Hello, A new project on the LCRC cluster has been requested. Please forward the information on to the LCRC Allocation sub-committee. Applicant's name: Mihai Anitescu Applicant's institution: ANL Applicant's division: MCS Project Name: UQCLM Project title: Uncertainty Quantification for Land Model in Climate Science Associated funding: Climate Science for a Sustainable Energy Future; PI: Mihai Anitescu; site PI: Ian Foster Other Systems: Science: In this project we study uncertainty quantification (UQ) methods applied to Community Land Model (CLM). CLM is the land model for the Community Earth System Model (CESM) and the Community Atmosphere Model (CAM). It consists of several sets of large and nonlinear complex systems. We aim to develop intrusive uncertainty quantification methods and to demonstrate their computational advantage over a differentiated version of CLM with automatic differentiation (OpenAD) which will open the way for gradient-based optimization for calibration in land testbed. Project description: The overall UQ approach comprises three parts: sensitivity characterization, uncertainty model calibration, and uncertainty propagation, each of which is computationally intensive. The first two tasks require the creation and use of derivative information. Our scientific goal is to applied the developed UQ tools to the land model. First, we will characterize the sensitivity of the outputs to model parameters, for example, the physiology inputs for vegetation and atmosphere forcing. By using OpenAD, we can use derivative information for obtaining the sensitivity and guide the choice of statistical models. Subsequently, we will use a parallel gradient-based optimization tool built around the stochastic programming code PIPS to do calibration of uncertain parameters with local data sets. After that we can use those tuned parameters to quantify prediction and finally, make overall estimate of prediction uncertainty. This approach is implemented as an iterative process, repeated as new observational data set become available. By doing so, land UQ results will be useful in guiding subsequent observational campaigns, leading to improved estimates of parameters with maximal impact on reducing the prediction uncertainty. We plan first to do comparisons between finite difference intrusive approaches versus a Monte Carlo approach for single point models and initial assessment of AD application to CLM. We will also make comparison between finite difference intrusive sensitivity analysis approaches versus a Monte Carlo sensitivity analysis approach for single point models. After validations for single points, we will proceed at the global gridded scale and coupled clm models. This project requires intensive computations due to the ensemble nature of the calculation and to the need to make assessments on steady-state quantities, which requires runs to stabilization. We need to prototype and test the interface with the OpenAD tool and test our numerical codes for runs on a global scale of more than 1,000 years with hundreds of perturbed parameters. Project URL: Requested allocation: 90000 Justification: The requester has used 0 hours of their initial startup project. In addition to approving an initial amount, please specify a Category and Subcategory for this project. For a list of the current selection of approved categories, please see: https://wiki.lcrc.anl.gov/wiki/Processes/Categories Once the Allocation committee has approved the project, please go to the Project Management page to create it: https://accounts.lcrc.anl.gov/projects.php Thank You, The LCRC Accounts System