SI2-SSI: Lightweight Infrastructure for Land Atmosphere Coupling (LILAC): A Tool for Easy Integration of the Community Land Model into Multiple Modeling Systems

Colorado State University, Fort Collins CO

Investigators

Abstract

Everyone on Earth lives on the ground, so interactions between the atmosphere and the land surface are a critically important part of the climate system and understanding these interactions is important for weather prediction, agriculture, and urban water management. Earth System Models (ESMs) are complex software systems representing complex natural systems. They are critical tools for diagnosing, understanding, and predicting interactions and change in the atmosphere, oceans, and land ecosystems. This project will develop a new software system for coupling the land and atmosphere components of Earth System models, specifically for the most widely-used climate model in the world: the Community Earth System Model (CESM). The new system will be capable of simulating climates near the ground including exchanges of heat, water ,and carbon between vegetated land and the air as well as streamflow and soil moisture. As an officially-supported component of CESM, it will be used by thousands of scientists and students around the world. Unlike its predecessor, the new system will be able to simulate small areas at high resolution for important applications and testing. The project will also support a computer science graduate student as well as academics and scientists who will help develop and test the software. The PIs will engage a global community of software developers and users through a series of workshops and webinars as well as through professional societies and publications. The PIs recognize the chronic under-representation of women and ethnic minorities in both Computer Science and Atmospheric Science. To address this, they will host a computer science summer camp for middle school students from underrepresented groups (URGs), and close collaboration with existing climate courses for K-12 teachers, science outreach in K-12 schools, and a highly successful REU-Site operated by the PI. This project will dramatically improve the usability of the most widely-used climate model in existence. Hundreds of developers and thousands of users around the world will benefit from the far greater flexibility and use cases for this model. The Community Land Model will be coupled to a much greater range of atmospheric codes for weather prediction, air quality applications, and climate projections that enhance the quality of life for people everywhere. The Community Earth System Model CESM) is a uniquely open ESM with a distributed community of hundreds of developers and thousands of users around the world. Compared to other ESMs, CESM is the "Linux of climate models." The land component of CESM, the Community Land Model (CLM), has its own vibrant and diverse user and development community, which has supported the construction of a particularly comprehensive terrestrial system model. It includes a rich array of processes that enable examination of the physical, biological, and chemical processes by which natural terrestrial ecosystems and human-managed land affect and are affected by climate and weather. CESM can only be run globally, but there is widespread interest in coupling CLM to alternative high-resolution atmosphere models to study the challenging scientific problems that exist at the interface between small and large spatial scales. A barrier to this research, however, is that CLM cannot readily be coupled to alternative atmosphere models. The CLM coupling interface only supports communication with the CESM coupling infrastructure (CPL7), which imposes strict requirements on how an atmospheric component can communicate with CLM. One key requirement is for the atmosphere model to complete a full time step before coupling can occur. This requirement necessitates significant refactoring of many atmospheric models in order for them to couple to CLM via CPL7. In addition, the tools to build and configure CLM are currently difficult to use outside of the CESM context. For all of these reasons, CLM has never been coupled to alternative atmospheric models in a sustainable fashion. Colorado State University (CSU), in partnership with the National Center for Atmospheric Research (NCAR), proposes to develop a Lightweight Infrastructure for Land-Atmosphere Coupling (LILAC) that will significantly simplify the coupling of CLM to alternative atmospheric models. The LILAC coupler and the associated proposed streamlining and simplification of the CLM tool chain will be developed and tested with a prototype high-resolution atmosphere model, the CSU SAM Cloud Resolving Model, and will be extended for use with any arbitrary Target Atmosphere Model. The development of LILAC and associated tools will enable numerous groups to couple CLM to their atmospheric models, and to quickly update to new state-of-the-art CLM model versions as they become available. This will open up new avenues of land-atmosphere research that can exploit the combined biogeophysical and biogeochemical capabilities of CLM with the strengths of high-resolution atmosphere models. It will enable research into land-atmosphere interactions across a variety of scales, ranging from turbulence-resolving simulations of tower and aircraft data to cloud-resolving simulations to study how small-scale land features such as hillslopes, valleys, lakes, rivers, urban areas and farms conspire to alter surface climate and atmospheric boundary layer characteristics to continental-scale simulations of the impact of land cover and land use change on weather and climate. This project is supported by the Office of Advanced Cyberinfrastructure in the Directorate for Computer & Information Science and Engineering and the Office of Polar Programs and Division of Atmospheric and Geospace Science in the Directorate for Geosciences.