Earth is a dynamic planet with two separate convecting domains, the rocky outer shell (the mantle) and inner metallic core, both of which are driven by a combination of primordial and radiogenic heat sources. These sources power a host of geological processes including the geodynamo, plate tectonics, mountain building, volcanism and degassing of volatiles to the atmosphere, all of which are familiar surface expressions of the internal dynamics. The purpose of CIDER 2014 is to bring together scientists from different disciplines to better understand how the complex interplay of processes in the interior of Earth and other terrestrial planets shape the internal structure of these bodies and govern their longterm evolution.

The CIDER summer program will address two aspects of planetary dynamics. The first part of the program will focus on models for the present-day convective flow and the criteria used to assess these models. Common constraints include the gravity field, plate motions, heat flow, surface topography, the three-dimensional structure of variations in elastic wave speeds, and the composition of meteorites, terrestrial samples and bulk planets. Additional information can be inferred from maps of seismic anisotropy, deflections in the depth of solid-state phase transition, and the seismicity of subducted oceanic plates that project into the mantle from ocean trenches. Less conventional constraints might include the curvature of trenches or paleomagnetic evidence for true polar wander. The overarching goal is to assess uncertainties and identify areas where future progress is required, both in the models and in the observations.

The second part of the program will deal with the interplay of processes that govern the evolution of mantle flow over geological time. What are the conditions for the onset of plate tectonics and when did this occur? How are new subduction zones initiated and what controls the reorganization of plate motions? How has heat flow varied over the age of the Earth? Has the Earth always had a magnetic field? Few direct observations are available, so greater reliance is placed on more indirect connections between models and measurements. For example, geochemical measurements of isotopic heterogeneity in basalt reveal distinct and persistent reservoirs in the mantle. The origin and spatial distribution of these reservoirs is not known, although secular trends in the chemistry of ancient lava flows provide constraints on the temporal evolution of at least some of these reservoirs. However, the persistence of these reservoirs appears to be inconsistent with estimates of convective mixing, based on models of present-day flow in the mantle. Such inconsistencies imply shortcomings in our understanding of the present-day dynamics and/or the evolution of these dynamic processes back in time. An analogous inconsistency arises in the thermal history of the Earth when cosmochemical abundances of heat-producing elements are combined with conventional models of convection. These inconsistencies reflect an incomplete understanding of planetary dynamics.

Recent advances across the disciplines motivate the proposed CIDER program. Refinements in seismological methods now provide more reliable images of the threedimensional elastic structure of the mantle, particularly at the largest scales. At the same time advances in our understanding of the physical properties of Earth materials at high pressure and temperature give us new insights into physical processes and better constrain numerical simulations. For example, a recent revision in the transport properties of liquid iron at high pressure and temperature has overturned the convention view of convection in Earth’s core and raised questions about how a field was generated early in the history of our planet. Interdisciplinary collaborations in the emerging field of neutrino geophysics have provided insights into the abundance and distribution of heat producing elements (Th & U) in the Earth.

Advances in computing have also enabled dramatic improvements in the numerical simulation of complex processes. Increasingly, numerical simulations are combined with data assimilation schemes and inverse methods to constrain initial conditions, estimate model parameters or identify missing physical components in the models. This emerging trend offers better integration of data and models, but it also imposes greater demands on both the models and the data. The limitations and uncertainties in the models and the observations need to be understood and quantified, requiring greater coordination across the disciplines. The proposed program will focus a broad base of expertise on several long-standing problems. The format is unique in that it enables in-depth discussion of competing ideas from a multidisciplinary perspective.