Earth Science Models Research Articles

SWIRRL. Managing Provenance-aware and Reproducible Workspaces

Abstract Modern interactive tools for data analysis and visualisation are designed to expose their functionalities as a service through the Web. We present in this paper a Web API (SWIRRL) that allows Virtual Research Environments (VREs) to easily integrate such tools in their websites and re-purpose them to their users. The API deals, on behalf of the clients, with the underlying complexity of allocating and managing resources within a target cloud platform. By combining storage and containerised services, offering analysis notebooks and other visualisation software, the API creates dedicated working sessions on-demand, which can be accessed collaboratively. Thanks to the API's support for workflow execution, SWIRRL workspaces can be automatically populated with data of interest collected from external data providers. The system keeps track of updates and changes affecting the data and the tools by adopting versioning and standard provenance technologies. Users are provided with interactive controls enabling traceability and recovery actions, including the possibility of creating executable snapshots of their environments. SWIRRL is built in cooperation with two research infrastructures in the field of solid earth science and climate data modeling. We report on the particular adoptions and use cases.

The Grids Python Tool for Querying Spatiotemporal Multidimensional Water Data

Scientific datasets from global-scale earth science models and remote sensing instruments are becoming available at greater spatial and temporal resolutions with shorter lag times. Water data are frequently stored as multidimensional arrays, also called gridded or raster data, and span two or three spatial dimensions, the time dimension, and other dimensions which vary by the specific dataset. Water engineers and scientists need these data as inputs for models and generate data in these formats as results. A myriad of file formats and organizational conventions exist for storing these array datasets. The variety does not make the data unusable but does add considerable difficulty in using them because the structure can vary. These storage formats are largely incompatible with common geographic information system (GIS) software. This introduces additional complexity in extracting values, analyzing results, and otherwise working with multidimensional data since they are often spatial data. We present a Python package which provides a central interface for efficient access to multidimensional water data regardless of the file format. This research builds on and unifies existing file formats and software rather than suggesting entirely new alternatives. We present a summary of the code design and validate the results using common water-related datasets and software.

Enabling High‐Performance Cloud Computing for Earth Science Modeling on Over a Thousand Cores: Application to the GEOS‐Chem Atmospheric Chemistry Model

AbstractCloud computing platforms can facilitate the use of Earth science models by providing immediate access to fully configured software, massive computing power, and large input data sets. However, slow internode communication performance has previously discouraged the use of cloud platforms for massively parallel simulations. Here we show that recent advances in the network performance on the Amazon Web Services cloud enable efficient model simulations with over a thousand cores. The choices of Message Passing Interface library configuration and internode communication protocol are critical to this success. Application to the Goddard Earth Observing System (GEOS)‐Chem global 3‐D chemical transport model at 50‐km horizontal resolution shows efficient scaling up to at least 1,152 cores, with performance and cost comparable to the National Aeronautics and Space Administration Pleiades supercomputing cluster.

Enabling Immediate Access to Earth Science Models through Cloud Computing: Application to the GEOS-Chem Model

AbstractCloud computing platforms can provide fast and easy access to complex Earth science models and large datasets. This article presents a mature capability for running the GEOS-Chem global 3D model of atmospheric chemistry on the Amazon Web Services (AWS) cloud. GEOS-Chem users at any experience level can get immediate access to the latest, standard version of the model in a preconfigured software environment with all needed meteorological and other input data, and they can analyze model output data easily within the cloud using Python tools in Jupyter notebooks. Users with no prior knowledge of cloud computing are provided with easy-to-follow, step-by-step instructions. They can learn how to complete a demo project in less than one hour, and from there they can configure and submit their own simulations. The cloud is particularly attractive for beginning and occasional users who otherwise may need to spend substantial time configuring a local computing environment. Heavy users with their own local clusters can also benefit from the cloud to access the latest standard model and datasets, share simulation configurations and results, benchmark local simulations, and respond to surges in computing demand. Software containers allow GEOS-Chem and its software environment to be moved smoothly between cloud platforms and local clusters, so that the exact same simulation can be reproduced everywhere. Because the software requirements and workflows tend to be similar across Earth science models, the work presented here provides general guidance for porting models to cloud computing platforms in a user-accessible way.

Spatial-feature data cube for spatiotemporal remote sensing data processing and analysis

With the rapid development in Earth observation technology, a variety of satellite sensors have provided large and open sets of remote sensing data. However, traditional methods of analysis are no longer available for time-serial remote sensing data analysis that typically handles multidimensional spatio-temporal data models. Moreover, researchers have found it trivial and tedious to obtain ready-to-analyze data for Earth science models from regular Earth observation data. For an easy and efficient time-serial remote sensing data analysis, a spatial-featured data cube analysis tool based on multidimensional data model is proposed for time-serial remote sensing data processing and analysis. For the performance consideration, a distributed execution engine was also used for efficient implementation of large-scale tasks in parallel. Finally, through experiments on both normalized difference vegetation index product and water detection within a 20-year period, we confirmed that our approach is efficient and scalable for a long time-series analysis.

31 years of hourly spatially distributed air temperature, humidity, and precipitation amount and phase from Reynolds Critical Zone Observatory

Abstract. Thirty-one years of spatially distributed air temperature, relative humidity, dew point temperature, precipitation amount, and precipitation phase data are presented for the Reynolds Creek Experimental Watershed, which is part of the Critical Zone Observatory network. The air temperature, relative humidity, and precipitation amount data are spatially distributed over a 10 m lidar-derived digital elevation model at an hourly time step using a detrended kriging algorithm. This 21 TB dataset covers a wide range of weather extremes in a mesoscale basin (238 km2) that encompasses the rain–snow transition zone and should find widespread application in earth science modeling communities. Spatial data allow for a more holistic analysis of basin means and elevation gradients, compared to weather station data measured at specific locations. Files are stored in the NetCDF file format, which allows for easy spatiotemporal averaging and/or subsetting. Data are made publicly available through an OPeNDAP-enabled THREDDS server hosted by Boise State University Libraries in support of the Reynolds Creek Critical Zone Observatory (https://doi.org/10.18122/B2B59V).

Using a Similarity Matrix Approach to Evaluate the Accuracy of Rescaled Maps

Rescaled maps have been extensively utilized to provide data at the appropriate spatial resolution for use in various Earth science models. However, a simple and easy way to evaluate these rescaled maps has not been developed. We propose a similarity matrix approach using a contingency table to compute three measures: overall similarity (OS), omission error (OE), and commission error (CE) to evaluate the rescaled maps. The Majority Rule Based aggregation (MRB) method was employed to produce the upscaled maps to demonstrate this approach. In addition, previously created, coarser resolution land cover maps from other research projects were also available for comparison. The question of which is better, a map initially produced at coarse resolution or a fine resolution map rescaled to a coarse resolution, has not been quantitatively investigated. To address these issues, we selected study sites at three different extent levels. First, we selected twelve regions covering the continental USA, then we selected nine states (from the whole continental USA), and finally we selected nine Agriculture Statistical Districts (ASDs) (from within the nine selected states) as study sites. Crop/non-crop maps derived from the USDA Crop Data Layer (CDL) at 30 m as base maps were used for the upscaling and existing maps at 250 m and 1 km were utilized for the comparison. The results showed that a similarity matrix can effectively provide the map user with the information needed to assess the rescaling. Additionally, the upscaled maps can provide higher accuracy and better represent landscape pattern compared to the existing coarser maps. Therefore, we strongly recommend that an evaluation of the upscaled map and the existing coarser resolution map using a similarity matrix should be conducted before deciding which dataset to use for the modelling. Overall, extending our understanding on how to perform an evaluation of the rescaled map and investigation of the applicability of the rescaled map compared to the existing land cover map is necessary for users to most effectively use these data in Earth science models.

31 years of hourly spatially distributed air temperature, humidity, and precipitation amount and phase from Reynolds Critical Zone Observatory

Thirty one years of spatially distributed air temperature, relative humidity, dew point temperature, precipitation amount, and precipitation phase data are presented for the Reynolds Creek Experimental Watershed, which is part of the Critical Zone Observatory network. The air temperature, relative humidity, and precipitation amount data are spatially distributed over a 10 m Lidar-derived digital elevation model at an hourly time step using a detrended kriging algorithm. This dataset covers a wide range of weather extremes in a mesoscale basin (237 km<sup>2</sup>) that encompasses the rain-snow transition zone and should find widespread application in earth science modeling communities. Spatial data allows for a more holistic analysis of basin means and elevation gradients, compared to weather station data measured at specific locations. Files are stored in the NetCDF file format, which allows for easy spatiotemporal averaging and/or subsetting. Data are made publicly available through an OPeNDAP-enabled THREDDS server hosted by Boise State University Libraries in support of the Reynolds Creek Critical Zone Observatory (<a href="https://doi.org/doi:10.18122/B2B59V" target ="_blank">https://doi.org/10.18122/B2B59V</a>).

CyberConnector: a service-oriented system for automatically tailoring multisource Earth observation data to feed Earth science models

Feeding multisource Earth observation (EO) data into Earth science models (ESM) remains a daunting challenge. This paper presents a service-oriented approach as an alternative solution. It uses geospatial web services to process the EO data and geoprocessing workflow for automation. Different from existing approaches, it takes advantage of virtual data products (VDP) to release modelers from intensive data processing. It can directly connect ESMs to public EO sources via Cyberinfrastructure. A prototype called CyberConnector is implemented. CyberConnector supports intuitive building of VDP, automatic execution of workflows and effortless retrieval of model-ready input files. We used it to stream multiple datasets to several ESMs including finite-volume coastal ocean model (FVCOM) and cloud-resolving model (CRM). The results show that CyberConnector can truly benefit modelers on time saving and effort minimizing.

Increasing the Use of Earth Science Data and Models in Air Quality Management

ABSTRACTIn 2010, the U.S. National Aeronautics and Space Administration (NASA) initiated the Air Quality Applied Science Team (AQAST) as a 5-year, $17.5-million award with 19 principal investigators. AQAST aims to increase the use of Earth science products in air quality-related research and to help meet air quality managers’ information needs. We conducted a Web-based survey and a limited number of follow-up interviews to investigate federal, state, tribal, and local air quality managers’ perspectives on usefulness of Earth science data and models, and on the impact AQAST has had. The air quality managers we surveyed identified meeting the National Ambient Air Quality Standards for ozone and particulate matter, emissions from mobile sources, and interstate air pollution transport as top challenges in need of improved information. Most survey respondents viewed inadequate coverage or frequency of satellite observations, data uncertainty, and lack of staff time or resources as barriers to increased use of satellite data by their organizations. Managers who have been involved with AQAST indicated that the program has helped build awareness of NASA Earth science products, and assisted their organizations with retrieval and interpretation of satellite data and with application of global chemistry and climate models. AQAST has also helped build a network between researchers and air quality managers with potential for further collaborations.Implications: NASA’s Air Quality Applied Science Team (AQAST) aims to increase the use of satellite data and global chemistry and climate models for air quality management purposes, by supporting research and tool development projects of interest to both groups. Our survey and interviews of air quality managers indicate they found value in many AQAST projects and particularly appreciated the connections to the research community that the program facilitated. Managers expressed interest in receiving continued support for their organizations’ use of satellite data, including assistance in retrieving and interpreting data from future geostationary platforms meant to provide more frequent coverage for air quality and other applications.

Application of an evolutionary algorithm for parameter optimization in a gully erosion model

Herein we demonstrate how to use model optimization to determine a set of best-fit parameters for a landform model simulating gully incision and headcut retreat. To achieve this result we employed the Covariance Matrix Adaptation Evolution Strategy (CMA-ES), an iterative process in which samples are created based on a distribution of parameter values that evolve over time to better fit an objective function. CMA-ES efficiently finds optimal parameters, even with high-dimensional objective functions that are non-convex, multimodal, and non-separable. We ran model instances in parallel on a high-performance cluster, and from hundreds of model runs we obtained the best parameter choices. This method is far superior to brute-force search algorithms, and has great potential for many applications in earth science modeling. We found that parameters representing boundary conditions tended to converge toward an optimal single value, whereas parameters controlling geomorphic processes are defined by a range of optimal values.

The abiotic and biotic drivers of rapid diversification in Andean bellflowers (Campanulaceae).

Summary The tropical Andes of South America, the world's richest biodiversity hotspot, are home to many rapid radiations. While geological, climatic, and ecological processes collectively explain such radiations, their relative contributions are seldom examined within a single clade.We explore the contribution of these factors by applying a series of diversification models that incorporate mountain building, climate change, and trait evolution to the first dated phylogeny of Andean bellflowers (Campanulaceae: Lobelioideae). Our framework is novel for its direct incorporation of geological data on Andean uplift into a macroevolutionary model.We show that speciation and extinction are differentially influenced by abiotic factors: speciation rates rose concurrently with Andean elevation, while extinction rates decreased during global cooling. Pollination syndrome and fruit type, both biotic traits known to facilitate mutualisms, played an additional role in driving diversification. These abiotic and biotic factors resulted in one of the fastest radiations reported to date: the centropogonids, whose 550 species arose in the last 5 million yr.Our study represents a significant advance in our understanding of plant evolution in Andean cloud forests. It further highlights the power of combining phylogenetic and Earth science models to explore the interplay of geology, climate, and ecology in generating the world's biodiversity.

A community-oriented workflow reuse and recommendation technique

NASA Earth Exchange (NEX) is a collaborative compute platform aiming to improve the availability of Earth science data, models, analysis tools and scientific results through a centralised environment that fosters knowledge sharing, collaboration, innovation and direct access to compute resources. One of the main objectives of NEX is to help Earth scientists leverage and reuse various data processing software modules developed by their peers, in order to quickly run value-added executable experiments (workflows). Toward this goal, this paper reports our efforts of leveraging social network analysis to intelligently extract hidden information from data processing workflows. By modelling Earth science workflow modules as social entities and their dependencies as social relationships, this research opens up new vistas for applying social science to facilitate software reuse and distributed workflow development. As a proof of concept, a prototyping system has been developed as a plug-in to the NEX workflow design and management system (VisTrails) to aid Earth scientists in discovering and reusing workflow modules and extending them to solve more complex science problems.

Model validation based on in situ radionuclide migration tests in Boom Clay: status of a large-scale migration experiment, 24 years after injection

Abstract Demonstration of the long-term safety of a nuclear waste repository relies on earth science models, integrated in a performance assessment model chain. These models are subject to quality assurance procedures and principles of which model validation, qualification and verification are essential elements. However, in the context of performance assessment, model validation is often limited owing to extreme timescales and the use of natural barriers that can never be entirely characterized. Nevertheless, it is often possible to demonstrate that the models are valid or qualified to describe the processes at hand. In case of geological disposal in Boom Clay, the host formation is the dominant barrier for radionuclide migration and releases to the biosphere. Therefore, it has to be demonstrated that migration of solutes through Boom Clay at relevant scale is adequately understood. Large-scale and long-term in situ migration experiments, such as the CP1 experiment, form a cornerstone in this confidence-building process. In this experiment, accurate predictions of the tracer's breakthrough curves up to 3 m from the source have been obtained using the conventional advection–dispersion–reaction equation to describe solute transport and parameters obtained from (small-scale) migration experiments in the laboratory.

Towards Data and Sensor Planning Service for Coupling Earth Science Models and Earth Observations

How to obtain sensor observation data easily and quickly is one of the challenging research areas in the Earth science community. Data and Sensor Planning Service aims to generate requirements on Earth observations data from which the geophysical products needed by Earth science models can derive. Then, the service discovers relevant data sources which meet these requirements, schedules data acquisition by ordering and retrieving data from these sources, and processes and returns the customized data which will be used in the Earth science models. In this general framework, Sensor Web Enablement specifications of Open Geospatial Consortium are implemented and integrated with other open Web geospatial data services seamlessly to access real or virtual Earth observation sensors. Additionally, an Ajax-enabled browser client is built to facilitate the standard sensor Web operations, user account management along with data acquisition. The prototype system is demonstrated to retrieve near real-time satellite observation data from Comprehensive Large Array-data Stewardship System through standard Web services, and also is applied in the bird migration model which uses climate data, vegetation data, as well as other remotely sensed and in-situ observation data to determine the bird migration routes and identify the possible resting areas.

Coordination Through Geospatial Web Service Workflow in the Sensor Web Environment

The Sensor Web connects live sensors and Earth Science models (ESM). A workflow is one approach for designing, implementing, and constructing a live link between sensors and ESM. The Sensor Web consists of many individual Web services. A workflow coordinates them to enable bi-directional connectivity between sensors and ESMs. This study uses Business Process Execution Language as the base scripting language to design, author, deploy, and execute workflows. Message-level coordination mechanisms are developed, implemented, and demonstrated in two scenarios-bird migratory modeling and georeferenceable imagery workflow. The advantage of message-level coordination is short-cutting the data flow.

Middleware-Based Sensor Web Integration

The Earth observation sensor web enables multiple-way interaction between earth observing sensors, sensor networks, Earth science models, and decision support systems. To achieve this goal, flexible and reliable integration between these disparate components is needed. In this study, a middleware-based, message-driven integration paradigm is proposed and implemented with the Land Information Sensor Web (LISW), to link a high-performance land surface modeling system with sensor simulators and other sensor web components, under a service-oriented architecture. OGC Sensor Web Enablement standard is adopted for interoperability. The middleware played a key role in enabling an integrated real-time sensor web with demonstrated simplicity, resilience and flexibility. We recommend that middleware-based integration should be adopted as a standard model in a wide range of sensor web applications, to replace the conventional point-to-point, client-server model.

CHEPROO: A Fortran 90 object-oriented module to solve chemical processes in Earth Science models

Accurate prediction of contaminant migration in surface and ground water bodies, including interaction with aquifer and hyporheic zone materials requires reactive transport modeling. The increasing complexity and the procedure-oriented type of programming often used in reactive transport hinder codes reuse and transportability. We present a Fortran 90 module using object-oriented concepts that simulates complex hydrobiogeochemical processes (CHEPROO, CHEmical PRocesses Object-Oriented). CHEPROO consists of a general structure with two classes. The Nodal Chemistry class accounts for the description of local chemistry and geochemical state variables. As such, it provides many functions related to basic operations (evaporation, mixing, etc.) and can easily grow on this direction (extreme dry conditions, biochemical state variables, etc.). The Chemical System class includes kinetic and thermodynamic models that describe reactions between and within phases. As such, it can grow in the direction of increasingly complex chemical systems (solid solutions, microorganisms as individual phases, etc.), without loss in the handling of simple problems. These two classes are overlaid by CHEPROO, a general structure designed for interaction with other codes. CHEPROO can be used as a geochemical tool for the modeling of complex processes such as biodegradation or evaporation at high salinities. However, many functions CHEPROO are devoted to coupling a broad range of chemical processes to other phenomena (flow, transport, mechanical). We have shown that reactive transport (based on either DSA or SIA approaches) could be easily implemented into existing conservative transport code with a minimal number of changes.

Effective interpolation of incomplete satellite-derived leaf-area index time series for the continental United States

Many earth science modeling applications employ continuous input data fields derived from satellite data. Environmental factors, sensor limitations and algorithmic constraints lead to data products of inherently variable quality. This necessitates interpolation of one form or another in order to produce high quality input fields free of missing data. The present research tests several interpolation techniques as applied to satellite-derived leaf-area index (LAI), an important quantity in many global climate and ecological models. The study evaluates and applies a variety of interpolation techniques for the Moderate Resolution Imaging Spectroradiometer (MODIS) LAI product over the time period of 2001–2006 for a region containing the conterminous United States. Results indicate that the accuracy of an individual interpolation technique depends upon the underlying land cover. Spatial interpolation provides better results in forested areas, while temporal interpolation performs more effectively over non-forest cover types. Combination of spatial and temporal approaches offers superior interpolative capabilities to any single method, and in fact, generation of continuous data fields requires a hybrid approach such as this.

Modelling marine ecosystems as a discipline in Earth Science

Faced with the imperatives of sustainable development, Earth Science must open to the study of ecosystems and their interactions with a multiscale geophysical environment. The indispensable development of interdisciplinary Earth Science models requires that the crafts and skills of physicists, chemists and biologists merge and cross-fertilize, with often a long way to go for each discipline to win over the others. This paper contains the reflections, queries and suggestions of a marine hydrodynamicist trying to develop coupled physical, chemical and biological marine models and negotiating admittance in the Biogeochemistry Club.