FastOpt (Germany)

In satellite remote sensing of land surface Essential Climate Variables (ECVs) using optical sensors, an atmospheric correction step is typically required to convert top-of-atmosphere (TOA) bi-directional reflectances into top-of-canopy (TOC) bi-directional reflectances. We analyse the error covariance structure of TOC reflectances that arises specifically from uncertainties in atmospheric correction. Using SMAC as the atmospheric correction model and Automatic Differentiation (AD) for efficient Jacobian computation, we quantify these error covariances across different scenarios. Our results show that uncertainties in the atmospheric state introduce non-negligible error correlations of both signs between different satellite bands. Since these error correlations are often overlooked, explicitly accounting for them could improve the accuracy of land surface ECV retrievals.

Meaningful study and quantification of climatic trends requires decades of observational data. Within satellite remote sensing this motivates the generation fundamental climate data records (FCDRs) - datasets containing the combined observations from a series of missions of a given sensor, to span the required duration for study. Ensuring the radiometric stability of such series is vital for such applications. To this end a consistent in-flight retrospective recalibration of all the sensors in such a series is required, called harmonisation. Presented here is a methodology for achieving this by analysing match-ups between sensors in the series and applicable sensors with modern, well-calibrated reference sensors, fitting new calibration parameters for each sensor in the series. Such a problem is not tractable in the most optimal, metrologically rigorous manner by existing solvers, due to the possibly complex error and geophysical correlation structures and high data volume (potentially >10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">8</sup> match-ups). Discussed are a palette of novel optimisation algorithms developed to overcome this, which investigate alternative approaches to handling the full problem. Preliminary results are shown for one these approaches, <i xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Fast EIV</i> (Errors in Variables), for the recalibration of the AVHRR sensor series.

The past two decades saw a steady decrease of summer Arctic sea ice extent. The 2007 value was yet considerably lower than expected from extrapolating the long‐term trend. We present a quantitative analysis of this extraordinary event based on the adjoint of a coupled ocean‐sea ice model. This new approach allows to efficiently assess the sensitivity of the ice‐covered area in September 2007 with respect to any potential influence factor. We can trace back 86% of the ice area reduction to only four of these factors: May and June wind conditions, September 2‐meter temperature, and March ice thickness. Two thirds of the reduction are determined by factors that are already known at the end of June, suggesting a high potential for an early prediction.

We present a computer‐efficient software package enabling us to assimilate operational remote‐sensing flux products into a state‐of‐the‐art two‐stream radiation transfer scheme suitable for climate models. This package implements the adjoint and Hessian codes, generated using automatic differentiation techniques, of a cost function balancing (1) the deviation from the a priori knowledge on the model parameter values and (2) the misfit between the observed remote‐sensing fluxes and the two‐stream model simulations. The individual weights of these contributions are specified notably via covariance matrices of the uncertainties in the a priori knowledge on the model parameters and the measurements. The proposed procedure delivers a Gaussian approximation of the PDFs of the retrieved model parameter values. The a posteriori covariance matrix is further exploited to evaluate, in turn, the posterior probability density functions of the radiant fluxes simulated by the two‐stream model, including those that are not measured, for example, the fraction of radiation absorbed in the ground. Applications are conducted using Moderate Resolution Imaging Spectroradiometer (MODIS) and Multiangle Imaging Spectroradiometer (MISR) broadband surface albedo products. It turns out that the differences between these two albedo sets may translate into discernible signatures on some retrieved model parameters. Meanwhile, adding the Joint Research Centre (JRC)‐Fraction of Absorbed Photosynthetically Active Radiation (FAPAR) Sea‐viewing Wide Field‐of‐view Sensor (SeaWiFS) products into the measurements yields a significant reduction of uncertainties. Results from these applications indicate that the products retrieved from the two‐stream inversion procedure (1) exhibit much less variability than those generated by the operational algorithms for the LAI and FAPAR, and (2) are in good agreement with the available ground‐based estimates.

The NASA finite-volume General Circulation Model (fvGCM) is a three-dimensional Navier-Stokes solver that is being used for quasi-operational weather forecasting at NASA/GMAO. By means of the automatic differentiation tool TAF, efficient tangent linear and adjoint versions are generated from the Fortran-90 source code of fvGCM’s dynamical core. fvGCM’s parallelisation capabilities based on OpenMP and MPI have been transferred to the tangent linear and adjoint codes. For OpenMP, TAF automatically inserts corresponding OpenMP directives in the derivative code. For MPI, TAF generates interfaces to hand-written tangent linear and adjoint wrapper routines. TAF also generates a scheme that allows the tangent linear and adjoint models to linearise around an external trajectory of the model state. The generation procedure is set up in an automated way, allowing quick updates of the derivative codes after modifications of fvGCM.

International audience