estimates are routinely validated, with model and data misfits available through web-based displays (online at http://www.ecco-group.org/). The solutions are close to the available data within a priori estimates of the data noise.
To complement the estimates of freshwater flux from ECCO, we have also utilized satellite-derived evaporation based on objectively analyzed air–sea fluxes (OAFlux; Yu and Weller 2007), precipitation datasets from the Global Precipitation Climatology Project (GPCP) (version 2.2; Huffman et al. 2009; Adler et al. 2003) and the Climate Prediction Center (CPC) Merged Analysis of Precipitation (CMAP) dataset (Xie and Arkin 1997), and one reanalysis product from the NOAA/National Centers for Environmental Prediction (NCEP) and the National Center for Atmospheric Research (NCAR) (NCEP–NCAR reanalysis; Kalnay et al. 1996).
b. Budget framework
One of the advantages of ECCO is that, unlike common data assimilation products, ECCO does not introduce sequential state increments of unknown physical origin and thus is a dynamically consistent state estimate (Wunsch 2006). This allows one to formulate closed budgets for any tracer (including salinity) making ECCO budget estimates a valuable tool to study the link between the surface freshwater fluxes and salinity changes, as well as for determining general physical mechanisms driving tracer evolution.
An important novelty of the budget approach presented here is its formulation under realistic representation of the effects of freshwater fluxes at the ocean surface. Previous modeling studies (e.g., Hasson et al. 2013; Vinogradova and Ponte 2013; Qu et al. 2011) examined salinity budget using “virtual” salt flux surface boundary conditions, which are generally speaking unphysical because they create or destroy salt mass (Roullet and Madec 2000). To conserve mass in such formulation (as well as to reduce salinity drifts), models often linearly restore sea surface salinity to observations. However, this can introduce inappropriate sources into the salinity conservation equation by adding an artificial salt flux through the air–sea interface. In the real ocean, however, there are no fluxes of salt at the surface boundary—precipitated water is essentially fresh and evaporation does not remove salt from the ocean but changes its concentration. An exception is sea ice–covered regions, where salt can be added at the ocean surface during brine rejection. In the present ECCO solution, the real freshwater flux across the upper surface is specified as the vertical velocity boundary condition for the continuity equation (Campin et al. 2004), and since no additional salinity flux is introduced at the surface, the ocean salt content is conserved (Roullet and Madec 2000).
Furthermore, the real freshwater flux also affects the thickness of the surface layer and thus the model volume, which is commonly expressed in a free-surface formulation (Campin et al. 2004). Because displacements in free-surface height are much smaller than the average depth of the ocean, models traditionally linearize the equation for the free-surface height (Griffies et al. 2001), which, while simplifying computations (Marshall et al. 1997), can alter the exact closure of the salt budget (Huang 1993). To avoid these issues, another innovation to the budget approach is introduced here as in Forget et al. (2015): we allow ocean (layer) thickness to be time-varying and include nonlinearities in free-surface calculations. Such innovation ensures consistency between the salt conservation and the ocean volume.
The resulting salinity equation defines the rate of salinity change (i.e., salinity tendency in time) as a combination of (i) dilution by air–sea freshwater fluxes, (ii) fluxes of salt in ice-covered regions, and ocean fluxes of salinity due to (iii) advective and (iv) diffusive convergences. Full derivation of the salinity equation in the real freshwater flux formulation is given in the appendix. For more intuitive analysis, we will integrate the salinity Eq. (A5) in time, converting each budget term from tendency space to salinity space.
c. Linear trends and changes
In what follows we examine the changes in salinity, ocean salinity fluxes, and the ocean water cycle that occurred during the last two decades spanning from the beginning of 1993 to the end of 2010. The changes are estimated as a linear fit against time and then multiplied by the period’s length (18 yr) to show total change since 1993. To ensure statistical significance of our results, we perform trend distribution testing. Tests for the deviation of distribution means from zero are done using the two-tailed Student’s t test at 95% confidence interval, with t intervals used as trend errors at the desired confidence interval. However, instead of the traditional white-noise random draw, we assume that the process under question is a red-noise process, represented as an autoregressive model of order 1 (AR1). The use of an AR1 instead of a white-noise process is motivated by its relatively long memory, as the current member is correlated with all previous members although with decreasing coefficients. This allows us to account for serial correlation of the residuals (Calafat and Chambers 2013), hence making it more appropriate for the analyzed time-integrated ocean variables (C. Piecuch 2016, personal communication).
Here, surface changes will refer to those that occur within the top 10 m, which is the thickness of the top model layer in ECCO solution. While mixed layer depth is often used as another representation of the ocean “surface” properties (e.g., Vinogradova and Ponte 2013; Qu et al. 2011; Kim et al. 2006), the choice of 10-m depth is deliberate and fits our purpose of staying as close to the air–sea boundary as the ECCO solution allows, where the effect of freshwater flux on salinity changes is expected to be the strongest.
3. Recent changes in the ocean water cycle
In the last two decades (1993–2010), there have been robust patterns of change in the ocean water cycle, as represented by estimates of surface freshwater flux (e.g., Fig. 1a). A number of tropical regions are experiencing an increase in precipitation (red in Fig. 1b), including the intertropical convergence zone (ITCZ), the western Pacific warm pool, and parts of the South Pacific convergence zone (SPCZ), where the wetting is enhanced by a decrease in evaporation (blue in Fig. 1b). In contrast, a number of subtropical regions are becoming drier, associated with a general increase in evaporation (e.g., western Australia), decrease in precipitation (e.g., the U.S. West Coast), or a combination of the two (e.g., the U.S. East Coast).
Total change in (a) oceanic component of the global water cycle, (c) precipitation, (d) evaporation, and (e) surface salinity from 1993 to end of 2010, computed as a linear trend multiplied by the period length. Negative values in (a) indicate increased freshwater input since 1993. Negative values in (c) show surface freshening. Only significant changes, judged using a 2σ effect at 95% confidence intervals and accounting for serial correlation, are shown (otherwise shaded gray). (b) Average rates at which the freshwater enters (blue) or leaves (red) the ocean. Matching colors in (a) and (b) indicate that the ocean water cycle has amplified—wet regions are getting wetter and dry regions are getting drier. (f) Spatial patterns of the steady-state surface salinity (annual mean) are characterized by salty subtropics and fresher tropical and high-latitude regions.
Citation: Journal of Climate 30, 14; 10.1175/JCLI-D-16-0626.1
Some of these patterns are expected in a warming climate (Trenberth and Asrar 2014), caused in part by human activities (IPCC 2013; Trenberth 2011). Warming increases the water-holding capacity of the atmosphere, which in turn leads to a positive trend in water vapor over the ocean (where the water supply is virtually unlimited), according to the Clausius–Clapeyron equation (Trenberth 2011). The increased amount of water vapor also gives rise to increased convergence of moisture (Trenberth et al. 2003) and thus more intense precipitation (Liu et al. 2009). An increase in water vapor also affects Earth’s radiative forcing and thus the amount of water that evaporates from the surface. Water vapor feedback eventually exceeds net cooling from atmospheric and temperature changes, leading to a net increase in evaporation (Colman 2015). Such changes in water vapor are evident in observations (Mieruch et al. 2014). Thus, with more moisture available to be transported from subtropics (divergence zone) to the tropics (convergence zones) and storm-track regions at higher latitudes, more freshwater is brought to wet regions and removed from dry regions. This mechanism, when wet regions are getting wetter and dry regions are getting drier, has been labeled the “rich get richer” mechanism (Giorgi et al. 2011) and is often mentioned along with enhancements to extreme events, such as floods and droughts (Emori and Brown 2005; Allen and Ingram 2002).
An intensification of existing annual patterns of freshwater flux is visible, as blue and red regions in Fig. 1a match blue and red regions in Fig. 1b. And while there is a general tendency of drying of arid subtropical regions and wetting in the tropics and mid-to-high latitudes, it does not seem to be the case everywhere over the globe. For clarity, we highlight the places with amplified ocean water cycle as orange in Fig. 2a, shading it if a region with negative anomalies in freshwater flux (compared to the global mean) experiences a negative trend, or if a region with positive freshwater anomaly is trending up. From Fig. 2a, intensification in the water cycle occurred in more than half of all ocean regions. However, in about 40% of all cases, weakening in freshwater flux patterns was observed (shaded gray), including the North Atlantic, large areas of the tropical Pacific, and parts of the Indian and Southern Oceans.
Ocean regions where the mean climatological patterns in (a) freshwater flux and (b) surface salinity amplified during the last two decades (otherwise shaded light gray). (c),(d) The strength of the pattern amplification: the x axis represents anomalies (relative to the global average) in zonal ocean basin averages of climatological mean (c) freshwater flux and (d) surface salinity, whereas the y axis represents their corresponding changes since 1993. Colors show different ocean basins. Total pattern amplification and pattern correlation are computed as the slope (black line) and correlation coefficient of their linear regression, following Durack et al. (2012) but including full Arctic and Southern Oceans.
Citation: Journal of Climate 30, 14; 10.1175/JCLI-D-16-0626.1
Strong dependence on location in pattern amplification holds toward larger, basinlike spatial scales, as evident from Table 1 and Fig. 2c. Here, similar to Durack et al. (2012) and Durack (2015) but including differentiation by regions, we examine zonal ocean basin averages of climatological mean freshwater flux anomaly versus its recent changes (Fig. 2c). The highest correlations, with pattern amplification of 5% and 11%, are found in the South Atlantic and South Pacific Oceans. In contrast, the North Atlantic and Indian Ocean display scattered behavior, and their mean patterns have actually weakened since 1993 by −7% and −4%, respectively.
Amplification in mean climatological patterns in the ocean water cycle (E − P − R) and surface salinity between 1993–2010 for different ocean basins.
Overall, our estimates suggest that globally, during the past 20 years the ocean water cycle has amplified by about 5% with the spatial pattern correlation coefficient around 0.5, computed as the slope and correlation coefficient of the linear regression between all points in Fig. 2c. The result implies that on average wet areas did get wetter and dry areas did get drier in the last two decades, although significant regional variations exist. Average amplification in the ocean water cycle of 5% is consistent with surface warming of about 0.65°C between 1993 and the end of 2010 based on the ECCO sea surface temperature estimates (not shown). The inferred ocean water cycle intensification of 7.6% °C−1 is close to that predicted by the Clausius–Clapeyron relationship (~7% °C−1).
4. Recent changes in surface salinity
What are the changes in surface salinity over the past two decades? The most notable features are strong surface freshening in many subtropical regions and salinification in large extents of the tropical Pacific, reaching 0.3 ± 0.01 psu in magnitude and stretching across almost the whole ocean basins (Fig. 1e). What is interesting is that some of these changes in surface salinity seem counterintuitive to those in the freshwater flux. For example, drying in the subtropical South Pacific and North Atlantic did not result in higher salinities. Likewise, moisture increase in the tropical regions of the western Pacific warm pool and ITCZ did not lead to a decrease in salinity, and the Atlantic–Pacific interbasin contrast—a well-known attribute of surface climatology with implications for the global thermohaline ocean circulation and climate (Seidov and Haupt 2002; Gordon 2001)—seemed to weaken, particularly in the tropics.
Similar to the freshwater flux, Figs. 2b,d show the regions and the strength of the pattern amplification in sea surface salinity. In the absence of changes in ocean circulation and mixing, it is expected that the patterns in surface salinity follow similar rich-get-richer concept where salty ocean regions are getting saltier and fresh regions are getting fresher (Durack and Wijffels 2010; Hosoda et al. 2009; Schmitt 2008). From Fig. 2b, vast portions of the Southern Ocean, central tropical Pacific, and parts of the Atlantic and Arctic Oceans do exhibit amplification of the mean salinity patterns over the 20-yr period. The most dramatic intensification is found in high-latitude regions, with basin-average amplification of 54% in the Southern Ocean and 24% in the Arctic Ocean (Table 1). However, like for freshwater fluxes, there are significant geographical variations in salinity pattern amplification, and even more so in their strengths. Moderate amplifications are found in the South Pacific (11%) and South Atlantic (6%), and a strong basin-average weakening of the salinity patterns is a prominent feature of the North Atlantic (−40%). This wide spread in the amplification strength, from −40% to +54%, and generally scattered behavior between broad-scale salinity patterns in Fig. 2d lead to low values in global-average correlations. Unlike freshwater flux patterns, overall intensification of the surface salinity pattern is less than 1%, implying that, globally, the use of 20-yr salinity trends to fingerprint intensification of the water cycle might be problematic.
The lack of correspondence between the changes in surface salinity and the ocean water cycle is expected if freshwater flux is not the only driver of salinity changes, and salt fluxes associated with mixing and advection processes internal to the ocean are also involved (Ponte and Vinogradova 2016; Gordon 2016; Vinogradova and Ponte 2013). Analysis of the surface salinity budget in the ECCO solution is pursued next to shed light on what controls the estimated trends over the 1993–2010 period.
5. Nature of recent salinity changes
a. Salinity budget: The role of advection
As it is clear from the budget equations in the appendix, decadal trends in salinity are the result of small imbalances in the cumulative effects of forcing, advection, and diffusion terms over the period studied. In a simplified rain-gauge concept, forcing wins the battle with mixing to drive same-sign salinity changes, with advection not affecting this correlation. However, from the surface salinity budget analysis in Fig. 3, not only are advection contributions to the salinity changes not negligible, but ocean fluxes in general are very effective in blurring the changes imposed by surface forcing during the last two decades. Dissipation of atmospheric effects by mixing is most common, as seen from their opposite signs in Figs. 3b,d, including compensation of atmospheric freshening in the tropics and drying in parts of the subtropics. But the sum of forcing and diffusion terms (Fig. 3f) does not look for the most part like the pattern of forcing in Fig. 3b, as one would expect in a simple rain-gauge concept. Instead, the salinity balance is far from passive and the role of advection is clear (Fig. 3c).
Surface salinity budget. Changes in (a) surface salinity from 1993 to the end of 2010 induced by time-integrated change in (b) freshwater flux, (c) advective, and (d) diffusive convergences resulting from ocean currents and mixing, respectively, computed according to Eq. (A5). White line is the zero contour. [Black boxes in (a) delineate regions discussed in Fig. 4.] For a more intuitive analysis, integration in time is performed to show an accumulative effect of each process in driving salinity changes. (f) Changes in salinity due to a combined effect of surface forcing and mixing are shown to see the regions where the rain-gauge concept is expected. (e) Places of active–passive salinity budget (i.e., where ocean advection overcompensates atmospheric forcing leading to trends in surface salinity that are of the opposite sign than those imposed by surface forcing).
Citation: Journal of Climate 30, 14; 10.1175/JCLI-D-16-0626.1
To further highlight the importance of advection, we identify places of active ocean response, which we define here as regions where despite favorable conditions for salinity to follow the changes in atmospheric forcing (i.e., the forcing term exceeds diffusion) the resulting salinity trends are of the opposite sign from those imposed by the freshwater flux. We find that the ocean response is active in about 25% of the global ocean area. This is generally confined within two zonal belts between approximately 15° and 40° latitude in the Southern and Northern Hemispheres (orange in Fig. 3e). Here, advective processes, including wind-induced Ekman transport, overcompensate atmospheric forcing leading to “counterintuitive” salinity trends: the ocean surface is getting fresher despite strong atmospheric drying. It is worth noting that active ocean response results in predominantly negative salinity trends, which is of course related to the fact that the advection is mostly negative, although with noticeable exceptions in the subtropics (Fig. 3c). The result is interesting and suggests that, despite the variable nature of horizontal currents, ocean dynamics have been actively decreasing salinity in the last 20 years in the tropical and high-latitude regions. The finding somewhat echoes Ponte and Vinogradova (2016), who also puzzle over the “large-scale, single-sign behavior of the advective processes in the tropics” in the time-mean salinity budget.
b. Regional changes: Salinity maxima and minima
We now take a closer look into the “salty gets saltier, fresh gets fresher” concept by examining several regions of salinity extremes. We pick five regions corresponding to the subtropical salinity maxima and five regions corresponding to the tropical and high-latitude salinity minima. Our choice of regions was partly governed by the analysis of the subtropical maxima by Gordon et al. (2015), as well as the location of recent NASA field programs Salinity Processes in the Upper-Ocean Regional Study, experiments 1 and 2 (SPURS-1 and SPURS-2, respectively; Lindstrom et al. 2015). Preference was also given to regions where the change in salinity was large (see Fig. 3a for locations).
A number of different regimes are present depending on regions considered (Fig. 4). Apart from the Arctic and South Atlantic (SATL) locations, low-salinity regions are getting saltier and high-salinity regions are getting fresher implying that the rich-get-richer paradigm is not valid here. Recent surface salinification in low-salinity regions arises from strong diffusive fluxes overriding opposite tendencies from forcing and advection [e.g., SPCZ, ITCZ, Bay of Bengal (BOB), and subpolar North Atlantic (SPNA)]. In contrast, recent surface freshening in high-salinity regions is due to a collective effort of ocean advection and mixing [e.g., South and North Pacific (SPAC and NPAC, respectively), North Atlantic (NATL), and south Indian Ocean (SIND)]. The balance in these salinity maxima regions is active: advection is tipping the scale, which leads to surface freshening despite a strong push for salinification from surface forcing. Notice also that the salinity intensification at the Arctic location, where fresh is getting fresher, is not atmospherically driven but rather is dominated by strong advection that overwhelms both mixing and surface forcing.
Changes in surface salinity (SSS; red), time-integrated changes in freshwater flux (yellow), advection (purple), and mixing (green) averaged over the 10 boxed regions delineated in Fig. 3a. Units are psu. The five bottom regions correspond to the subtropical salinity maxima in SPAC, SATL, NPAC, NATL, and SIND Oceans. The five top regions correspond to the tropical and high-latitude salinity minima in the SPNA, east Arctic Ocean (EARC), BOB, ITCZ, and SPCZ. For clarity, the changes in surface salinity are multiplied by a factor of 100.
Citation: Journal of Climate 30, 14; 10.1175/JCLI-D-16-0626.1
The results clearly show that the changes in surface salinity in the last two decades reflect a delicate balance between all the terms in the salinity equation without following the pattern of any particular term. In addition to the terms in the salinity equation, there might be other processes affecting the ocean circulation that are related to forcing other than freshwater fluxes (e.g., surface winds and heat fluxes). Therefore, even though these factors are not explicitly present in the salinity equation, they can affect the surface salinity budgets and thus complicate a one-to-one correspondence between freshwater fluxes and salinity even further.
c. Influence of natural variability: Interdecadal oscillations
Some of these changes may in fact be related to natural variability at shorter time scales, which can mask long-term salinity trends related to secular changes in the ocean water cycle. Indeed, a large volume of literature indicates the importance of climate variability in driving salinity changes at interannual to decadal time scales, including clear signatures of ENSO (e.g., Qu and Yu 2014; Zhu et al. 2014; Hackert et al. 2011; Ballabrera-Poy et al. 2002) and NAO (e.g., Buckley and Marshall 2016; Sarafanov et al. 2008; Häkkinen and Rhines 2004) events in the contemporary salinity patterns. Similarly, the influence of climate variability at interdecadal time scales is well recognized, including structural changes associated with the interdecadal Pacific oscillation (IPO; Henley et al. 2015; Trenberth and Fasullo 2013; Deser et al. 2012, 2010; Mantua et al. 1997). The IPO patterns are typically expressed in terms of alternating anomalies in sea surface temperature, extending from the tropical Indo-Pacific to the North and South Pacific. In addition, the IPO impacts on other climate variables, such as sea surface height, sea level pressure, precipitation, and drought, have also been identified (Wang et al. 2014; England et al. 2014; Dai 2012). Most recently, the importance of the IPO has been highlighted in the context of a recent slowdown in surface warming (e.g., England et al. 2014; Trenberth and Fasullo 2013; L’Heureux et al. 2013; Meehl et al. 2011; Katsman and van Oldenborgh 2011; Kaufmann et al. 2011; Folland et al. 2002), although with significantly less attention paid to the corresponding interdecadal changes in salinity (e.g., Overland et al. 1999; Lagerloef 1995).
Role of natural climate variability in driving salinity trends in the Pacific Ocean: influence of the IPO mode. (a) Expression of the IPO loading pattern in surface salinity, computed by linearly regressing monthly ECCO salinity anomalies upon (b) the IPO time series in normalized units of standard deviation (SD). The IPO time series are defined by their leading principal component (Henley et al. 2015; available at www.esrl.noaa.gov/psd/data/climateindices/list/). The black curve is a 61-month moving average as in Trenberth and Fasullo (2013). Salinity anomalies are obtained by removing the climatological annual cycle and the global mean salinity anomaly at each grid point (following, e.g., Deser et al. 2010). (c) Trends in the IPO loading salinity patterns. (d) Residual trends after the relevant IPO loading pattern is removed. To be consistent with Fig. 1, (c) and (d) represent total change, computed as a linear trend multiplied by the period length.
Citation: Journal of Climate 30, 14; 10.1175/JCLI-D-16-0626.1
The index monthly time series [the term IPO in Eq. (1)] in Fig. 5b reveal interdecadal regimes of the IPO over the study period, with positive phases before 1998 and (mostly) negative phases after 1999 onward (see black line for clarity). During its negative phase, the IPO is manifested by strengthened winds in the tropical Pacific Ocean, which in turn leads to an acceleration of the equatorial currents and intensification of the subtropical gyres. This shift in phase and an acceleration of the trade winds have been identified as the leading cause in driving trends in sea surface temperature (Wu et al. 2012) and sea level (England et al. 2014; McGregor et al. 2012) during 1992–2011, which both exhibit the expected pattern of warming (rise) in the western boundary and cooling (fall) over the tropical central and eastern Pacific.
To examine the corresponding IPO contribution to the salinity trends, we contrast the IPO-related trends in Fig. 5c with the residual trends in Fig. 5d, computed as a linear fit of αIPO and the residual ε, respectively. Although not as pronounced as for temperature (and sea level) trends [cf., e.g., England et al. (2014) and their Fig. 2], the IPO influence on the salinity trends in the Pacific Ocean is clearly detectable: on average, about 30% of changes in the corresponding tripole salinity trend pattern (freshening in the subtropics and salinification in the tropics in Fig. 1e) can be explained by the IPO variability. The highest IPO contribution to the bidecadal trends (>60%) is found in the equatorial part of the warm pool and in the central equatorial Pacific (red in Fig. 5c). Here, IPO-related trend largely accounts for the salinification during the last two decades and can partly explain positive salinity trends despite an increase in moisture and the input of freshwater into the ocean (cf. Fig. 1a). In many other regions, however, a large part of the total salinity change pattern does not seem to be related to the IPO, including subtropical regions that exhibit strong freshening since 1992. In such regions, it would be of merit to investigate in the future the influence of other modes of climate variability, as well as to use more rigorous and quantitative methods of isolating internal variability from secular trends. Overall, from the simple analysis here, we conclude that the change in the salinity in the past two decades detected here is indeed partly explained by the natural climate variability, including IPO and its phase shift in the late 1990s.
It is interesting to note that, unlike for salinity, changes in the ocean water cycle over the last 20 years reported here have more similarities with the long-term changes in E − P driven by increased radiative forcing (Durack et al. 2012; Greve and Seneviratne 2015). This might suggest that the time of emergence for the ocean water cycle might be substantially earlier than that for surface salinity. In other words, for the ocean water cycle 20 years might be sufficient for the anthropogenic signals to rise above the “noise” level of natural variability. A more detailed analysis is left for future study.
d. Uncertainties and limitations
An important caveat worth mentioning is the general difficulty in determining the robustness of our estimates of recent trends in the ocean water cycle and surface salinity. In particular, until full Argo coverage starts around 2005, surface salinity data are much sparser in space and time, limiting the ability to accurately resolve changes in the salinity during the first decade of the period considered. The ECCO estimate used here partially mitigates for the inhomogeneous salinity data by including constraints to other datasets, namely satellite altimetry and sea surface temperature, that have a relatively dense and homogenous coverage since 1993.
Similar to salinity, the determination of trends in the ocean water cycle over the period of analysis (1993–2010) may involve substantial uncertainty. As mentioned above, existing estimates of E and P still suffer from large uncertainties, whether they are based on reanalysis, combined reanalysis- and satellite-derived observational products, or model solutions (e.g., Yu et al. 2017; Chaudhuri et al. 2016; Skliris et al. 2014; Josey et al. 2013; Trenberth et al. 2005, and references herein). Driven by differences in product methodologies, sources of input data, inadequate observational coverage, and other issues, there are major discrepancies among reported estimates of freshwater fluxes, which limits our ability to accurately resolve variability, including trends in the hydrological cycle presented here.
For example, Fig. 6 compares the trends in the ocean water cycle in the last two decades from ECCO with those using various precipitation products, including GPCP (Huffman et al. 2009; Adler et al. 2003) and CMAP (Xie and Arkin 1997) precipitation datasets, and the NCEP–NCAR reanalysis product (Kalnay et al. 1996). To estimate the net flux of freshwater into the oceans, each precipitation product was complemented by the evaporation fields from the satellite-based OAFlux synthesis derived by Yu and Weller (2007). A number of main features are common among all E − P trends in Fig. 6, such as an increase in moisture over the western tropical Pacific and parts of the SPCZ and a general drying over the central equatorial Pacific and near the U.S. eastern seaboard. However, differences in E − P patterns of change are also evident, particularly in the tropics and high latitudes. Using a variety of metrics, Yu et al. (2017) discuss potential sources for the observed discrepancies among various E and P products that an interested reader is further referred to. Important for our purposes, however, is the stability of the amplification estimates. Despite apparent differences in detail among various patterns of change in E − P, all products suggest a global amplification of the water cycle during the last two decades by 2% (OAFlux and CMAP), 3% (OAFlux and GPCP), and 4% (OAFlux and NCEP–NCAR reanalysis), although the spread in regional amplitudes is still large (not shown). These global values are also close to the estimates of 5% computed from the ECCO solution, indicating the robustness of our conclusions within data uncertainties.
Change in the amount of freshwater flux entering the oceans between 1993 and 2010 based on (a) ECCO solution and the difference between the satellite-based evaporation from OAFlux (Yu and Weller 2007) minus precipitation from (b) GPCP (Huffman et al. 2009), (c) CMAP (Xie and Arkin 1997), and (d) NCEP–NCAR reanalysis (Kalnay et al. 1996). Notice that, unlike in (b)–(d), fluxes from ECCO also include components associated with river and ice dynamics.
Citation: Journal of Climate 30, 14; 10.1175/JCLI-D-16-0626.1
6. Looking ahead
The use of salinity to monitor Earth’s water cycle has proven to be a challenge but has also offered a tremendous opportunity for the research community, opening new frontiers in salinity oceanography. The past decade has seen a wide range of salinity-based applications ranging from a simple view of salinity as a rain gauge on multidecadal time scales (Durack 2015; Skliris et al. 2014; Durack et al. 2012) to more elaborate techniques that use salinity observations to predict terrestrial drought and flood events (Li et al. 2016), to monitor river plumes and flood events (Gierach et al. 2013), to describe propagation of planetary and tropical instability waves (Lee et al. 2012), oceanic fronts (Kao and Lagerloef 2015), and eddies (Bryan and Bachman 2015), and to model transport of atmospheric moisture (Liu and Tang 2005) and ocean heat exchange (Köhl et al. 2014). Adding to this list, here we continue searching for more ways to use salinity as a fingerprint of the changing ocean circulation and the global hydrological cycle. As confirmed by our current results, as well as by the work of others (Gordon 2016; Hasson et al. 2013; Vinogradova and Ponte 2013; Yu 2011), the seas do not always get saltier if it rains less, nor do they always get fresher if it rains more. Variability in surface salinity alone does not track the changes in air–sea freshwater fluxes (at least on subannual–bidecadal time scales), and the linearity between the two variables is generally not expected. Therefore, a more evolved formulation of the rain-gauge concept is warranted.
One way to improve estimates of freshwater fluxes is through a combination of salinity variations in conjunction with information about ocean fluxes, including computations based on the divergence of ocean freshwater transports and ocean inverse calculations (Wijffels et al. 1992; Ganachaud and Wunsch 2003; Stammer et al. 2004; Köhl et al. 2014). Recent progress in the salinity-observing system (Vinogradova et al. 2016; Durack et al. 2016) includes global high-resolution spaceborne measurements of sea surface salinity from NASA’s Aquarius instrument on board Satélite de Aplicaciones Cientificas-D (SAC-D; 2011–15) (Lagerloef 2012), the European Space Agency’s Soil Moisture and Ocean Salinity (SMOS; 2009–present) (Reul et al. 2014), and the Soil Moisture Active–Passive (SMAP) launched in January 2015, as well as NASA’s dedicated field missions like SPURS-1 (Lindstrom et al. 2015) and SPURS-2 and unprecedented three-dimensional sampling by the network of Argo floats, which has provided near-global coverage since 2005 (Roemmich et al. 2009). Given these datasets, we now have an opportunity to include new information on salinity into climate models and ocean state estimations that ultimately can lead to a more accurate representation of ocean circulation in general and representation of the freshwater exchange within the ocean–atmosphere–land system in particular.
In our search for fingerprints of the changing climate using salinity measurements, we also advocate for the use of salinity as an implicit, rather than direct, indicator of changes affecting Earth’s climate system. Salinity partly determines seawater density and thus the ocean circulation and stratification. That, in turn, can affect the ocean heat transport, release, and storage, which can ultimately impact Earth’s thermal, energy, and carbon balance. Therefore, salinity trends of large-scale nature and considerable magnitudes reported here are yet more evidence of current alterations of the global climate. Such changes in salinity between now and the early 1990s are not only an important indicator of the ocean state but also provide a glimpse of what is happening within other fundamental cycles of Earth’s system, including the change of fluxes of CO2 (Salisbury et al. 2015). Using this information to project future changes in Earth’s climate system is the next challenge.
Acknowledgments
This work was supported by NASA’s Physical Oceanography Program and the Ocean Surface Salinity Project through Grant NNX16AP79G to Cambridge Climate Institute. We thank our ECCO colleagues Gaël Forget and Patrick Heimbach for all their efforts on the production of ECCO version 4 and Martha Buckley for her help with formulation of salinity budgets. We appreciate insightful suggestions from Julian Schanze and three anonymous reviewers who helped us improve our manuscript.
APPENDIX
Formulation of ECCO Salinity Budget under Real Freshwater Flux Boundary Condition
Surface salinity budget. Changes in (a) surface salinity tendency from 1993 to the end of 2010 induced by the change in (b) freshwater flux and (c) advective and (d) diffusive convergences resulting from ocean currents and mixing, respectively, computed according to Eq. (A5).
Citation: Journal of Climate 30, 14; 10.1175/JCLI-D-16-0626.1
REFERENCES
Adler, R. F., and Coauthors, 2003: The Version-2 Global Precipitation Climatology Project (GPCP) monthly precipitation analysis (1979–present). J. Hydrometeor., 4, 1147–1167, doi:10.1175/1525-7541(2003)004<1147:TVGPCP>2.0.CO;2.
Allen, M. R., and W. J. Ingram, 2002: Constraints on future changes in climate and the hydrologic cycle. Nature, 419, 224–232, doi:10.1038/nature01092.
Ballabrera-Poy, J., R. Murtugudde, and A. J. Busalacchi, 2002: On the potential impact of sea surface salinity observations on ENSO predictions. J. Geophys. Res., 107, 8007, doi:10.1029/2001JC000834.
Beranger, K., B. Barnier, S. Gulev, and M. Crépon, 2006: Comparing 20 years of precipitation estimates from different sources over the World Ocean. Ocean Dyn., 56, 104–138, doi:10.1007/s10236-006-0065-2.
Bingham, F. M., G. R. Foltz, and M. J. McPhaden, 2012: Characteristics of the seasonal cycle of surface layer salinity in the global ocean. Ocean Sci., 8, 915–929, doi:10.5194/os-8-915-2012.
Bryan, F., and S. Bachman, 2015: Isohaline salinity budget of the North Atlantic salinity maximum. J. Phys. Oceanogr., 45, 724–736, doi:10.1175/JPO-D-14-0172.1.
Buckley, M. W., and J. Marshall, 2016: Observations, inferences, and mechanisms of Atlantic meridional overturning circulation variability: A review. Rev. Geophys., 54, 5–63, doi:10.1002/2015RG000493.
Calafat, F. M., and D. P. Chambers, 2013: Quantifying recent acceleration in sea level unrelated to internal climate variability. Geophys. Res. Lett., 40, 3661–3666, doi:10.1002/grl.50731.
Campin, J.-M., A. Adcroft, C. Hill, and J. Marshall, 2004: Conservation of properties in a free surface model. Ocean Modell., 6, 221–244, doi:10.1016/S1463-5003(03)00009-X.
Chaudhuri, A., R. M. Ponte, and G. Forget, 2016: Impact of uncertainties in atmospheric boundary conditions on ocean model solutions. Ocean Modell., 100, 96–108, doi:10.1016/j.ocemod.2016.02.003.
Colman, R. A., 2015: Climate radiative feedbacks and adjustments at the Earth’s surface. J. Geophys. Res. Atmos., 120, 3173–3182, doi:10.1002/2014JD022896.
Dai, A., 2011: Drought under global warming: A review. Wiley Interdiscip. Rev.: Climate Change, 2, 45–65, doi:10.1002/wcc.81.
Dai, A., 2012: The influence of the inter-decadal Pacific oscillation on US precipitation during 1923–2010. Climate Dyn., 41, 633–646, doi:10.1007/s00382-012-1446-5.
Delcroix, T. M., and J. Picaut, 1998: Zonal displacement of the western equatorial Pacific “fresh pool.” J. Geophys. Res., 103, 1087–1098, doi:10.1029/97JC01912.
Delcroix, T. M., and M. J. McPhaden, 2002: Interannual sea surface salinity and temperature changes in the western Pacific warm pool during 1992–2000. J. Geophys. Res., 107, 8002, doi:10.1029/2001JC000862.
Deser, C., M. A. Alexander, S. P. Xie, and A. S. Phillips, 2010: Sea surface temperature variability: Patterns and mechanisms. Annu. Rev. Mar. Sci., 2, 115–143, doi:10.1146/annurev-marine-120408-151453.
Deser, C., and Coauthors, 2012: ENSO and Pacific decadal variability in Community Climate System Model version 4. J. Climate, 25, 2622–2651, doi:10.1175/JCLI-D-11-00301.1.
Durack, P. J., 2015: Ocean salinity and the global water cycle. Oceanography, 28, 20–31, doi:10.5670/oceanog.2015.03.
Durack, P. J., and S. E. Wijffels, 2010: Fifty-year trends in global ocean salinities and their relationship to broad-scale warming. J. Climate, 23, 4342–4362, doi:10.1175/2010JCLI3377.1.
Durack, P. J., S. E. Wijffels, and R. J. Matear, 2012: Ocean salinities reveal strong global water cycle intensification during 1950 to 2000. Science, 336, 455–458, doi:10.1126/science.1212222.
Durack, P. J., T. Lee, N. T. Vinogradova, and D. Stammer, 2016: Keeping the lights on for global ocean salinity observation. Nat. Climate Change, 6, 228–231, doi:10.1038/nclimate2946.
Emori, S., and S. J. Brown, 2005: Dynamic and thermodynamic changes in mean and extreme precipitation under changed climate. Geophys. Res. Lett., 32, L17706, doi:10.1029/2005GL023272.
England, M. H., and Coauthors, 2014: Recent intensification of wind-driven circulation in the Pacific and the ongoing warming hiatus. Nat. Climate Change, 4, 222–227, doi:10.1038/nclimate2106.
Folland, C. K., J. A. Renwick, M. J. Salinger, and A. B. Mullan, 2002: Relative influences of the interdecadal Pacific oscillation and ENSO on the South Pacific convergence zone. Geophys. Res. Lett., 29, 211–214, doi:10.1029/2001GL014201.
Forget, G., J.-M. Camping, P. Heimbach, C. N. Hill, R. M. Ponte, and C. Wunsch, 2015: ECCO version 4: An integrated framework for non-linear inverse modeling and global ocean state estimation. Geosci. Model Dev., 8, 3071–3104, doi:10.5194/gmd-8-3071-2015.
Ganachaud, A., and C. Wunsch, 2003: Large-scale ocean heat and freshwater transports during the World Ocean Circulation Experiment (WOCE). J. Climate, 16, 696–705, doi:10.1175/1520-0442(2003)016<0696:LSOHAF>2.0.CO;2.
Gaspar, P., Y. Gregoris, and J. M. Lefevre, 1990: A simple eddy kinetic energy model for simulations of the oceanic vertical mixing: Tests at station Papa and long-term upper ocean study site. J. Geophys. Res., 95, 16 179–16 193, doi:10.1029/JC095iC09p16179.
Gent, P. R., and J. C. McWilliams, 1990: Isopycnal mixing in ocean circulation models. J. Phys. Oceanogr., 20, 150–155, doi:10.1175/1520-0485(1990)020<0150:IMIOCM>2.0.CO;2.
Gierach, M. M., J. Vazquez-Cuervo, T. Lee, and V. M. Tsontos, 2013: Aquarius and SMOS detect effects of an extreme Mississippi River flooding event in the Gulf of Mexico. Geophys. Res. Lett., 40, 5188–5193, doi:10.1002/grl.50995.
Gimeno, L., and Coauthors, 2015: Oceanic and terrestrial sources of continental precipitation. Rev. Geophys., 50, RG4003, doi:10.1029/2012RG000389.
Giorgi, F., E.-S. Im, E. Coppola, N. S. Diffenbaugh, X. J. Gao, L. Mariotti, and Y. Shi, 2011: Higher hydroclimatic intensity with global warming. J. Climate, 24, 5309–5324, doi:10.1175/2011JCLI3979.1.
Gordon, A. L., 2001: Interocean exchange. Ocean Circulation and Climate: Observing and Modelling the Global Ocean, G. Siedler, J. Church, and J. Gould, Eds., Academic Press, 303–314.
Gordon, A. L., 2016: The marine hydrological cycle: The ocean’s floods and droughts. Geophys. Res. Lett., 43, 7649–7652, doi:10.1002/2016GL070279.
Gordon, A. L., C. F. Giulivi, J. Busecke, and F. M. Bingham, 2015: Differences among subtropical surface salinity patterns. Oceanography, 28, 32–39, doi:10.5670/oceanog.2015.02.
Greve, P., and S. I. Seneviratne, 2015: Assessment of future changes in water availability and aridity. Geophys. Res. Lett., 42, 5493–5499, doi:10.1002/2015GL064127.
Griffies, S., R. Pacanowski, M. Schmidt, and V. Balaji, 2001: Tracer conservation with an explicit free surface method for z-coordinate ocean models. Mon. Wea. Rev., 129, 1081–1098, doi:10.1175/1520-0493(2001)129<1081:TCWAEF>2.0.CO;2.
Grist, J. P., and S. A. Josey, 2003: Inverse analysis adjustment of the SOC air–sea flux climatology using ocean heat transport constraints. J. Climate, 16, 3274–3295, doi:10.1175/1520-0442(2003)016<3274:IAAOTS>2.0.CO;2.
Hackert, E., J. Ballabrera-Poy, A. J. Busalacchi, R.-H. Zhang, and R. Murtugudde, 2011: Impact of sea surface salinity assimilation on coupled forecasts in the tropical Pacific. J. Geophys. Res., 116, C05009, doi:10.1029/2010JC006708.
Häkkinen, S., and P. B. Rhines, 2004: Decline of subpolar North Atlantic circulation during the 1990s. Science, 304, 555–559, doi:10.1126/science.1094917.
Hasson, A., T. Delcroix, and R. Dussin, 2013: An assessment of the mixed layer salinity budget in the tropical Pacific Ocean: Observations and modelling (1990–2009). Ocean Dyn., 63, 179–194, doi:10.1007/s10236-013-0596-2.
Helm, K. P., N. L. Bindoff, and J. A. Church, 2010: Changes in the global hydrological-cycle inferred from ocean salinity. Geophys. Res. Lett., 37, L18701, doi:10.1029/2010GL044222.
Henley, B. J., J. Gergis, D. J. Karoly, S. B. Power, J. Kennedy, and C. K. Folland, 2015: A tripole index for the interdecadal Pacific oscillation. Climate Dyn., 45, 3077–3090, doi:10.1007/s00382-015-2525-1.
Hosoda, S., T. Suga, N. Shikama, and K. Mizuno, 2009: Global surface layer salinity change detected by Argo and its implication for hydrological cycle intensification. J. Oceanogr., 65, 579–586, doi:10.1007/s10872-009-0049-1.
Huang, R. X., 1993: Real freshwater flux as a natural boundary condition for the salinity balance and thermohaline circulation forced by evaporation and precipitation. J. Phys. Oceanogr., 23, 2428–2446, doi:10.1175/1520-0485(1993)023<2428:RFFAAN>2.0.CO;2.
Huffman, G. J., R. F. Adler, D. T. Bolvin, and G. Gu, 2009: Improving the global precipitation record: GPCP version 2.1. Geophys. Res. Lett., 36, L17808, doi:10.1029/2009GL040000.
Huntington, T. G., 2006: Evidence for intensification of global water cycle: Review and synthesis. J. Hydrol., 319, 83–95, doi:10.1016/j.jhydrol.2005.07.003.
IPCC, 2013: Climate Change 2013: The Physical Science Basis. Cambridge University Press, 1535 pp., doi:10.1017/CBO9781107415324.004.
Josey, S. A., S. Gulev, and L. Yu, 2013: Exchanges through the ocean surface. Ocean Circulation and Climate: A 21st Century Perspective, G. Siedler et al., Eds., International Geophysics Series, Vol. 103, Academic Press, 115–140, doi:10.1016/B978-0-12-391851-2.00005-2.
Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437–471, doi:10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2.
Kao, H.-Y., and G. S. E. Lagerloef, 2015: Salinity fronts in the tropical Pacific Ocean. J. Geophys. Res. Oceans, 120, 1096–1106, doi:10.1002/2014JC010114.
Katsman, C. A., and G. J. van Oldenborgh, 2011: Tracing the upper oceans “missing heat.” Geophys. Res. Lett., 38, L14610, doi:10.1029/2011GL048417.
Kaufmann, R. K., H. Kauppib, M. L. Mann, and J. H. Stock, 2011: Reconciling anthropogenic climate change with observed temperature 1998–2008. Proc. Natl. Acad. Sci. USA, 108, 11 790–11 793, doi:10.1073/pnas.1102467108.
Kim, S.-B., I. Fukumori, and T. Lee, 2006: The closure of the ocean mixed layer temperature budget using level-coordinate model fields. J. Atmos. Oceanic Technol., 23, 840–853, doi:10.1175/JTECH1883.1.
Köhl, A., M. S. Martins, and D. Stammer, 2014: Impact of assimilating surface salinity from SMOS on ocean circulation estimates. J. Geophys. Res. Oceans, 119, 5449–5464, doi:10.1002/2014JC010040.
Lagerloef, G., 1995: Interdecadal variations in the Alaska Gyre. J. Phys. Oceanogr., 25, 2242–2258, doi:10.1175/1520-0485(1995)025<2242:IVITAG>2.0.CO;2.
Lagerloef, G., 2012: Satellite mission monitors ocean surface salinity. Eos, Trans. Amer. Geophys. Union, 93, 233–234, doi:10.1029/2012EO250001.
Lago, V. S., S. Wijffels, P. Durack, J. Church, N. Bindoff, and S. Marsland, 2015: Simulating the role of surface forcing on observed multidecadal upper-ocean salinity changes. J. Climate, 29, 5575–5588, doi:10.1175/JCLI-D-15-0519.1.
Lee, T., G. Lagerloef, M. M. Gierach, H.-Y. Kao, S. Yueh, and K. Dohan, 2012: Aquarius reveals salinity structure of tropical instability waves. Geophys. Res. Lett., 39, L12610, doi:10.1029/2012GL052232.
L’Heureux, M. L., S. Lee, and B. Lyon, 2013: Recent multidecadal strengthening of the Walker circulation across the tropical Pacific. Nat. Climate Change, 3, 571–576, doi:10.1038/nclimate1840.
Li, L., R. W. Schmitt, C. C. Ummenhofer, and K. B. Karnauskas, 2016: Implications of North Atlantic sea surface salinity for summer precipitation over the U.S. Midwest: Mechanisms and predictive value. J. Climate, 29, 3143–3159, doi:10.1175/JCLI-D-15-0520.1.
Lindstrom, E., F. Bryan, and R. Schmitt, 2015: SPURS: Salinity Processes in the Upper-Ocean Regional Study—the North Atlantic experiment. Oceanography, 28, 14–19, doi:10.5670/oceanog.2015.01.
Liu, S. C., C. Fu, C. J. Shiu, J. P. Chen, and F. Wu, 2009: Temperature dependence of global precipitation extremes. Geophys. Res. Lett., 36, L17702, doi:10.1029/2009GL040218.
Liu, W. T., and W. Tang, 2005: Estimating moisture transport over oceans using space-based observations. J. Geophys. Res., 110, D10101, doi:10.1029/2004JD005300.
Mantua, N. J., S. R. Hare, Y. Zhang, J. M. Wallace, and R. C. Francis, 1997: A Pacific interdecadal oscillation with impacts on salmon production. Bull. Amer. Meteor. Soc., 78, 1069–1079, doi:10.1175/1520-0477(1997)078<1069:APICOW>2.0.CO;2.
Marshall, J. C., C. Hill, L. Perelman, and A. Adcroft, 1997: Hydrostatic, quasi-hydrostatic and non-hydrostatic ocean modeling. J. Geophys. Res., 102, 5733–5752, doi:10.1029/96JC02776.
McGregor, S., A. S. Gupta, and M. H. England, 2012: Constraining wind stress products with sea surface height observations and implications for Pacific Ocean sea level trend attribution. J. Climate, 25, 8164–8176, doi:10.1175/JCLI-D-12-00105.1.
Meehl, G. A., J. M. Arblaster, J. Y. Fasullo, A. Hu, and K. E. Trenberth, 2011: Model-based evidence of deep-ocean heat uptake during surface-temperature hiatus periods. Nat. Climate Change, 1, 360–364, doi:10.1038/nclimate1229.
Mieruch, S., M. Schröder, S. Noël, and J. Schulz, 2014: Comparison of decadal global water vapor changes derived from independent satellite time series. J. Geophys. Res. Atmos., 119, 12 489–12 499, doi:10.1002/2014JD021588.
Overland, J. E., S. Salo, and J. M. Adams, 1999: Salinity signature of the Pacific decadal oscillation. Geophys. Res. Lett., 26, 1337–1340, doi:10.1029/1999GL900241.
Pierce, D. W., P. J. Gleckler, T. P. Barnett, B. D. Santer, and P. J. Durack, 2012: The fingerprint of human-induced changes in the ocean’s salinity and temperature fields. Geophys. Res. Lett., 39, L21704, doi:10.1029/2012GL053389.
Ponte, R. M., and N. T. Vinogradova, 2016: An assessment of basic processes controlling mean surface salinity over the global ocean. Geophys. Res. Lett., 43, 7052–7058, doi:10.1002/2016GL069857.
Qu, T. D., and J. Y. Yu, 2014: ENSO indices from sea surface salinity observed by Aquarius and Argo. J. Oceanogr., 70, 367–375, doi:10.1007/s10872-014-0238-4.
Qu, T. D., S. Gao, and I. Fukumori, 2011: What governs the North Atlantic salinity maximum in a global GCM? Geophys. Res. Lett., 38, L07602, doi:10.1029/2011GL046757.
Redi, M. H., 1982: Oceanic isopycnal mixing by coordinate rotation. J. Phys. Oceanogr., 12, 1154–1158, doi:10.1175/1520-0485(1982)012<1154:OIMBCR>2.0.CO;2.
Reul, N., and Coauthors, 2014: Sea surface salinity observations from space with the SMOS satellite: A new means to monitor the marine branch of the water cycle. Surv. Geophys., 35, 681–722, doi:10.1007/s10712-013-9244-0.
Rodell, M., and Coauthors, 2015: The observed state of the water cycle in the early twenty-first century. J. Climate, 28, 8289–8318, doi:10.1175/JCLI-D-14-00555.1.
Roemmich, D., and Coauthors, 2009: Argo: The challenge of continuing 10 years of progress. Oceanography, 22, 46–55, doi:10.5670/oceanog.2009.65.
Roullet, G., and G. Madec, 2000: Salt conservation, free surface, and varying levels: A new formulation for ocean general circulation models. J. Geophys. Res., 105, 23 927–23 942, doi:10.1029/2000JC900089.
Salisbury, J., and Coauthors, 2015: How can present and future satellite missions support scientific studies that address ocean acidification? Oceanography, 28, 108–121, doi:10.5670/oceanog.2015.35.
Sarafanov, A., A. Falina, A. Sokov, and A. Demidov, 2008: Intense warming and salinification of intermediate waters of southern origin in the eastern subpolar North Atlantic in the 1990s to mid-2000s. J. Geophys. Res., 113, C12022, doi:10.1029/2008JC004975.
Schanze, J. J., R. W. Schmitt, and L. L. Yu, 2010: The global oceanic freshwater cycle: A state-of-the-art quantification. J. Mar. Res., 68, 569–595, doi:10.1357/002224010794657164.
Schmitt, R. W., 2008: Salinity and the global water cycle. Oceanography, 21, 12–19, doi:10.5670/oceanog.2008.63.
Seidov, D., and B. J. Haupt, 2002: On the role of inter-basin surface salinity contrasts in global ocean circulation. Geophys. Res. Lett., 29, 1800, doi:10.1029/2002GL014813.
Skliris, N., R. Marsh, S. A. Josey, S. A. Good, C. Liu, and R. P. Allan, 2014: Salinity changes in the World Ocean since 1950 in relation to changing surface freshwater fluxes. Climate Dyn., 43, 709–736, doi:10.1007/s00382-014-2131-7.
Stammer, D., K. Ueyoshio, A. Köhl, W. G. Large, S. A. Josey, and C. Wunsch, 2004: Estimating air-sea fluxes of heat, freshwater, and momentum through global ocean data assimilation. J. Geophys. Res., 109, C05023, doi:10.1029/2003JC002082.
Terray, L., L. Corre, S. Cravatte, T. Delcroix, G. Reverdin, and A. Ribes, 2012: Near-surface salinity as nature’s rain gauge to detect human influence on the tropical water cycle. J. Climate, 25, 958–977, doi:10.1175/JCLI-D-10-05025.1.
Trenberth, K. E., 2011: Changes in precipitation with climate change. Climate Res., 47, 123–138, doi:10.3354/cr00953.
Trenberth, K. E., and J. T. Fasullo, 2013: An apparent hiatus in global warming? Earth’s Future, 1, 19–32, doi:10.1002/2013EF000165.
Trenberth, K. E., and G. R. Asrar, 2014: Challenges and opportunities in water cycle research: WCRP contributions. Surv. Geophys., 35, 515–532, doi:10.1007/s10712-012-9214-y.
Trenberth, K. E., A. Dai, R. M. Rasmussen, and D. B. Parsons, 2003: The changing character of precipitation. Bull. Amer. Meteor. Soc., 84, 1205–1217, doi:10.1175/BAMS-84-9-1205.
Trenberth, K. E., J. Fasullo, and L. Smith, 2005: Trends and variability in column-integrated atmospheric water vapor. Climate Dyn., 24, 741–758, doi:10.1007/s00382-005-0017-4.
Vinogradova, N. T., and R. M. Ponte, 2013: Clarifying the link between surface salinity and freshwater fluxes on climate scales. J. Geophys. Res. Oceans, 118, 3190–3201, doi:10.1002/jgrc.20200.
Vinogradova, N. T., T. Lee, P. J. Durack, J. Boutin, and D. Stammer, 2016: Ocean salinity and the water cycle: Recent progress and future challenges. GEWEX News, No. 28, International GEWEX Project Office, Silver Spring, MD, 6–8.
Wang, L., J. Huang, Y. He, and Y. Guan, 2014: Combined effects of the Pacific decadal oscillation and El Niño-Southern Oscillation on global land dry–wet changes. Sci. Rep., 4, 6651, doi:10.1038/srep06651.
Wijffels, S. E., R. W. Schmitt, H. L. Bryden, and A. Stigebrandt, 1992: Transport of freshwater by the oceans. J. Phys. Oceanogr., 22, 155–162, doi:10.1175/1520-0485(1992)022<0155:TOFBTO>2.0.CO;2.
Willett, K. W., N. P. Gillett, P. D. Jones, and P. W. Thorne, 2007: Attribution of observed humidity changes to human influence. Nature, 449, 710–721, doi:10.1038/nature06207.
Wu, L., and Coauthors, 2012: Enhanced warming over the global subtropical western boundary currents. Nat. Climate Change, 2, 161–166, doi:10.1038/nclimate1353.
Wunsch, C., 2006: Discrete Inverse and State Estimation Problems: With Geophysical Fluid Applications. Cambridge University Press, 384 pp.
Wunsch, C., and P. Heimbach, 2007: Practical global oceanic state estimation. Physica D, 230, 197–208, doi:10.1016/j.physd.2006.09.040.
Wunsch, C., P. Heimbach, R. M. Ponte, and I. Fukumori, 2009: The global general circulation of the ocean estimated by the ECCO-consortium. Oceanography, 22, 88–103, doi:10.5670/oceanog.2009.41.
Xie, P., and P. A. Arkin, 1997: Global precipitation: A 17-year monthly analysis based on gauge observations, satellite estimates, and numerical model outputs. Bull. Amer. Meteor. Soc., 78, 2539–2558, doi:10.1175/1520-0477(1997)078<2539:GPAYMA>2.0.CO;2.
Yu, L., 2011: A global relationship between the water cycle and near-surface salinity. J. Geophys. Res., 116, C10025, doi:10.1029/2010JC006937.
Yu, L., and R. A. Weller, 2007: Objectively analyzed air–sea heat fluxes for the global ice-free oceans (1981–2005). Bull. Amer. Meteor. Soc., 88, 527–539, doi:10.1175/BAMS-88-4-527.
Yu, L., X. Jin, S. Josey, T. Lee, A. Kumar, C. Wen, and Y. Xue, 2017: The global ocean water cycle in atmospheric reanalysis, satellite, and ocean salinity. J. Climate, 30, 3829–3852, doi:10.1175/JCLI-D-16-0479.1
Zhang, X. B., F. W. Zwiers, G. C. Hegerls, F. H. Lambert, N. P. Gillett, S. Solomon, P. A. Scott, and T. Nozawa, 2007: Detection of human influence on twentieth-century precipitation. Nature, 448, 461–464, doi:10.1038/nature06025.
Zhu, J., B. Huang, R.-H. Zhang, Z.-Z. Hu, A. Kumar, M. A. Balmaseda, L. Marx, and J. L. Kinter, 2014: Salinity anomaly as a trigger for ENSO events. Sci. Rep., 4, 6821, doi:10.1038/srep06821.
