Appendix E
Panel Presentation Abstracts

Tropical Pacific decadal variability and the global warming hiatus:

Shang-Ping Xie, Scripps Institution of Oceanography, UC San Diego

Global mean surface temperature (GMST) is known to rise following a major El Niño event. The tropical Pacific cooling that began in the late 1990s emerged as the leading mechanism for the slowdown of the GMST increase for the recent 15 years. An important question is how we can test this hypothesis among other mechanisms for the global warming hiatus. Much attention has been given to the annual mean GMST, but it is too narrow a focus to quantify the relative importance of the zoo of mechanisms.

We need to go beyond the annual mean GMST by unpacking it into seasonal and spatial dimensions and develop distinctive fingerprints of these various mechanisms. The pacemaker experiments with a GFDL climate model reveal the following fingerprints of the tropical Pacific cooling on the recent hiatus:

• The seasonal contrast between the GMST decrease in boreal winter and increase in summer;
• The decadal droughts over the Southwest U. S. (including California and Texas) for the past 15 years.

We show that the seasonal fingerprint is present in all the GMST datasets including the one recently released from NOAA.

We also need to develop metrics that distinguish forced change and internal variability. For example, planetary/ocean heat uptake is an important aspect of the transient climate response to anthropogenic radiative forcing, but is it also an essential element of internal decadal variability as is widely assumed in hiatus studies? Modeling studies suggest that the answer is probably no. This has important implications for observations.

Comparing simulated and observed and decadal trends:

John Fyfe, CCCma

Should it be done, and if so how do we separate, quantify and communicate the influences of uncertainty (model, forcing and observational) and internal variability? This question will be considered in the context of decadal trends in Pacific SST, GMST and Arctic sea ice extent. I'll touch on the Karl et al. result, and finish with a forecast for the end of the current GMST hiatus.

Tropical Pacific decadal variability: Oceanic processes and the possible important role of climate noise:

Antonietta Capotondi, University of Colorado/CIRES and NOAA/ESRL/Physical Sciences Division

In this talk I will start by reviewing some of the mechanisms proposed for tropical Pacific decadal variability, with emphasis on oceanic processes. Focus will be on the 1976/77 climate shift, as an example. I will then discuss the possible influence of the slowly varying mean tropical climate state upon ENSO characteristics, and discuss some of the proposed

theories for the resulting decadal ENSO modulation. Finally, using a Linear Inverse Modeling (LIM) approach I will show that apparent changes in ENSO characteristics over decadal periods are within the expected range of noise-driven variations. Thus, we cannot reject the null hypothesis that decadal ENSO modulation may merely result from sampling variability, with important implications for predictability.

Pacific decadal climate variability: Phenomenon, evidence, and impacts

Yochanan Kushnir, Lamont-Doherty Earth Observatory, Columbia University

The concept of Pacific Decadal Climate Variability (PDV) was introduced in a series of high-visibility articles during 1990s (though J. Namias already discussed evidence for the existence of such low-frequency behavior in 1978. The phenomenon (initially referred to as the Pacific Decadal Oscillation—PDO) was identified when studying climate variability in the North Pacific and contrasting it with the strong interannual variability (ENSO) in the tropics. The PDV was found connected with important environmental impacts in the countries surrounding the Pacific Basin and with changes in ocean circulation patterns and ocean biology. The PDV however also affects the tropics as a slow and relatively small (compared to ENSO) fluctuation in the tropical Pacific east west SST gradient (referred to as the Inter-Decadal Pacific Oscillation, IPO), consistent variations in the strength of the trade winds, and consequently changes in convection and precipitation patterns. These changes in surface variables and tropical diabatic heating gradients make PDV an important forcing agent of a global climate dynamical response. Broadly speaking, the PDV is considered as an internally driven natural mode of variability though it may also be invoked by slow changes in external forcing. It is not fully understood whether there is a single unique form of PDV. Also, because of the relatively short instrumental record it’s not clear what the time scale of PDV is, if there is a distinct one, and what controls this time scale. Moreover, already early after it was defined, the PDV was identified as associated with what appeared to be a perplexing, distinct rapid shift (around 1976) in North Pacific sea level pressure, winds, ocean temperatures and ocean currents. Paleoclimate proxies provide useful information in better characterizing the time scale and spatial pattern of PDV.

Robust and non-robust aspects of AMOC intrinsic variability and mechanisms in the Community Earth System Model (CESM)

Gokhan Danabasoglu, National Center for Atmospheric Research

Atlantic Meridional Overturning Circulation (AMOC) is presumed to play a major role in decadal and longer time scale climate variability and in prediction of the earth’s future climate on these time scales. The primary support for such a prominent role for AMOC comes from coupled model simulations. They show rich AMOC variability, but time scales of variability and mechanisms differ substantially among models. A topic that remains largely unexplored is the role that an ocean model’s subgrid scale parameterizations play in AMOC intrinsic variability. Here, we present an assessment of the impacts of several, loosely-constrained ocean model parameter choices on AMOC characteristics in CESM with the primary goal of identifying both robust and non-robust elements of AMOC variability and mechanisms. Specifically, we change parameter values in mesoscale, submesoscale, vertical mixing, and lateral viscosity parameterizations in the ocean model. The characteristics of AMOC from these simulations are then compared with a three-member ensemble of experiments in which the initial atmospheric temperature field is slightly perturbed. We find that both the amplitude and time scale of AMOC variability differ considerably among all these experiments with dominant time scales of variability ranging from decadal to centennial. There are also substantial differences in the relative contributions of temperature and salinity anomalies to the positive density anomalies created in the model’s deep-water formation (DWF) region prior to AMOC intensifications.

Nevertheless, we identify some robust elements of AMOC variability mechanisms. These include: i) The Labrador Sea is the key region with upper-ocean density and boundary layer anomalies preceding AMOC anomalies; ii) Enhanced Nordic Sea overflow transports do not lead to an increase in AMOC maximum transports; iii) Persistent positive phase of the North Atlantic Oscillation plays a significant role in setting up the density anomalies that lead to AMOC intensification via surface buoyancy fluxes; and iv) After AMOC intensification, subsequent weakening is due to advection of positive temperature anomalies into the model’s DWF region.

Understanding tropical Atlantic decadal variability: The role of tropical Pacific versus subpolar Atlantic:

Mingfang Ting, Lamont-Doherty Earth Observatory, Columbia University

The Atlantic Multidecadal Variability (AMV) has been shown to affect precipitation globally. In particular, the frequency and severity of droughts across North America has been modulated by the phase of the Atlantic Multidecadal Variability (AMV) over the historical period. The decadal oscillations in U. S. West hydroclimate (associated with ENSO) reach extreme severity during the warm and neutral phases of AMV, such as in the 1930s and the 1950s when the U. S. Great Plains and the Southwest experienced the extremely dry conditions of the Dust Bowl and the persistent Texas drought, respectively. When AMV was in its cold phase in the early 1900s and from 1965 to 1995 droughts were less frequent or severe. The hydroclimate impacts of AMV are believed to be dominated by its tropical component through changes in tropical convection and related circulation changes.

This study explores the inter-connection between the tropical Pacific and North Atlantic using both available historical observations and the Climate Model Intercomparison Project Phase 5 (CMIP5) climate models. The interconnection between the tropical Pacific and the tropical Atlantic on decadal time scale is found to be crucial in realistically representing the hydroclimate impacts of the AMV on North America. We found that decadal ENSO variability plays a more dominant role in CMIP5 models compared to observations in causing the decadal tropical Atlantic SST anomalies. Depending on how decadal tropical Atlantic SST anomalies are generated in CMIP5 models, whether it is dominated by ENSO conditions in the tropical Pacific or subpolar SST anomalies, the warm AMV-dry North America relationship as observed can be severely underestimated in models. By examining how the tropical component of the AMV is generated, it provides a useful metric for evaluating the realism of the model AMV as well as understanding its physical mechanisms.

The impact of the North Atlantic Oscillation on climate through its influence on the Atlantic Meridional Overturning Circulation:

Tom Delworth, GFDL

Prominent multidecadal climate variations have been observed over the Atlantic and Arctic oceans and surrounding continents over the last 130+ years. Here we use climate model simulations to explore the possible role of multidecadal variations of the North Atlantic Oscillation (NAO) for this observed variability through its effect on the Atlantic Meridional Overturning Circulation (AMOC. Perturbation experiments are conducted in which patterns of anomalous fluxes corresponding to the NAO are added to the model ocean; in companion experiments no such fluxes are added. Differences between the experiments illustrate how the model ocean and climate system respond to the NAO. A positive phase of the NAO tends to strengthen the AMOC by extracting heat from the subpolar gyre, thereby increasing deepwater formation, horizontal density gradients, and the AMOC.

The flux forcings have the spatial structure of the observed NAO, but the amplitude of the forcing varies in time. The temporal variation of the imposed fluxes is one of the following types: (a) sudden switch on of the flux forcing, (b) vary the amplitude of the flux forcing sinusoidally in time with distinct periods varying from 2 to 200 years, (c) vary the flux forcing to match the observed time sequence of the NAO over the 20th and early 21st centuries. In the idealized experiments we show that the response of the AMOC to NAO variations is small at short time scales, but increases up to the dominant time scale of internal AMOC variability (20-30 years for the models used. The amplitude of the response of the AMOC, and associated oceanic heat transport, is approximately constant as the time scale of the forcing is increased further. In contrast, the response of other properties, such as hemispheric surface air temperature or Arctic sea ice, continues to increase as the time scale of the forcing becomes progressively longer. The larger response of temperature and sea ice is associated with an increased impact of radiative feedback processes at progressively longer time scales. The impact of the NAO on the AMOC and climate is a function of the dominant time scale of internal AMOC variability, as well as the background mean state. In the experiments using the observed sequence of the NAO we estimate the contribution of NAO-induced AMOC anomalies to climate variations in the 20th and early 21st centuries. We show that NAO-induced AMOC variations may have contributed substantially to multidecadal warming and cooling of the Northern Hemisphere, including cooling from the 1960s through the 1980s, and warming from the 1980s through the 2000s. We further show that such NAO-induced AMOC variations could have contributed to the observed reduction of sea ice in the 1990s and 2000s, as well as a possible remote influence on the Southern Ocean, including sea ice.

Predictability of the recent slowdown and subsequent recovery of large-scale surface warming using statistical methods

Michael E. Mann, Pennsylvania State University

The recent, temporary slowdown in large-scale surface warming has been attributed to both external and internal sources of climate variability. Using semi-empirical estimates of the internal low-frequency variability component in Atlantic, Pacific, and Northern Hemisphere surface temperature in concert with statistical hindcast experiments, we investigate whether the slowdown and its recent recovery were predictable in advance, and conclude that they likely were not. The internal variability of the North Pacific, which played a critical role in the slowdown, does not appear to be predictable in advance using statistical forecast methods. An additional minor contribution from the North Atlantic, by contrast, appears to exhibit some predictability. While our analyses focus on combining semi-empirical estimates of internal climatic variability with statistical hindcast experiments, some possible implications for initialized predictions are also discussed.

Decadal variability in Pacific trade winds inferred from coral Mn/Ca: Implications for the rate of global warming

Diane M. Thompson

Decadal variations in zonal wind strength and direction may play an important role in modulating the El Niño-Southern Oscillation (ENSO) and the rate of global temperature rise. However, historical observations of tropical Pacific winds are limited, and existing datasets disagree on long-term trends, emphasizing the need for independent data to assess zonal wind variability. Earlier work suggested that the ratio of manganese to calcium in corals from islands with westerly facing lagoons may record westerly winds associated with the onset and maintenance of El Niño events. These westerly wind anomalies trigger strong physical mixing and release of Mn from the Mn-enriched lagoonal sediments, which is incorporated into the coral skeleton. Here I present a new ~90 year Mn/Ca record from

Tarawa that provides further support for the link between the frequency of westerly winds and coral Mn/Ca. This new Mn/Ca record provides a means to assess westerly wind anomalies before the mid-20th century, when instrumental data from the tropical Pacific are scarce. Along with a Sr/Ca-SST reconstruction from the eastern tropical Pacific, this wind reconstruction corroborates and extends the idea, developed from models and analyses of the well-observed late 20th century, that periods of strong Pacific trade winds are associated with cooler equatorial Pacific SSTs and a slower rate of global warming, and vice versa. By adding Mn/Ca to the suite of coral tracers measured for paleoclimate reconstructions from appropriate sites, we can expand our view of past climate variability to include westerly winds, along with the more commonly reconstructed variables of SST and salinity. Development of additional Mn/Ca records from other equatorial atolls with westerly facing lagoons will be used to obtain a broader multivariate perspective on the dynamics of recent decadal climate variability.

Paleo-constraints on decadal climate variability in the tropical Pacific

Kim Cobb, Georgia Tech

The tropical Pacific is a prominent source of decadal-scale global climate variability, with a variety of coupled ocean-atmosphere dynamical processes giving rise to the Pacific Decadal Oscillation (PDO; Mantua et al., 1997) and the North Pacific Gyre Oscillation (NPGO; Di Lorenzo et al., 2008. Indeed, Pacific decadal variability has been implicated in the observed slow-down of global surface temperature over the last decade (Kosaka and Xie, 2013; England et al., 2014; Nieves et al., 2015), which is turn may be linked to the magnitude and spatial footprint of recent ENSO extremes (e. g. McPhaden and McClurg, 2011. Here we assess the characteristics of Pacific decadal variability over the last millennium using coral paleoclimate records of SST and hydrology, and compare these records to the evolution of 20th century Pacific decadal variability, with an eye towards isolating potential anthropogenic trends in Pacific climate.

Radiative forcing contributions to changes in recent rates of global warming:

Susan Solomon, Massachusetts Institute of Technology

This talk will briefly survey what is known and what is not known about radiative forcing changes during the period from 2000-2014, and will summarize how these can contribute to the decadal rates of global warming. In addition to greenhouse gases, changes in volcanic aerosol impacts, solar forcing, stratospheric water vapor and tropospheric aerosols will be discussed. Implications for future observational needs will be briefly described.

How long could the current hiatus in global warming last?

Thomas R. Knutson, NOAA GFDL

Global mean temperature did not rise steadily since the late 1800s but rose primarily during two rapid warming periods (early 20th century and late 20th century) which were separated by a pause in warming from about 1940-1970. Could another such multidecadal pause occur at the beginning of the 21st century, and if so by what processes could this occur? At one extreme, the current global warming “hiatus” could end shortly (or may have already ended. However, at the other extreme we ask: How long could the current hiatus in global warming potentially last? To explore this issue, we analyze the internal multidecadal variability of global mean temperature in the GFDL CM3 model control run and test the potential influence of such internal variability on 21st century global mean temperature evolution, including current projections of future warming from anthropogenic forcings (e. g., CMIP5 models. We also explore the plausibility of CM3’s multidecadal variability based on comparisons with historical trends.

Impact of data coverage and quality control on global surface temperature trends: Part 1—Overview and sea surface temperature aspects

Huai-Min Zhang, NOAA National Centers for Environmental Information (NCEI)

The recent paper by Karl et al. (2015) highlighted the importance of data homogenization and bias correction in resolving the so-called global “Warming Hiatus” from observational analyses. In this talk we present the details of these impacts on the global and regional surface temperature trends in various time scales. The impacts are studied using the data quality control and bias correction processes used in the NOAA’s centennial time scale sea surface temperature (SST) products, as well as the data gaps in the majorly available international datasets. Additional analysis of subsurface observations, mainly obtained by the Argo floats in recent decades, also shows continued warming over previous decades. Lastly, we analyze the consistency and discrepancy of satellite and in-situ based SSTs since the early 1980s when satellite data became available, and clarify their utilization limitations in determining the trends and other variabilities such as El Niño signals.

Impact of data coverage and quality control on global surface temperature trends: Part 2—Land surface air temperature aspects

Matt Menne, NOAA National Centers for Environmental Information (NCEI)

Land surface temperature air temperature (LSAT) records have been compiled from a variety of sources over the past few decades. Here we discuss the recent effort to improve land surface station temperature data holdings known as the International Surface Temperature Initiative (ISTI) and how these holdings are being used to produce a new NOAA analysis of land surface air temperature since the late 19th Century. A comparison of this latest analysis to other datasets will be discussed as well as efforts to extend global surface air temperature analysis over the Arctic Ocean.

Pacific temporarily hid heat below surface

Veronica Nieves, JPL

The recent hiatus in global warming was caused by a sequestration of heat in the subsurface tropical Pacific waters and was symptomatic of decadal variability. This natural variability is superimposed on the long-term human-caused warming trend, and dominates on a decadal time scale with large regional societal impacts. Heat traveled west in the subsurface 100-300 m depth layer (from the eastern Pacific to the central/western Pacific and Indian Ocean) due to unusually strong trade winds during the early 21st century. The important question is whether the trapped heat will move up to the surface when the Pacific changes to a warm phase or will it be absorbed into the deeper layers of the ocean in the next decade or two. If it mixes down, the significant unknown is how rapidly it will be vertically mixed into the ocean and how it will moderate global temperatures.

Understanding decadal climate variability using formal model-data synthesis

Patrick Heimbach, UT Austin

Formal model-data synthesis (loosely termed data assimilation) seeks to optimally combine information contained in observations from heterogeneous (and sparse) data streams and models that obey known conservation laws exactly. Different techniques lead to different pitfalls in the use of these products. In particular, so-called ocean reanalyses, like atmospheric reanalyses do not conserve properties over time, in particular heat and freshwater, thus rendering their use for assessing decadal changes in properties problematic. After illustrating the issue, we present results from a global bidecadal (1992-2011) dynamically consistent ocean state estimate, with an emphasis on global heat

content changes and vertical redistribution of heat. Both show large lateral and vertical variations. Net vertical cooling at depth may be an expression of long term oceanic memory processes. We discuss challenges for designing an observing system suitable for understanding decadal climate variability and future requirements for estimation systems.

Consequences of uncertainty in air-sea exchange

Baylor Fox-Kemper, Brown University

The heat capacity of the ocean greatly exceeds that of the atmosphere, which leads to significant exchanges and variability of the coupled system on seasonal and longer timescales. I will describe some of the key processes in the air-sea exchange, emphasizing in particular those processes which are poorly observed and modeled—due to their intermittency and small scale—and insufficiently understood to be parameterized. I will then estimate their cumulative effect on the global heat budget and surface temperature, emphasizing the decadal and longer timescales.

The ocean’s role in polar climate change: asymmetric Arctic and Antarctic responses to greenhouse gas and ozone forcing.

John Marshall, Massachusetts Institute of Technology

In recent decades, the Arctic has been warming and sea ice disappearing. By contrast, the Southern Ocean around Antarctica has been (mainly) cooling and sea-ice extent growing. We argue here that inter-hemispheric asymmetries in the mean ocean circulation, with sinking in the northern North Atlantic and upwelling around Antarctica, strongly influence the sea-surface temperature (SST) response to anthropogenic greenhouse gas (GHG) forcing, accelerating warming in the Arctic while delaying it in the Antarctic. Furthermore, while the amplitude of GHG forcing has been similar at the poles, significant ozone depletion only occurs over Antarctica. We suggest that the initial response of SST around Antarctica to ozone depletion is one of cooling and only later adds to the GHG-induced warming trend as upwelling of sub-surface warm water associated with stronger surface westerlies impacts surface properties.

Arctic changes and mid-latitude weather linkages in the coming decades:

James Overland, NOAA/Pacific Marine Environmental Laboratory, Seattle, WA

Ongoing temperature changes in the Arctic are large relative to lower latitudes; a process known as Arctic Amplification. Arctic temperatures have increased 2-3 times the rate of mid-latitude temperatures relative to the late 20th century, due to multiple interacting feedbacks driven by modest global change. Even if global temperature increases are contained to +2 C by 2040, Arctic (North of 60 N) monthly mean temperatures in fall will increase by +5 C. The Arctic is very likely to be sea ice free during summer before 2040 and snow cover will be absent in May and June on most land masses. Thus for the next few decades out to 2040, continuing rapid environmental changes in the Arctic are very likely, despite mitigation activities, and the appropriate response is to plan for adaptation to meet mean and extreme event changes. Mitigation is essential to forestall further disasters in the second half of the century. Whether these changes impact mid-latitude extreme weather events is complex and controversial, as the time period for observing such linkages is short [<10 years] and involves understanding direct forcing by Arctic changes on a chaotic climatic system. There is general agreement that there will be no net mid-latitude cooling, only a potential for impacting severe events. Linkages will be regional, episodic, and based on amplification of existing weather patterns such as Greenland atmospheric blocking and the Siberian High. It is important to note such future rapid Arctic amplification and the potential for environmental surprises.

Indian Ocean variability and its impact on regional climate

Caroline C. Ummenhofer, Woods Hole Oceanographic Institution

The Indian Ocean has sustained robust surface warming in recent decades, with warming rates exceeding those of other tropical ocean basins. However, it remains unclear how multi-decadal variability in upper-ocean thermal characteristics has contributed to these Indian Ocean trends. Temperatures and heat content exhibit extensive subsurface cooling for much of the tropical Indian Ocean since the 1950s, likely due to remote Pacific wind changes associated with the Interdecadal Pacific Oscillation/Pacific Decadal Oscillation. As such, multi-decadal wind forcing has masked increases in Indian Ocean heat content due to thermal forcing since the 1960s. However, wind and thermal forcing both contribute positively to Indian Ocean heat content since the turn of the century. Drastic increases in the heat content in coming decades are therefore likely; in fact, they have been implicated to play a role in the recent warming hiatus.

Multi-decadal variability in Indian Ocean characteristics has implications for regional climate: strength of the Austral-Asian monsoon system, regional hydroclimate, sea-level variations, and marine ecosystems are modulated by Indian Ocean variability. Better decadal predictions of Indian Ocean properties are therefore likely of considerable benefit to vulnerable societies in Indian Ocean rim-countries.

Science is not finished until it is communicated

Susan Hassol, Climate Communication

The enormous societal implications of climate science make effective communication essential. Deeply ingrained misconceptions and decades of disinformation make this more challenging, particularly when communicating about the complex topic of decadal climate variability. Providing context for this topic by reiterating what is known about recent climate change can help avoid people becoming confused or misled by details and uncertainties. This discussion will focus on ways to provide the needed context and communicate what is known about both human-induced climate change and natural decadal climate variability in simple, clear terms. Coming at the end of the workshop, it will be informed by the presentations and discussions of the latest science.