Search

Developed at the National Oceanic and Atmospheric Administration (NOAA) and the Joint Center for Satellite Data Assimilation (JCSDA), the Community Global Observing System Simulation Experiment (OSSE) Package (CGOP) provides a vehicle to quantitatively evaluate the impacts of emerging environmental observing systems or emerging in situ or remote sensing instruments on NOAA numerical weather prediction (NWP) forecast skill. The typical first step for the OSSE is to simulate observations from the so-called nature run. Therefore, the observation spatial, temporal, and view geometry are needed to extract the atmospheric and surface variables from the nature run, which are then input to the observation forward operator (e.g., radiative transfer models) to simulate the new observations. This is a challenge for newly proposed systems for which instruments are not yet built or platforms are not yet deployed. To address this need, this study introduces an orbit simulator to compute these parameters based on the specific hosting platform and onboard instrument characteristics, which has been recently developed by the NOAA Center for Satellite Applications and Research (STAR) and added to the GCOP framework. In addition to simulating existing polar-orbiting and geostationary orbits, it is also applicable to emerging near-space platforms (e.g., stratospheric balloons), cube satellite constellations, and Tundra orbits. The observation geometry simulator includes not only passive microwave and infrared sounders but also global navigation satellite system/radio occultation (GNSS/RO) instruments. For passive atmospheric sounders, it calculates the geometric parameters of proposed instruments on different platforms, such as time varying location (latitude and longitude), scan geometry (satellite zenith and azimuth angles), and ground instantaneous field of view (GIFOV) parameters for either cross-track or conical scanning mechanisms. For RO observations, it determines the geometry of the transmitters and receivers either on satellites or stratospheric balloons and computes their slant paths. The simulator has been successfully applied for recent OSSE studies (e.g., evaluating the impacts of future geostationary hyperspectral infrared sounders and RO observations from stratospheric balloons). Significance Statement An orbit simulator for satellite and near-space platforms for supporting observing system simulation experiments (OSSE) is developed in this study. It can compute spatial, temporal, and view geometry parameters of the new observations based on the specific hosting platform and onboard instrument characteristics. These parameters are important for the OSSE to extract atmospheric profiles and surface properties from nature run data and simulate new observations. In addition to simulating existing polar-orbiting and geostationary orbits, it is applicable to emerging near-space platforms (e.g., stratospheric balloons), cube satellite constellations, and Tundra orbits. The observation geometry simulator includes not only passive microwave and infrared sounders but also radio occultation instruments.

In this article, we leverage and apply state-of-the-art artificial intelligence (AI) techniques to satellite remote sensing of temperature, moisture, surface, and cloud parameters in all-weather, all-surface conditions, from both microwave and infrared sensors. The multi-instrument inversion and data assimilation preprocessing system, artificial intelligence version, or MIIDAPS-AI for short, is valid for both polar and geostationary microwave and infrared sounders and imagers as well as for pairs of combined infrared and microwave sounders. The algorithm produces vertical profiles of temperature and moisture as well as surface temperature, surface emissivity, and cloud parameters. Additional products from hyperspectral infrared sensors include selected trace gases. From microwave sensors, additional products such as rainfall rate, first year/multiyear sea ice concentration, and soil moisture can be derived from primary products. The MIIDAPS-AI algorithm is highly efficient with no noticeable decrease in accuracy compared to traditional operational sounding algorithms. The automatically generated Jacobians from this deep-learning algorithm could provide an explainability mechanism to build trustworthiness in the algorithm, and to quantify uncertainties of the algorithm's outputs. The computation gain is estimated to be two orders of magnitude, which opens the door to either 1) process massively larger amounts of satellite data, or to 2) offer improvements in timeliness and significant saving in computing power (and therefore cost) if the same amount of data is processed. Here, we present an overview of the MIIDAPS-AI implementation, discuss its applicability to various sensors and provide an initial performance assessment for a select number of sensors and geophysical parameters.

Developing trustworthy artificial intelligence for weather and ocean forecasting, as well as for long-term environmental sustainability, requires integrating collaborative efforts from many sources.

In 2011, the National Oceanic and Atmospheric Administration (NOAA) began a cooperative initiative with the academic community to help address a vexing issue that has long been known as a disconnection between the operational and research realms for weather forecasting and data assimilation. The issue is the gap, more exotically referred to as the “valley of death,” between efforts within the broader research community and NOAA’s activities, which are heavily driven by operational constraints. With the stated goals of leveraging research community efforts to benefit NOAA’s mission and offering a path to operations for the latest research activities that support the NOAA mission, satellite data assimilation in particular, this initiative aims to enhance the linkage between NOAA’s operational systems and the research efforts. A critical component is the establishment of an efficient operations-to-research (O2R) environment on the Supercomputer for Satellite Simulations and Data Assimilation Studies (S4). This O2R environment is critical for successful research-to-operations (R2O) transitions because it allows rigorous tracking, implementation, and merging of any changes necessary (to operational software codes, scripts, libraries, etc.) to achieve the scientific enhancement. So far, the S4 O2R environment, with close to 4,700 computing cores (60 TFLOPs) and 1,700-TB disk storage capacity, has been a great success and consequently was recently expanded to significantly increase its computing capacity. The objective of this article is to highlight some of the major achievements and benefits of this O2R approach and some lessons learned, with the ultimate goal of inspiring other O2R/R2O initiatives in other areas and for other applications.

The objective of this research is to evaluate the potential for CubeSats and other small satellite constellations to mitigate a potential data gap and/or improve numerical weather forecasts. This is conducted as a collaborative project involving NOAA AOML and NESDIS/STAR, CIMSS, and the University of MD. To this end, a series of Observing System Simulation Experiments (OSSEs) is being conducted. These experiments utilize both global and regional OSSE systems. Specific instruments being evaluated include the CubeSat Infrared Atmospheric Sounder (CIRAS), the Micro-sized Microwave Atmospheric Sounder-2 (MicroMAS-2), the Cyclone Global Navigation Satellite System (CYGNSS), and the Time-Resolved Observations of Precipitation and storm Intensity with a Constellation of Smallsats (TROPICS) mission. The diverse set of OSSEs that have been performed to date show potential for each of these systems to mitigate a data gap and to contribute to improved prediction and begin to quantify their relative impacts.

Test beds have emerged as a critical mechanism linking weather research with forecasting operations. The U.S. Weather Research Program (USWRP) was formed in the 1990s to help identify key gaps in research related to major weather prediction problems and the role of observations and numerical models. This planning effort ultimately revealed the need for greater capacity and new approaches to improve the connectivity between the research and forecasting enterprise. Out of this developed the seeds for what is now termed “test beds.” While many individual projects, and even more broadly the NOAA/National Weather Service (NWS) Modernization, were successful in advancing weather prediction services, it was recognized that specific forecast problems warranted a more focused and elevated level of effort. The USWRP helped develop these concepts with science teams and provided seed funding for several of the test beds described. Based on the varying NOAA mission requirements for forecasting, differences in the organizational structure and methods used to provide those services, and differences in the state of the science related to those forecast challenges, test beds have taken on differing characteristics, strategies, and priorities. Current test bed efforts described have all emerged between 2000 and 2011 and focus on hurricanes (Joint Hurricane Testbed), precipitation (Hydrometeorology Testbed), satellite data assimilation (Joint Center for Satellite Data Assimilation), severe weather (Hazardous Weather Testbed), satellite data support for severe weather prediction (Short-Term Prediction Research and Transition Center), mesoscale modeling (Developmental Testbed Center), climate forecast products (Climate Testbed), testing and evaluation of satellite capabilities [Geostationary Operational Environmental Satellite-R Series (GOES-R) Proving Ground], aviation applications (Aviation Weather Testbed), and observing system experiments (OSSE Testbed).

The NOAA Microwave Integrated Retrieval System (MiRS) retrieves the atmospheric profiles of state variables (temperature, moisture, liquid cloud, hydrometeors, surface emissivity spectrum and skin temperature) simultaneously and coherently with one-dimension Variational (1DVar) method based on passive microwave radiance observation from multiple polar-orbiting operational satellites, including NOAA18, NOAA19, MetOp (A and B), DMSP (F16 and F18) and SNPP. In MiRS, the geophysical consistency between the retrieved state variables is obtained through the constraint from the state variable background covariance matrices, which define the variability of each state variable and the relationship between them. The current atmospheric covariance matrices were mainly built based on the ECMWF 60 layer sample data set (EC60). Because no rain data available in EC60, the MM5 model output rain was used as a substitute. Although MiRS performs well based on current matrices, obviously, there is room for improvement if the matrices could be rebuilt based more representative, higher resolution dataset, and most importantly, all variables are from one datasets, so the cross relationship between parameters could be properly represented. The ECMWF IFS-137 dataset (EC137) is the most recent ECMWF sample data set. Besides several major modifications and improvements in ECMWF forecasting system, comparing with EC60, the profiles in EC137 have more than doubled vertical resolution. The hydrometeor variables are included in EC137, rather than in EC60. With upgraded sampling method, the profiles in EC137 are more evenly distributed in both temporal and spatial domains. And the profile population in EC137 shows significant different statistic characters than its precedents. To generate the new matrices, EC137 data is interpolated from 137 layers to 100 layers. The variable units are transferred to be used for radiation transfer calculation and match with those used in MiRS. For example, the unit of rain and snow is transferred from mass flux (Kg/(m2*s)) to water content (Kg/m2) with the assumption of rain (snow) drop speed as 4.0 (1.0) m/s. Comparing the matrix based on EC137 and that of current MiRS, except the variation in details, there are two significant distinct features are noticeable. First, the snow variable becomes effective in EC137 matrix because the snow variable is provided in EC137 rather than in EC60 and MM5. Second, the covariance between rain and other variables becomes meaningful in EC137 matrix. In the old matrix, rain is not well correlated with other variables, as rain data is from MM5, but other parameters are from EC60, two unrelated datasets. All these differences imply different, most likely better performance of MiRS system if employing the background covariance matrices based on EC137. Tuning is conducted after implementation of the new EC137 matrices. And the impact of the new matrices is assessed by comparing the retrieval products based on the new and old matrices, and with colloated dropsonde, radiosonde and ECMWF analysis profiles.

Recently, the Optimal Spectral Sampling (OSS) method was implemented in a development version of the Community Radiative Transfer Model (CRTM) at JCSDA. This presentation describes the way that the OSS is implemented in CRTM, and some preliminary evaluation of the performance of the CRTM-OSS in comparison with CRTM-ODPS method. One of the important benefits of the OSS method is its capability to simulate unapodized radiance, which could lead to better sensitivity in retrievals and data assimilation for numerical weather prediction (NWP). The accuracy and speed of the OSS simulations are two trade-off factors, depending on how many spectral nodes are used during training and simulation. In general, the OSS simulations of hyper spectral IR sensors, such as IASI and CrIS, can archive similar accuracy and double speed as compared to the CRTM-ODPS simulations.

To retrieve microwave land emissivity, infrared surface skin temperatures have been used as surface physical temperature since there is no global information on physical vegetation/soil temperature profiles. However, passive microwave emissions originate from deeper layers with respect to the skin temperature. So, this inconsistency in sensitivity depths between skin temperatures and microwave temperatures may introduce a discrepancy in the determined emissivity. Previous studies showed that this inconsistency can lead to significant differences between day and night retrievals of land emissivity which can exceed 10%. This study proposes an approach to address this inconsistency and improve the retrieval of land emissivity using microwave observations from Advanced Microwave Scanning Radiometer–Earth Observing System (AMSR-E). The diurnal cycle of the microwave brightness temperature ( Tb ) was constructed over the globe for different frequencies/polarizations using a constellation of satellites. Principal component analysis (PCA) was conducted to evaluate the spatial variation of the Tb diurnal cycle. The diurnal amplitudes of microwave temperatures observed in desert areas were not consistent with the larger amplitudes of the diurnal cycle of skin temperature. Densely vegetated areas with more moisture have shown smaller amplitudes. A lookup table of effective temperature ( T eff ) anomalies is constructed based on the Tb diurnal cycle to resolve the inconsistencies between infrared and Tb diurnal variation. This lookup table of T eff anomalies is a weighted average over the layers contributing to the microwave signal, for each channel and month. The integration of this T eff in the retrieval of land emissivity reduced the differences between day and night retrieved emissivities to less than 0.01 for AMSR-E observations.

The proposed Temporal Experiment for Storms and Tropical Systems (TEMPEST) satellite mission addresses key science needs related to cloud and precipitation processes using a constellation of five CubeSats with identical five-frequency millimeter-wave radiometers spaced 5–10 minutes apart in orbit. This CubeSat constellation will directly observe the time evolution of clouds to study the conditions that control the transition of clouds to precipitation. The TEMPEST millimeter-wave radiometers will penetrate into the cloud to directly observe changes as the cloud begins to precipitate or ice accumulates inside the storm. TEMPEST provides observations at five millimeter-wave frequencies from 90 to 183 GHz using a single compact instrument that is well suited for a 6U CubeSat architecture and fits well within the NASA CubeSat Launch Initiative capabilities.