SimilarityExplorer: A Visual Inter‐Comparison Tool for Multifaceted Climate Data

Poco, Jorge; Dasgupta, Aritra; Wei, Yaxing; Hargrove, William W.; Schwalm, Christopher R.; Cook, R.B.; Bertini, Enrico; Silva, C.

doi:10.1111/cgf.12390

Cited by 30 publications

(15 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In practice, however, this was no issue because ocean modelers typically study weekly, monthly, or even seasonal averages. This approach is also applied by climate scientists (as described by Poco et al [33]) and significantly reduces the number of time steps per year.…”

Section: Discussionmentioning

confidence: 99%

“…Kehrer et al [24,25] and Ladtstädter et al [27] support visual analysis and comparison of different variables of climate model output by multiple linked views. Recent work by Poco et al [33] focusses on the comparison of output from different climate models. Their approach concentrates on the analysis of correlations between data sets.…”

Section: Related Workmentioning

confidence: 99%

“…Although the consolidation overview component provides them with important information that guides subsequent analysis steps, they can always decide to make an educated guess and readily study any potentially interesting aspect. Before they had our tool, such an educated guess was not feasible because it involved time-consuming scripting and plotting procedures, a problem that also exists in related domains such as climate research [33,37]. The means of interaction provided in our tool effectively remove this hurdle.…”

Section: User Feedbackmentioning

confidence: 99%

See 2 more Smart Citations

Visual Analytics for Comparison of Ocean Model Output with Reference Data: Detecting and Analyzing Geophysical Processes Using Clustering Ensembles

Köthur

Sips

Dobslaw

et al. 2014

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

Researchers assess the quality of an ocean model by comparing its output to that of a previous model version or to observations. One objective of the comparison is to detect and to analyze differences and similarities between both data sets regarding geophysical processes, such as particular ocean currents. This task involves the analysis of thousands or hundreds of thousands of geographically referenced temporal profiles in the data. To cope with the amount of data, modelers combine aggregation of temporal profiles to single statistical values with visual comparison. Although this strategy is based on experience and a well-grounded body of expert knowledge, our discussions with domain experts have shown that it has two limitations: (1) using a single statistical measure results in a rather limited scope of the comparison and in significant loss of information, and (2) the decisions modelers have to make in the process may lead to important aspects being overlooked.In this article, we propose a visual analytics approach that broadens the scope of the analysis, reduces subjectivity, and facilitates comparison of the two data sets. It comprises three steps: First, it allows modelers to consider many aspects of the temporal behavior of geophysical processes by conducting multiple clusterings of the temporal profiles in each data set. Modelers can choose different features describing the temporal behavior of relevant processes, clustering algorithms, and parameterizations. Second, our approach consolidates the clusterings of one data set into a single clustering via a clustering ensembles approach. The consolidated clustering presents an overview of the geospatial distribution of temporal behavior in a data set. Third, a visual interface allows modelers to compare the two consolidated clusterings. It enables them to detect clusters of temporal profiles that represent geophysical processes and to analyze differences and similarities between two data sets.This work is the result of a close collaboration with ocean modelers. They employed our concept to find aspects of improvement in a new version of the Ocean Model for Circulation and Tides (OMCT).

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: User Feedbackmentioning

confidence: 99%

See 1 more Smart Citation

Visual Analytics for Comparison of Ocean Model Output with Reference Data: Detecting and Analyzing Geophysical Processes Using Clustering Ensembles

Köthur

Sips

Dobslaw

et al. 2014

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

show abstract

“…[1][2][3][4] There is inherent uncertainty and disagreement in the parameterization of these models and in understanding their effects on outputs. However, gauging consensus among model outputs is critical for achieving high accuracy about prediction of environmental events, climate change patterns, and so on.…”

Section: Climate Science Backgroundmentioning

confidence: 99%

Reducing the Analytical Bottleneck for Domain Scientists: Lessons from a Climate Data Visualization Case Study

et al. 2016

View full text Add to dashboard Cite

R ecent advances in high-performance computing have helped produce huge amounts of data at a very fast rate. But the relative speeds of data analysis and exploration, especially in scientific disciplines, have been disproportionately low. For example, using a modern supercomputer, it only takes a few days to produce terabytes of scientific simulation output data, but it then takes weeks or months of tedious scripting for domain scientists to perform analysis and exploration. Richard Hamming famously said, "The purpose of computing is insight, not numbers" (https://en.wikipedia.org/wiki/ Richard_Hamming). But most current scientific data analysis tools have focused exclusively on the data scalability problem and not so much on the complexity and variety of the data space that affect the process of deriving insights from the data.In domains such as biology and climate, many scientists still adopt manual, time-consuming data analysis processes or use tools that don't tightly integrate interactive and analytical capabilities. To let domain scientists analyze, explore, and synthesize insights from large simulation data, there's a need for new techniques and analytical abstractions that will significantly speed up the

show abstract

“…Ensemble-Vis [3], Noodles [4], SimilarityExplorer [2] and EnsembleGraph [5] were built to provide insights on 2D numerical weather simulation models. They are efficient in presenting insights of overall uncertainty but are inefficient in handling various distributions of uncertainty as their spatial visualization techniques are not effective in identifying and tracking outliers.…”

Section: Problem Statementmentioning

confidence: 99%

Visual scalability of spatial ensemble uncertainty

Anreddy

Zhang

Mercer

et al. 2015

2015 IEEE Conference on Visual Analytics Science and Technology (VAST)

View full text Add to dashboard Cite

Weather Research and Forecasting (WRF) models simulate weather conditions by generating 2D numerical weather prediction ensemble members either through perturbing initial conditions or by changing different parameterization schemes, e.g., cumulus and microphysics schemes. These simulations are often used by weather analysts to analyze the nature of uncertainty attributed by these simulations to forecast weather conditions with good accuracy. The number of simulations used for forecasting is growing with the advent of increase in computing power. Hence, there is a need for providing better visual insights of uncertainty with growing number of ensemble members. We propose a geo visual analytical framework that uses visual analytics approach to resolve visual scalability of these ensemble members. Our approach naturally fits with the workflow of an analyst analyzing ensemble spatial uncertainty. Meteorologists evaluated our framework qualitatively and found it to be effective in acquiring insights of spatial uncertainty associated with multiple ensemble runs that are simulated using multiple parameterization schemes. PROBLEM STATEMENTWeather forecasters simulate weather conditions using numerical weather prediction models. These models are run multiple times in order to reduce error and uncertainty attributed by one single run. Each simulation run is termed an ensemble member that is generated either by perturbing initial conditions or by changing parameterization schemes. Some of the tasks that are performed with ensemble members and parameterization schemes are to find an ensemble run that represents consensus of all ensemble runs and also to find outliers from the ensemble. Therefore, in this paper "uncertainty" refers to the degree to which 2D scalar ensemble members agree or disagree with each other. A meteorologist is interested in understanding uncertainty attributed by parameterization schemes whereas an operational weather forecaster is interested in forecasting weather using these simulations. The climatologists also use these simulations to predict climate. Ensemble-Vis[3], Noodles[4], SimilarityExplorer[2]and EnsembleGraph [5] were built to provide insights on 2D numerical weather simulation models. They are efficient in presenting insights of overall uncertainty but are inefficient in handling various distributions of uncertainty as their spatial visualization techniques are not effective in identifying and tracking outliers. These frameworks initially depend on providing insights based on spatial characteristics of uncertainty and use those insights to understand multiple parameterization schemes. So, if the spatial * visualization techniques used in these frameworks do not yield better representations of underlying distributions of uncertainty, it would affect a meteorologist s understanding of parameterization schemes.The use of visual metaphors such as graduated glyphs, ribbons are good at presenting the overall uncertainty or spread of ensemble members on single large display, but are not helpful i...

show abstract

SimilarityExplorer: A Visual Inter‐Comparison Tool for Multifaceted Climate Data

Cited by 30 publications

References 30 publications

Visual Analytics for Comparison of Ocean Model Output with Reference Data: Detecting and Analyzing Geophysical Processes Using Clustering Ensembles

Visual Analytics for Comparison of Ocean Model Output with Reference Data: Detecting and Analyzing Geophysical Processes Using Clustering Ensembles

Reducing the Analytical Bottleneck for Domain Scientists: Lessons from a Climate Data Visualization Case Study

Visual scalability of spatial ensemble uncertainty

Contact Info

Product

Resources

About