Session B62: Detecting Signals in a Noisy Climate System

Insights Regarding Observational Requirements For Climate Change Signal Detection

Climate change signals must be detected against both the noise of internal natural variability as well as a wide range of uncertainties in the observing system. Observing system uncertainties include SI traceable accuracy, instrument stability over a decade or longer, calibration across data gaps, changing instrument designs, changing sampling of the earth, and changing data analysis methods. Most current observations used for climate change signals were not designed with climate change uncertainties in mind (National Academy of Sciences Continuity Report for NASA, Nov. 2015). Examples are given of how to set requirements for climate change observations relative to anticipated climate change signals and climate system natural variability. Given that we currently lack an observing system designed specifically for climate change, what would such a system look like? Design principles are provided, and examples are given of how more accurate observations can narrow key scientific uncertainties in climate change. Examples will also be presented of observations that can meet these much more challenging climate change requirements. The lack of a designed climate observing system naturally raises the question of the societal return on investment of providing such a system. Recent published estimates using state of the art economic integrated assessment models (IAMs) suggest a return of ~ $50 per $1 invested. Society will be managing the Earth’s Climate System actively or passively, wisely or unwisely, for the indefinite future. The world has had an internationally designed, committed, and shared weather observing system for many decades. It is time to begin an equivalent international climate change observing system. The motivation for such systems are the same: a better future for society.   

ENSO Change in Climate Projections: Forced Response or Internal Variability?

The El Niño-Southern Oscillation (ENSO) is the dominant driver of interannual variability globally and has effects which are felt in many remote regions of the world. As such it is vital to assess the potential future changes of ENSO. However, there is little consensus on how ENSO sea surface temperature (SST) may change in a future with increasing greenhouse gas emissions and underlying warming (Bellenger et al., 2014; Collins et al., 2010; Guilyardi et al., 2012; Ham & Kug, 2016), with large differences found between different Coupled Model Intercomparison Project 5 (CMIP5) model projections (Collins et al., 2010; Guilyardi et al., 2012). The range of projections of ENSO in the future could be due to differences in model physics, resulting in different projections from different models. However, the role of internal variability must also be considered. Such internal variations can result in different projections from single ensemble members of the same climate model. A large ensemble of a single model can be used to estimate this internal variability, and together with other model ensembles can be used to address uncertainties in model physics.

Two large ensembles are used to quantify the extent to which internal variability can contribute to long-term changes in ENSO characteristics. The range of simulated ENSO amplitude changes in the large ensemble historical simulations encompasses 90% of the Coupled Model Intercomparison Project 5 historical simulations and 80% of moderate (RCP4.5) and strong (RCP8.5) warming scenarios. When considering projected ENSO pattern changes, model differences are also important. We find that ENSO has high internal variability and that single realizations of a model can produce very different results to the ensemble mean response. Due to this variability, 30-40 ensemble members of a single model are needed to robustly compute absolute ENSO variance to a 10% error when 30-year analysis periods are used.   

From months to Milankovitch: how timescale-dependent interactions in the coupled Earth system determine the spectrum of climate variability and response.

Acting as both signal and noise, stochastic internal variability dominates the observational record of Earth's energy budget. While variability confounds estimates of anthropogenic climate change, it can also be leveraged for insight into the underlying physics, provided one understands both the governing stochastic processes, and the ways in which they are encoded in the statistics of observable quantities.

I will show how the frequency spectrum of Earth's temperature variability is determined by - and informs on - the climate system's radiative damping efficiency. This damping efficiency determines how much radiation the system sheds to space for a given change in surface temperature, and it is set by how the different components of the climate system - atmosphere, ocean, cryosphere, and the carbon cycle, interact across a range of timescales. The theoretical model for a timescale-dependent radiative damping efficiency is constrained by a combination of observations of broad-band variability drawn from both instrumental records as well as proxies of past climate change.
  

The climate change signal in hurricanes

In this presentation, I will discuss results from our ongoing research on detecting climate change signal in hurricane activity in the recent history. A question that is frequently asked during or after an extreme hurricane season like 2017 is whether such extreme season becomes more frequent or is simply one of the statistical flukes. Such question is hard to answer by using observational data alone as the length of high-quality hurricane record is too short. Therefore, we study this question by a combination of observations, statistical-dynamical downscaling modeling and high-resolution global dynamical modeling approaches. Simulations driven by reanalysis data will be used to estimate historical trends. Simulations driven by outputs from global climate models as well as those directly output from high-resolution global models are used to estimate the roles of radiative forcing, natural variability, and their combination. I will show the comparison of various climatological measures of TC activity from observations and model-based estimates for the current, pre-industrial, and near future periods. In addition to our work, I will also discuss other recent studies on detecting climate change signal on hurricanes.   

Detecting changes in marine ecosystems and ocean colour

Monitoring changes in marine phytoplankton is important as they form the foundation of the marine food web and are crucial in the carbon cycle. Climate change is affecting marine phytoplankton by altering their nutrient, temperature, light and chemical environments. Often Chlorophyll-a (Chl-a) is used to track changes in phytoplankton, since there are global, regular satellite-derived estimates, and such studies suggesting complex, but as yet limited, patterns of long-term change over the last two decades. However, satellite sensors do not measure Chl-a directly. Instead, Chl-a is estimated from remote sensing reflectance (R_RS): the ratio of upwelling radiance to the downwelling irradiance at the ocean’s surface. We use a unique ocean physics, biogeochemistry and ecosystem model that explicitly includes a representation of the ocean’s optical properties to explore how climate change signals are manifested in Chl-a, phytoplankton communities, and ocean colour over the course of the 21^st century. We show that R_RS in the blue-green spectrum is likely to have a stronger and earlier climate change-driven signal than Chl-a. This is because R_RS integrates not only changes to Chl-a, but also alterations in other optically important constituents. Phytoplankton community structure, which strongly affects ocean optics, is likely to show one of the clearest and most rapid signatures of changes to the base of the marine ecosystem.