Search

The grassland is an important land-use type that plays an important role in the ecosystem service supply in China. It is of great significance to the grassland management to determine the changing trend of grassland productivity and its response to land-use change and climate change. In this chapter, we first examined changes in grassland productivity due to climate and land-use change in the Three-River Headwaters region (TRHR) of Qinghai Province. In the macrocontext of climatic change, we analyzed the possible changing trends of the net primary productivity (NPP) of local grasslands under four representative concentration pathways (RCPs) scenarios (i.e., RCP2.6, RCP4.5, RCP6.0, and RCP8.5) during 2010–2030 with the model estimation, and the grass yield and theoretical grazing capacity under each scenario were further qualitatively and quantitatively analyzed. The results indicate that the grassland productivity in the TRHR will be unstable under all the four scenarios. The grassland productivity will be greatly influenced by the fluctuations of precipitation, and the temperature fluctuations will also play an important role during some periods. The local grassland productivity will decrease to some degree during 2010–2020 and then will fluctuate and increase slowly during 2020–2030. The theoretical grazing capacity was analyzed in this study and calculated on the basis of the grass yield. The result indicates that the theoretical grazing capacity ranges from 4 to 5 million sheep under the four scenarios and it can provide quantitative information reference for decision making on how to determine the reasonable grazing capacity, promote the sustainable development of grasslands, and so forth. Further, we estimated changes in economic returns of livestock production in the TRHR of China. The land surface in TRHR, a typical ecological fragile zone of China, is quite sensitive to the climate change which will destabilize certain ecosystem service valuable to the entire nation and neighboring countries. We analyzed the impacts of climate change and agents’ adaptive behaviors on the regional land-use change with the agent-based model (ABM). First, the main agents were extracted according to the production resource endowments and socioeconomic background. Then, the agents’ land-use behaviors were analyzed and parameterized. Thereafter, the ABM model was built to simulate the impacts of climate change on the regional land-use change and agents’ economic benefits. The results showed that the land-use change was mainly characterized by the increase of grassland and decrease of unused land area. Besides, the agents would get more wealth under the scenario without climate change in the long term, even though the total income is lower than that under the scenario with climate change. In addition, the sensitivity analysis indicated that the model is sensitive to the climatic conditions, market price of agricultural and animal husbandry products, government subsidies, and cost control. Finally, we predicted changes in grassland productivity in China. The results showed that, firstly, the relationship between grassland productivity and climate change, geographical conditions, and human activities was analyzed with the panel data of the whole China during 1980–2010. The result indicated that the temperature and precipitation were very important to grassland productivity at the national scale; secondly, the grassland in China was divided into seven grassland ecological–economic zones according to the ecosystem service function and climate characteristics. The relationship between grassland productivity and climate change was further analyzed at the regional scale. The result indicated that the temperature is more beneficial to the increase of the grassland productivity in the Qinghai–Tibet Plateau and the Southwest Karst shrubland region; thirdly, the increase of the temperature and precipitation can increase the grassland productivity and consequently relieve the pressure according to the climate factors of simulation with the community climate system model v4.0 (CCSM). However, the simulation result indicates that the human pressure on grasslands is still severe under the four RCPs scenarios and the grassland area would reduce sharply due to the conversion from the grassland to the cultivated land. What is more, there is still a great challenge to the increase of total grassland productivity in China.

Climate scientists have made substantial progress in understanding Earth's climate system, particularly at global and continental scales. Climate research is now focused on understanding climate changes over wider ranges of time and space scales. These efforts are generating ultra-scale data sets at very high spatial resolution. An insightful analysis in climate science depends on using software tools to discover, access, manipulate, and visualize the data sets of interest. These data exploration tasks can be complex and time-consuming, and they frequently involve many resources from both the modeling and observational climate communities. Because of the complexity of the explorations, provenance is critical, allowing scientists to ensure reproducibility, revisit existing computational pipelines, and more easily share analyses and results. In addition, as the results of this work can impact policy, having provenance available is important for decision-making. In this paper we describe, UV-CDAT, a workflow-based, provenance-enabled system that integrates climate data analysis libraries and visualization tools in an end-to-end application, making it easier for scientists to integrate and use a wide array of tools.

The advent of coupled earth system models has raised an important question in parallel computing: What is the most effective method for coupling many parallel models to form one high-performance coupled modeling system? We present our solution to this problem--The Model Coupling Toolkit (MCT). We describe how our effort to construct the Next-Generation Coupler for NCAR Community Climate System Model motivated us to create the Toolkit. We describe in detail the conceptual design of the MCT, and explain its usage in constructing parallel coupled models. We present some preliminary performance results for the Toolkit's parallel data transfer facilities. Finally, we outline an agenda for future development of the MCT.

We describe several common problems that we discovered during our efforts to refactor several large geofluid applications that are components of the Community Climate System Model (CCSM) developed at the National Center for Atmospheric Research (NCAR). We stress tested the weak scalability of these applications by studying the impact of increasing both the resolution and core counts by factors of 10–100. Several common code design and implementations issues emerged that prevented the efficient execution of these applications on very large microprocessor counts. We found that these problems arise as a direct result of disparity between the initial design assumptions made for low resolution models running on a few dozen processors, and today’s requirements that applications run in massively parallel computing environments. The issues discussed include non-scalable memory usage and execution time in both the applications themselves and the supporting scientific data tool chains.

A Google search for the keyword ‘climate’ on a cold summer day in August 2010 delivered more than 150 million links in 0.23 s, and ‘climate change’ brought another 58 million. Obviously it is no problem to find floods of information about these topics on the net, yet understanding the scientific concept of climate and climate modelling is not so easy. The trouble with ‘climate’ starts when it is mixed up with the idea of weather, and when extreme weather events and short-term trends in temperature or precipitation are interpreted as effects of climate change. Usually, these interpretations are linked to an individual’s memory of experiences in childhood and other periods of life. But the trouble results not from this individual definition, which does not accord with the World Meteorological Organization’s official definition of climate as the statistics of weather. The trouble is raised by the scientific concept of climate as a mathematical construct that cannot be experienced directly. This problem is hitting science now that socio-political demands are coming into play. For responding to such demands, science has to break down its statistical and general concepts into individual and local conclusions, but this is—at the moment at least—not possible. The reason lies in the top-down approach of modern science, which uses globally valid equations to achieve increasingly higher resolution. The great challenge for meteorology during the next years and decades will be to translate statistical and general results into individual and local knowledge. Or in other words, science has to connect its global view with local circumstances. Regional modelling and downscaling are just the beginning, although these methods are still far removed from any particular individual or local view of a particular city or area. Of course, one can ask why humans do not simply get used to the scientific concept of climate. But when concrete environmental activities are required, individual needs and local effects play the main role, not the annual mean global temperature.

The first workshop on Atmospheric and Oceanic Computational Science brings together computational and domain scientists who develop computational tools for the study of the atmosphere and oceans. These tools are essential for understanding and prediciting weather, air and water pollution, and the evolution of the planet's climate. The dynamics of the atmosphere and of the oceans is driven by a multitude of physical processes and is characterized by a multiple spatial and temporal scales. Moreover, the computations are very large scale: present day models track the time evolution of tens of millions to tens of billions variables. These factors make atmospheric and oceanic simulations a challenging, vibrant research field with a tremendous impact on society at large. Topics covered in this symposium include new methods for spatial and temporal discretization, parallel and high performance computing, advances with existing models, and data assimilation and observation targeting algorithms.

A framework for a fully implicit solution method is implemented into (1) the High Order Methods Modeling Environment (HOMME), which is a spectral element dynamical core option in the Community Atmosphere Model (CAM), and (2) the Parallel Ocean Program (POP) model of the global ocean. Both of these models are components of the Community Climate System Model (CCSM). HOMME is a development version of CAM and provides a scalable alternative when run with an explicit time integrator. However, it suffers the typical time step size limit to maintain stability. POP uses a time-split semi-implicit time integrator that allows larger time steps but less accuracy when used with scale interacting physics. A fully implicit solution framework allows larger time step sizes and additional climate analysis capability such as model steady state and spin-up efficiency gains without a loss in scalability. This framework is implemented into HOMME and POP using a new Fortran interface to the Trilinos solver library, ForTrilinos, which leverages several new capabilities in the current Fortran standard to maximize robustness and speed. The ForTrilinos solution template was also designed for interchangeability; other solution methods and capability improvements can be more easily implemented into the models as they are developed without severely interacting with the code structure. The utility of this approach is illustrated with a test case for each of the climate component models.

A global sea-ice modeling component of the Community Climate System Model was augumented with automatic differentiation (AD) technology. The numerical experiments were run with two problem sets of different grid sizes. Rigid ice regions with high viscous properties cause computational difficulty in the propagation of AD-based derivative computation. Pre-tuning step was required to obtain successful convergence behavior. Various thermodynamic and dynamic parameters were selected for multivariate sensitivity analysis. The major parameters controlling the sea-ice thickness/volume computation were ice and snow densitives, albedo parameters, thermal conductivities, and emissivity constant. Especially, the ice and snow albedo parameters are found to have stronger effect during melting seasons. This high seasonal variability of the thermodynamic parameters underlines the importance of the multivariate sensitivity approach in global sea-ice modeling studies.

The Espresso Scientific Modeling Interface (Espresso) is a scientific model productivity tool developed for climate modelers. Espresso was designed to be an extensible interface to both scientific models and Grid resources. It also aims to be a contemporary piece of software that relies on Globus.org's Java CoG Kit for a Grid toolkit, Sun's Java 2 API and is configured using XML. This article covers the design and implementation of Espresso's Grid functionality and how it interacts with existing scientific models. We give specific examples of how we have designed Espresso to perform climate simulations using the PSU/NCAR MM5 atmospheric model. Plans to incorporate the CCSM and FOAM climate models are also discussed.

In this paper we discuss the development of a high performance climate modeling system as an example of the application of Grid based technology to climate modeling. The climate simulation system at Argonne currently includes a scientific modeling interface (Espresso) written in Java which incorporates Globus middleware to facilitate climate simulations on the Grid. The climate modeling system also includes a high performance version of MM5v3.4 modified for long climate simulations on our 512 processor Linux cluster (Chiba City), an interactive web based tool to facilitate analysis and collaboration via the web, and an enhanced version of the Cave5D software capable of visualizing large climate data sets. We plan to incorporate other climate modeling systems such as the Fast Ocean Atmosphere Model (FOAM) and the National Center for Atmospheric Research's (NCAR) Community Climate Systems Model (CCSM) within Espresso to facilitate their application on computational grids.

The DOE Parallel Climate Model (PCM) is used to simulate the earth's climate system and has been used to study the climate of the 20th century and to project possible climate changes into the 21st century and beyond. It was designed for use on distributed memory, highly parallel, architectures. The computational requirements and design of the model are discussed, as well as its performance and scalability characteristics. A method for port validation is demonstrated. The shortcomings of the current model are summarized and future design plans are presented.