Current Challenges and Future Directions in Data Assimilation and Reanalysis

Valmassoi, Arianna; Keller, Jan D.; Kleist, Daryl; English, Stephen; Ahrens, Bodo; Ďurán, Ivan Bašták; Bauernschubert, Elisabeth; Bosilovich, Michael G.; Fujiwara, Masatomo; Hersbach, Hans; Lei, Lili; Löhnert, Ulrich; Mamnun, Nabir; Martin, Cory; Moore, Andrew M.; Niermann, Deborah; Ruiz, Juan; Scheck, Leonhard

doi:10.1175/bams-d-21-0331.1

Cited by 7 publications

(4 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More recently, machine learning (ML) methods have being used to support data assimilation systems (Düben et al, 2021;Valmassoi et al, 2022). Such methods can be applied directly to NWP with minimum design choice of predictor variables and are considered as promising to automate the processes, i.e., ML can be used as an emulator for some physical processes.…”

Section: Radiative Transfer Modelsmentioning

confidence: 99%

Focal-TSMP: Deep learning for vegetation health prediction and agricultural drought assessment from a regional climate simulation

Shams Eddin,

Gall

2023

Preprint

View full text Add to dashboard Cite

Abstract. In this study, we investigate applying deep learning (DL) models on a regional climate simulation produced by the Terrestrial Systems Modelling Platform (TSMP Ground to Atmosphere G2A) for vegetation health modeling and agricultural drought assessment. The TSMP simulation is performed in a free mode and the DL model is used in an intermediate step to synthesize Normalized Difference Vegetation Index (NDVI) and Brightness Temperature (BT) images from the TSMP simulation over Europe. These predicted images are then used to derive different vegetation and drought indices like NDVI anomaly, BT anomaly, Vegetation Condition Index (VCI), Thermal Condition Index (TCI), and Vegetation Health Index (VHI). To ensure reliability and to assess the model applicability with different seasonality and spatial variability, we provide an analysis of model biases and uncertainties across different regions over the Pan-Europe domain. We further provide an analysis about the contribution of the input variables from the TSMP model components to ensure a better understanding of the model prediction. A comprehensive evaluation on the long-term TSMP using reference remote sensing data showed sufficiently good agreements between the model predictions and observations. While model performance varies on the test set between different climate regions, it achieves a mean absolute error (MAE) of 0.027 and 1.90 K° with coefficient of determination (R2) scores of 0.88 and 0.92 for NDVI and BT, respectively, at 0.11° resolution for sub-seasonal predictions. Our study could be used as a complimentary evaluation framework for climate change simulations with TSMP. Moreover, the developed DL model could be integrated with data assimilation and used for down-stream tasks, i.e., modelling the impact of extreme events on vegetation responses with different climate change scenarios. In summary, we demonstrate the feasibility of using DL on a TSMP simulation to synthesize NDVI and BT, which can be used for agricultural drought forecasting. Our implementation is publicly available at the project page (https://hakamshams.github.io/Focal-TSMP).

show abstract

Section: Radiative Transfer Modelsmentioning

confidence: 99%

Focal-TSMP: Deep learning for vegetation health prediction and agricultural drought assessment from a regional climate simulation

Shams Eddin,

Gall

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…They provide local point‐scale weather information for the past (He & Kammen, 2014; Ramirez‐Vergara et al., 2023), while especially wind might show a pronounced spatial variability (Santos‐Alamillos et al., 2014; Yu et al., 2015), and both wind and solar might show pronounced changes in the future related to climate change (Yang et al., 2022). Data from meteorological station networks often serve as ground truth in REP‐related model validations or as assimilated observations in forecasts and reanalyzes (Valmassoi et al., 2023).…”

Section: Introductionmentioning

confidence: 99%

Convection‐Permitting ICON‐LAM Simulations for Renewable Energy Potential Estimates Over Southern Africa

Chen,

Poll,

Hendricks Franssen

et al. 2024

JGR Atmospheres

View full text Add to dashboard Cite

Renewable energy is recognized in Africa as a means for climate change mitigation, but also to provide access to electricity in sub‐Saharan Africa, where three‐quarters of the global population without electricity resides. Reliable and highly resolved renewable energy potential (REP) information is indispensable to support power plants expansion. Existing atmospheric data sets over Africa that are used for REP estimates are often characterized by data gaps, or coarse resolution. With the aim to overcome these challenges, the ICOsahedral Nonhydrostatic (ICON) Numerical Weather Prediction (ICON‐NWP) model in its Limited Area Mode (ICON‐LAM) is implemented and run over southern Africa in a hindcast dynamical downscaling setup at a convection‐permitting 3.3 km horizontal resolution. The simulation time span covers contrasting solar and wind weather years from 2017 to 2019. To assess the suitability of the novel simulations for REP estimates, the simulated hourly 10 m wind speed (sfcWind) and hourly surface solar irradiance (rsds) are extensively evaluated against a large compilation of in situ observations, satellite, and composite data products. ICON‐LAM reproduces the spatial patterns, temporal evolution, the variability, and absolute values of sfcWind sufficiently well, albeit with a slight overestimation and a mean bias (mean error (ME)) of 1.12 m s−1 over land. Likewise the simulated rsds with an ME of 50 W m−2 well resembles the observations. This new ICON simulation data product will be the basis for ensuing REP estimates that will be compared with existing lower resolution data sets.

show abstract

“…Efforts are ongoing to improve the exploitation of satellite observations that are currently underutilized, both in terms of assimilating already operational data under all conditions and over all surfaces and in using channels that are not yet directly assimilated at all (Valmassoi et al, 2022;Hu et al, 2022). Solar satellite channels fall into the latter category, mostly because sufficiently fast and accurate forward operators are missing or have only become available recently.…”

Section: Introductionmentioning

confidence: 99%

A neural network-based method for generating synthetic 1.6 μm near-infrared satellite images

Baur

Scheck

Stumpf

et al. 2023

Preprint

View full text Add to dashboard Cite

Abstract. This study presents an extension to the method for fast satellite image synthesis (MFASIS) to allow simulating reflectances for the 1.6 μm near-infrared channel based on a computationally efficient neural network with improved accuracy. Such a fast forward operator enables using 1.6 μm channels from different satellite instruments in applications like model evaluation or operational data assimilation. It thus paves the way for the exploitation of additional information at this frequency, e.g. on cloud phase and particle sizes, which is complementary to the visible and thermal infrared range. To achieve similar accuracy for 1.6 μm NIR as for the visible channels 0.4–0.8 μm, it is important to represent vertical gradients of effective cloud particle radii, as well as mixed-phase clouds and molecular absorption. A comprehensive dataset sampled from IFS forecasts is used to develop the method. A new approach for describing the complex vertical cloud structure with a two layer model of water, ice and mixed-phase clouds optimized to obtain small reflectance errors is described and the relative importance of the different input parameters describing the idealized profiles is analyzed. Additionally, a different parameterization of the effective water and ice particle radii was used for testing. Further evaluation uses a month of ICON-D2 hindcasts with effective radii directly determined by the two-moment microphysics scheme of the model. The fast neural network approach itself does not add any significant additional error compared to the profile simplifications. In all cases, the mean absolute reflectance error achieved is about 0.01 or smaller, which is an order of magnitude smaller than typical differences between reflectance observations and corresponding model values.

show abstract

Current Challenges and Future Directions in Data Assimilation and Reanalysis

Cited by 7 publications

References 16 publications

Focal-TSMP: Deep learning for vegetation health prediction and agricultural drought assessment from a regional climate simulation

Focal-TSMP: Deep learning for vegetation health prediction and agricultural drought assessment from a regional climate simulation

Convection‐Permitting ICON‐LAM Simulations for Renewable Energy Potential Estimates Over Southern Africa

A neural network-based method for generating synthetic 1.6 μm near-infrared satellite images

Contact Info

Product

Resources

About