Marcia L. Branstetter

Higher trends but larger uncertainty and geographic variability in 21st century temperature and heat waves

Generating credible climate change and extremes projections remains a high-priority challenge, especially since recent observed emissions are above the worst-case scenario. Bias and uncertainty analyses of ensemble simulations from a global earth systems model show increased warming and more intense heat waves combined with greater uncertainty and large regional variability in the 21st century. Global warming trends are statistically validated across ensembles and investigated at regional scales. Observed heat wave intensities in the current decade are larger than worst-case projections. Model projections are relatively insensitive to initial conditions, while uncertainty bounds obtained by comparison with recent observations are wider than ensemble ranges. Increased trends in temperature and heat waves, concurrent with larger uncertainty and variability, suggest greater urgency and complexity of adaptation or mitigation decisions.

Practical Application of Parallel Coordinates for Climate Model Analysis

The determination of relationships between climate variables and the identification of the most significant associations between them in various geographic regions is an important aspect of climate model evaluation. The EDEN visual analytics toolkit has been developed to aid such analysis by facilitating the assessment of multiple variables with respect to the amount of variability that can be attributed to specific other variables. EDEN harnesses the parallel coordinates visualization technique and is augmented with graphical indicators of key descriptive statistics. A case study is presented in which the focus is on the Harvard Forest site (42.5378N Lat, 72.1715W Lon) and the Community Land Model Version 4 (CLM4) is evaluated. It is shown that model variables such as land water runoff are more sensitive to a particular set of environmental variables than a suite of other inputs in the 88 variable analysis conducted. The approach presented here allows climate scientists to focus on the most important variables in the model evaluations.

A spectral transform dynamical core option within the Community Atmosphere Model (CAM4)

An ensemble of simulations covering the present day observational period using forced sea surface temperatures and prescribed sea-ice extent is configured with an 85 truncation resolution spectral transform dynamical core (T85) within the Community Atmosphere Model (CAM), version 4 and is evaluated relative to observed and model derived data sets and the one degree finite volume (FV) dynamical core. The spectral option provides a well-known base within the climate model community to assess climate behavior and statistics, and its relative computational efficiency for smaller computing platforms allows it to be extended to perform high-resolution climate length simulations. Overall, the quality of the T85 ensemble is similar to FV. Analyzing specific features of the T85 simulations show notable improvements to the representation of wintertime Arctic sea level pressure and summer precipitation over the Western Indian subcontinent. The mean and spatial patterns of the land surface temperature trends over the AMIP period are generally well simulated with the T85 ensemble relative to observations, however the model is not able to capture the extent nor magnitude of changes in temperature extremes over the boreal summer, where the changes are most dramatic. Biases in the wintertime Arctic surface temperature and annual mean surface stress fields persist with T85 as with the CAM3 version of T85, as compared to FV. An experiment to identify the source of differences between dycores has revealed that the longwave cloud forcing is sensitive to the choice of dycore, which has implications for tuning strategies of the physics parameter settings.

A Linked Science investigation: enhancing climate change data discovery with semantic technologies

Linked Science is the practice of inter-connecting scientific assets by publishing, sharing and linking scientific data and processes in end-to-end loosely coupled workflows that allow the sharing and re-use of scientific data. Much of this data does not live in the cloud or on the Web, but rather in multi-institutional data centers that provide tools and add value through quality assurance, validation, curation, dissemination, and analysis of the data. In this paper, we make the case for the use of scientific scenarios in Linked Science. We propose a scenario in river-channel transport that requires biogeochemical experimental data and global climate-simulation model data from many sources. We focus on the use of ontologies—formal machine-readable descriptions of the domain—to facilitate search and discovery of this data. Mercury, developed at Oak Ridge National Laboratory, is a tool for distributed metadata harvesting, search and retrieval. Mercury currently provides uniform access to more than 100,000 metadata records; 30,000 scientists use it each month. We augmented search in Mercury with ontologies, such as the ontologies in the Semantic Web for Earth and Environmental Terminology (SWEET) collection by prototyping a component that provides access to the ontology terms from Mercury. We evaluate the coverage of SWEET for the ORNL Distributed Active Archive Center (ORNL DAAC).

High performance statistical computing with parallel R: applications to biology and climate modelling

Ultrascale computing and high-throughput experimental technologies have enabled the production of scientific data about complex natural phenomena. With this opportunity, comes a new problem – the massive quantities of data so produced. Answers to fundamental questions about the nature of those phenomena remain largely hidden in the produced data. The goal of this work is to provide a scalable high performance statistical data analysis framework to help scientists perform interactive analyses of these raw data to extract knowledge. Towards this goal we have been developing an open source parallel statistical analysis package, called Parallel R, that lets scientists employ a wide range of statistical analysis routines on high performance shared and distributed memory architectures without having to deal with the intricacies of parallelizing these routines.

The atmospheric hydrologic cycle in the ACME v0.3 model

We examine the global water cycle characteristics in the Accelerated Climate Modeling for Energy v0.3 model (a close relative to version 5.3 of the Community Atmosphere Model) in atmosphere-only simulations spanning the years 1980–2005. We evaluate the simulations using a broad range of observational and reanalysis datasets, examine how the simulations change when the horizontal resolution is increased from 1° to 0.25

Geospatial–temporal dependence among weekly precipitation extremes with applications to observations and climate model simulations in South America

^{\circ }

Biospheric feedback effects in a synchronously coupled model of human and Earth systems

∘, and compare the simulations against models participating in the the Atmosphere Model Intercomparison Project of the 5th Coupled Model Intercomparison Project (CMIP5). Particular effort has been made to evaluate the model using the best available observational estimates and verifying model biases with additional datasets when differences are known to exist among the observations. Regardless of resolution, the model exhibits several biases: global-mean precipitation, evaporation, and precipitable water are too high, light precipitation occurs too frequently, and the atmospheric residence time of water is too short. Many of these biases are shared by the multi-model mean climate of models participating in CMIP5. The reasons behind regional biases in precipitation are discussed by examining how different fields, such as local evaporation and transport of water vapor, contribute to the bias. Although increasing the horizontal resolution does not drastically change the water cycle, it does lead to a few differences: an increase in global mean precipitation rate, an increase in the fraction of total precipitation that falls over land, more frequent heavy precipitation (>30 mm day