SEARCH · Engineering Papers

Results for “forecast error”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 955 records · Page 53

All sources NASA NTRS DOE OSTI

NASA NTRSDate not supplied

Implementing JEDI into NASA GMAO’s Real Time Production Suite

NASA’s Global Modeling and Assimilation Office (GMAO) has prepared their first production system involving the Joint Effort for Data assimilation Integration (JEDI) framework. In this system the central analysis, that drives the deterministic forecast, will be provided using JEDI. This talk outlines the phased approach to implementing JEDI into production that GMAO has designed, and how this approach will allow for a careful analysis of the system against the existing data assimilation framework (GSI). In the first phase of implementation the existing data assimilation system will perform certain actions that are still under development in JEDI. These include thinning the observations and producing satellite bias correction coefficients. JEDI is hooked up to the existing workflow so a single line switch can activate whether the existing or JEDI-based analysis is cycled. Outside of the monumental effort to construct JEDI that is ongoing at the Joint Center for Satellite Data Assimilation (JCSDA), GMAO have undertaken two areas of considerable effort. The talk will describe these efforts and highlight the main challenges that have been encountered. The first area of work is to implement the background error model from the existing data assimilation system into JEDI. The second is to validate the observing system in JEDI against the one in GSI, which has involved several new features being added to the observation operators in JEDI. While the longer-term plans involve trying to improve on the GSI in these two areas, GMAO is keen to have JEDI start from a trusted baseline. This is also key to implementing JEDI quickly so other priorities, such as increasing the number of model levels, can be easily worked on in parallel. GMAO is actively working on a framework to shepherd in the next generation coupled data assimilation system and model. As JEDI is implemented for the first time the plan is to ambitiously cycle through implementations, frequently bringing JEDI features to production. Details of these plans will be given in the talk and we will highlight key implementation and product milestones that we hope to achieve, as well as touch on the development environment that we will use to support frequent refreshing of the production system.

JEDI↗

NASA NTRS2001

DAO's Next Generation Physical-Space/Finite-Volume Data Assimilation System: Formulation and Initial Evaluation

The Physical-space/Finite-volume Data Assimilation System (fvDAS) is the next generation global atmospheric data assimilation system in development at the Data Assimilation Office at NASA's Goddard Space Flight Center. It is based on a new finite-volume general circulation model jointly developed by NASA and NCAR, and on the Physical-Space Statistical Analysis System (PSAS) developed at the DAO. In this talk we will describe the general system formulation, the adaptive quality control and general aspects of the error covariance modeling. The NASA-NCAR GCM is a completely new model which replaces the CEOs GCM used in the previous GEOS-1/2/3 Data Assimilation systems. A particular configuration of adaptive Statistical Quality Control and the Physical-space Statistical Analysis System (PSAS) are currently implemented in DAO's operational Data Assimilation System. However, the unique finite-volume formulation of the NASA-NCAR GCM, combined with the generality of the observation-space formulation of PSAS, provides for a very simple and accurate model-analysis interface. The system assimilates a variety of conventional and satellite observations. In particular, TOVS Level 1B radiances are assimilated using a 1-D variational scheme, both in clear sky and cloudy conditions. Computationally, the fvDAS runs approximately 10 times faster than the operational GEOS-Terra system. We will show that the next-generation fvDAS has much improved observation-minus-6hr forecast (O-F) statistics, as well as 5-day forecast skills. Top of the atmosphere radiation fields are in closer agreement with CERES measurements, with realistic precipitation and moisture fields. We will also show that the finite-volume formulation of the fvDAS produce assimilated fields which are more suitable for driving constituent transport models.

daSilva, A.↗

DOE OSTI2020

Reinforcement learning based automated history matching for improved hydrocarbon production forecast

History matching aims to find a numerical reservoir model that can be used to predict the reservoir performance. An engineer and model calibration (data inversion) method are required to adjust various parameters/properties of the numerical model in order to match the reservoir production history. In this study, we develop deep neural networks within the reinforcement learning framework to achieve automated history matching that will reduce engineers’ efforts, human bias, automatically and intelligently explore the parameter space, and remove the need of large set of labeled training data. To that end, a fast-marching-based reservoir simulator is encapsulated as an environment for the proposed reinforcement learning. The deep neural-network-based learning agent interacts with the reservoir simulator within reinforcement learning framework to achieve the automated history matching. Reinforcement learning techniques, such as discrete Deep Q Network and continuous Deep Deterministic Policy Gradients, are used toth, used to train the learning agents. The continuous actions enable the Deep Deterministic Policy Gradients to explore more states at each iteration in a a learning episode; consequently, a better history matching is achieved using this algorithm as compared to Deep Q Network. For simplified dual-target composite reservoir models, the best history-matching performances of the discrete and continuous learning methods in terms of normalized root mean square errors are 0.0447 and 0.0038, respectively. Furthermore, our study shows that continuous action space achieved by the deep deterministic policy gradient drastically outperforms deep Q network.

42 ENGINEERING↗

DOE OSTI2022

Fermi–Löwdin orbital self-interaction correction of adsorption energies on transition metal ions

Density functional theory (DFT)-based descriptions of the adsorption of small molecules on transition metal ions are prone to self-interaction errors. Here, we show that such errors lead to a large over-estimation of adsorption energies of small molecules on Cu + , Zn + , Zn 2+ , and Mn + in local spin density approximation (LSDA) and Perdew, Burke, Ernzerhof (PBE) generalized gradient approximation calculations compared to reference values computed using the coupled-cluster with single, doubles, and perturbative triple excitations method. These errors are significantly reduced by removing self-interaction using the Perdew–Zunger self-interaction correction (PZ-SIC) in the Fermi–Löwdin Orbital (FLO) SIC framework. In the case of FLO-PBE, typical errors are reduced to less than 0.1 eV. Furthermore, analysis of the results using DFT energies evaluated on self-interaction-corrected densities [DFT(@FLO)] indicates that the density-driven contributions to the FLO-DFT adsorption energy corrections are roughly the same size in DFT = LSDA and PBE, but the total corrections due to removing self-interaction are larger in LSDA.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

NASA NTRS2019

Assimilation of Sentinel 1 and SMAP–based satellite soil moisture retrievals into SWAT hydrological model: the impact of satellite revisit time and product spatial resolution on flood simulations in small basins

In runoff generation process, soil moisture plays an important role as it controls the magnitude of the flood events in response to the rainfall inputs. In this study, we investigated the ability of a new era of satellite soil moisture retrievals to improve the Soil & Water Assessment Tool (SWAT) daily discharge simulations via soil moisture data assimilation for two small (< 500 km2) and hydrologically different catchments located in Central Italy. We ingested 1) the Soil Moisture Active and Passive (SMAP) Enhanced L3 Radiometer Global Daily 9 km EASE-Grid soil moisture, 2) the Advanced SCATterometer (ASCAT) H113 soil moisture product released within the EUMETSAT Satellite Application Facility on Support to Operational Hydrology and Water Management (HSAF) which has a nearly daily temporal resolution and sampling of 12.5 km, and 3) a fused ASCAT/Sentinel-1 (S1) satellite soil moisture product named SCATSAR-SWI with temporal and spatial sampling of 1 day and 1 km, respectively into SWAT hydrological model via the Ensemble Kalman Filter (EnKF). Different configurations were tested with the aim of exploring the effect of the hydrological regime, the land use conditions, the spatial sampling and the revisit time of the products (which controls the amount of available data to be potentially ingested). Results show a general improvement of SWAT discharge simulations for all products in terms of error and Nash Sutcliffe efficiency index. In particular, we found a relatively good behavior of both the active and the passive products in terms of low flows improvement especially for the catchment characterized by a higher baseflow component. The benefit of the higher spatial resolution of SCATSAR-SWI obtained via S1 over ASCAT was small, likely due to very challenging areas for the S1 retrieval. Eventually, better performances were obtained for the passive product in the more forested catchment. With the aim of exploring the benefit of having more frequent satellite soil moisture observations to be ingested, we tested the performance of the ASCAT product with a reduced temporal sampling obtained by temporally matching ASCAT observations to that of SMAP. The results show a significant reduction of the performance of ASCAT, suggesting that the correction frequency (due to the higher number of observations available) for small catchments is an important aspect for improving flood forecasting as it helps to adjust more frequently the pre-storm soil moisture conditions.

Shima Azimi↗

NASA NTRS2007

Comparison of Different Global Information Sources Used in Surface Radiative Flux Calculation: Radiative Properties of the Surface

Direct estimates of surface radiative fluxes that resolve regional and weather-scale variabilty over the whole globe with reasonable accuracy have only become possible with the advent of extensive global, mostly satellite, datasets within the past couple of decades. The accuracy of these fluxes, estimated to be about 10-15 W per square meter is largely limited by the accuracy of the input datasets. The leading uncertainties in the surface fluxes are no longer predominantly induced by clouds but are now as much associated with uncertainties in the surface and near-surface atmospheric properties. This study presents a fuller, more quantitative evaluation of the uncertainties for the surface albedo and emissivity and surface skin temperatures by comparing the main available global datasets from the Moderate-Resolution Imaging Spectroradiometer product, the NASA Global Energy and Water Cycle Experiment Surface Radiation Budget project, the European Centre for Medium-Range Weather Forecasts, the National Aeronautics and Space Administration, the National Centers for Environmental Prediction, the International Satellite Cloud Climatology Project (ISCCP), the Laboratoire de Meteorologie Dynamique, NOAA/NASA Pathfinder Advanced Very High Resolution Radiometer project, NOAA Optimum Interpolation Sea Surface Temperature Analysis and the Tropical Rainfall Measuring Mission (TRMM) Microwave Image project. The datasets are, in practice, treated as an ensemble of realizations of the actual climate such that their differences represent an estimate of the uncertainty in their measurements because we do not possess global truth datasets for these quantities. The results are globally representative and may be taken as a generalization of our previous ISCCP-based uncertainty estimates for the input datasets. Surface properties have the primary role in determining the surface upward shortwave (SW) and longwave (LW) flux. From this study, the following conclusions are obtained. Although land surface albedos in the near near-infrared remain poorly constrained (highly uncertain), they do not cause too much error in total surface SW fluxes; the more subtle regional and seasonal variations associated with vegetation and snow are still on doubt. The uncertainty of the broadband black-sky SW albedo for land surface from this study is about 7%, which can easily induce 5-10 W per square meter uncertainty in (upwelling) surface SW flux estimates. Even though available surface (broadband) LW emissivity datasets differ significantly (3%-5% uncertainty), this disagreement is confined to wavelengths greater than 20 micrometers so that there is little practical effect (1-3 W per square meters) on the surface upwelling LW fluxes. The surface skin temperature is one of two leading factors that cause problems with surface LW fluxes. Even though the differences among the various datasets are generally only 2-4 K, this can easily cause 10-15 W per square meter uncertainty in calculated surface (upwelling) LW fluxes. Significant improvements could be obtained for surface LW flux calculations by improving the retrievals of (in order of decreasing importance): (1) surface skin temperature, (2) surface air and near-surface-layer temperature, (3) column precipitable water amount and (4) broadband emissivity. And for surface SW fluxes, improvements could be obtained (excluding improved cloud treatment) by improving the retrievals of (1) aerosols (from our sensitivity studies but not discussed in this work), and (2) surface (black-sky) albedo, of which, NIR part of the spectrum has much larger uncertainty.

Zhang, Yuanchong↗

DOE OSTI2023

Forecasts on the Dark Matter Density Profiles of Dwarf Spheroidal Galaxies with Current and Future Kinematic Observations

We forecast parameter uncertainties on the mass profile of a typical Milky Way dwarf spheroidal galaxy (dSph) using the spherical Jeans equation and Fisher matrix formalism. For a Draco-like system we show that radial velocity measurements for 1000 individual stars can constrain the mass contained within the effective radius of a dSph to within 5%. This is consistent with constraints extracted from current observational data. We compare two systems, a cusp and core, and demonstrate that a minimum sample of 100,000 (10,000) stars with both radial and proper motions measurements is required to disentangle their inner slopes at the 2σ (1σ) level. If using the log-slope measured at the half-light radius as a proxy for differentiating between a core or cusp slope, only 1000 line-of-sight and proper motions measurements are required; however, we show this choice of radius does not always unambiguously differentiate between core and cusped profiles. Once observational errors are below half the value of the intrinsic dispersion, improving the observational precision yields little change in the density profile uncertainties. The choice of priors in our profile shape analysis plays a crucial role when the number of stars in a system is less than 100 but does not affect the resulting uncertainties for larger kinematic samples. Our predicted 2D confidence regions agree well with those from a full likelihood analysis run on a mock kinematic data set taken from the Gaia Challenge, validating our Fisher predictions. Our methodology is flexible, allowing us to predict density profile uncertainties for a wide range of current and future kinematic data sets.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2020

SCOPE: Spectral cloud optical property estimation using real-time GOES-R longwave imagery

The output of ground-based, solar power generation systems is strongly dependent on cloud cover, which is the main contributor to solar power variability and uncertainty. Cloud optical properties are typically over-simplified in forecasting applications due to the lack of realtime, accurate estimates. In this work, we introduce a method, the Spectral Cloud Optical Property Estimation (SCOPE), for estimating cloud optical properties directly from high-resolution (5-min, 2 km) imagery from Geostationary Operational Environmental Satellite (GOES)-R, which is the newest generation of the GOES system. The SCOPE method couples a two-stream, spectrally resolved radiative model with the longwave GOES-R sensor output to simultaneously estimate the cloud optical depth, cloud top height, and cloud thickness during both day and night at 5-min intervals. The accuracy of SCOPE is evaluated using one year (2018) of downwelling longwave (DLW) radiation measurements from the Surface Radiation Budget Network, which consists of seven sites spread across climatically diverse regions of the contiguous United States. During daytime clear-sky periods, SCOPE predicts DLW within instrument uncertainty (10 W m -2 ) for four of the seven locations, with the remaining locations yielding errors of the order of 11.2, 17.7, and 20.2 W m -2 . For daytime cloudy-sky, daytime all-sky (clear or cloudy), and nighttime all-sky periods, SCOPE achieves root mean square error values of 23.0–34.5 W m -2 for all seven locations. These results, together with the low-latency of the method (1 s per sample), show that SCOPE provides a viable solution to real-time, accurate estimation of cloud optical properties for both day and night.

42 ENGINEERING↗

NASA NTRS2020

The Boreal Winter El Niño Precipitation Response over North America: Insights into Why January is More Difficult to Predict than February

This study examines the within−season monthly variation of the El Niño response over North America during December−March using the NASA/GEOS model. In agreement with previous studies, the skill of 1−month lead GEOS coupled model forecasts of precipitation over North America is largest (smallest) for February (January), with similar results in uncoupled mode. A key finding is that the relatively poor January skillis the result ofthe model placing the main circulation anomaly over the northeast Pacific slightly to the west of the observed, resulting inprecipitation anomalies that lie off the coast instead of over land as observed. In contrast, during February the observed circulation anomaly over the northeast Pacific shifts westward, lining up with the predicted anomaly which is essentially unchanged from January, resulting in both the observed and predicted precipitation anomalies remaining off the coast. Furthermore, the largest precipitation anomalies occur along the southern tier of states associated with an eastward extended jet–something that the models capture reasonably well. Simulations with a stationary wave model indicate that the placement of January El Niño response to the west of the observed over the northeast Pacificis the result of biases in the January climatological stationary waves, rather than errors in the tropical Pacific El Niño heating anomalies in January. Furthermore, evidence is provided that the relatively poor simulation of the observed January climatology, characterized by a strengthened North Pacific jet and enhanced ridge over western North America, can be traced back to biases in the January climatology heating over the Tibet region and the tropical western Pacific.

Young-kwon Lim↗

DOE OSTI2025

Uncertainty quantification and optimization of precipitating hydrometeor parameters for winter precipitation in a cloud microphysics scheme

The precipitating hydrometeor parameters used in cloud microphysics schemes carry inherent uncertainties. The quantification of these uncertainties, together with parameter optimization, can significantly improve precipitation forecasts. This study investigates the effects of 13 parameters in the Weather Research and Forecasting (WRF) Double-Moment 6-class (WDM6) microphysics scheme, which define the hydrometeor characteristics such as fall velocity–diameter and mass–diameter relationships, as well as the shape parameter of the drop size distribution for precipitating particles such as rain, snow, and graupel on simulated winter precipitation. A comparison between the model's pre-defined parameters and observations from the International Collaborative Experiments for the PyeongChang 2018 Olympic and Paralympic winter games (ICE-POP 2018) field campaign reveals that the fall velocity–diameter relationship for rain, the mass–diameter relationships for snow and graupel, and the shape parameters for all precipitating particles in the WDM6 scheme deviate from the median values observed by the two-dimensional video disdrometer (2DVD). To quantify parameter sensitivities, a perturbed parameter ensemble (PPE) of 256 simulations was conducted within parameter ranges constrained by 2DVD observations for three winter precipitation cases. Bayesian optimization was then applied to identify parameter sets that minimized the root mean square error (RMSE) for each case, achieving reductions of up to 30.2 %. These results demonstrate that ensemble-based uncertainty quantification and parameter optimization can help identify key parameters and provide a pathway to improving precipitation simulation performance. In addition, measurement sites can be strategically selected based on regions that show high sensitivity to variations in hydrometeor characteristic parameters.

Bayesian optimization↗

NASA NTRS2019

Using Wind Observations from Nearby Aircraft to Update the Optimal Descent Trajectory in Real-Time

The ability to meet a controlled time of arrival during a continuous descent operation will enable environmentally friendly and fuel efficient descent operations while simultaneously maintaining airport throughput. However, if the wind forecast used to compute the initial trajectory plan is not accurate enough, the guidance system will need to correct time deviations from the plan during the execution of the descent. Previous work proposed an on-board guidance strategy based on model predictive control, which repeatedly updates the trajectory plan in real-time from the current aircraft state and for the remainder of the descent. However, the wind conditions downstream, at altitudes not explored yet, were difficult to predict due to the lack of data. This paper shows the potential benefits of using wind observations, broadcast by nearby aircraft, to reconstruct the wind profile downstream. The wind profile in the trajectory optimization problem is modeled as a spline, which control points are updated to fit the observations before re-planning the trajectory. Results from simulations using realistic wind data show that the performance of model predictive control significantly improves when including up-to-date wind observations, in terms of time and energy errors at the metering fix and fuel consumption.

Dalmau, Ramon↗

DOE OSTI2023

Simulating organic aerosol in Delhi with WRF-Chem using the volatility-basis-set approach: exploring model uncertainty with a Gaussian process emulator

The nature and origin of organic aerosol in the atmosphere remain unclear. The gas–particle partitioning of semi-volatile organic compounds (SVOCs) that constitute primary organic aerosols (POAs) and the multigenerational chemical aging of SVOCs are particularly poorly understood. The volatility basis set (VBS) approach, implemented in air quality models such as WRF-Chem (Weather Research and Forecasting model with Chemistry), can be a useful tool to describe emissions of POA and its chemical evolution. However, the evaluation of model uncertainty and the optimal model parameterization may be expensive to probe using only WRF-Chem simulations. Gaussian process emulators, trained on simulations from relatively few WRF-Chem simulations, are capable of reproducing model results and estimating the sources of model uncertainty within a defined range of model parameters. In this study, a WRF-Chem VBS parameterization is proposed; we then generate a perturbed parameter ensemble of 111 model runs, perturbing 10 parameters of the WRF-Chem model relating to organic aerosol emissions and the VBS oxidation reactions. This allowed us to cover the model's uncertainty space and to compare outputs from each run to aerosol mass spectrometer observations of organic aerosol concentrations and O:C ratios measured in New Delhi, India. The simulations spanned the organic aerosol concentrations measured with the aerosol mass spectrometer (AMS). However, they also highlighted potential structural errors in the model that may be related to unsuitable diurnal cycles in the emissions and/or failure to adequately represent the dynamics of the planetary boundary layer. While the structural errors prevented us from clearly identifying an optimized VBS approach in WRF-Chem, we were able to apply the emulator in the following two periods: the full period (1–29 May) and a subperiod period of 14:00–16:00 h LT (local time) on 1–29 May. The combination of emulator analysis and model evaluation metrics allowed us to identify plausible parameter combinations for the analyzed periods. We demonstrate that the methodology presented in this study can be used to determine the model uncertainty and to identify the appropriate parameter combination for the VBS approach and hence to provide valuable information to improve our understanding of OA production.

54 ENVIRONMENTAL SCIENCES↗

NASA NTRS2021

Evaluation of Global Surface Water Temperature Data Sets for use in Passive Remote Sensing of Soil Moisture

Inland open water bodies often pose a systematic error source in the passive remote sensing retrievals of soil moisture. Water temperature is a necessary variable used to compute water emissions that is required to be subtracted from satellite observation to yield actual emissions from the land portion, which in turn generates accurate soil moisture retrievals. Therefore, overestimation of soil moisture can often be corrected using concurrent water temperature data in the overall mitigation procedure. In recent years, several data sets of lake water temperature have become available, but their specifications and accuracy have rarely been investigated in the context of passive soil moisture remote sensing on a global scale. For this reason, three lake temperature products were evaluated against in-situ measurements from 2007 to 2011. The data sets include the lake surface water temperature (LSWT) from Global Observatory of Lake Responses to Environmental Change (GloboLakes), the Copernicus Global Land Operations Cryosphere and Water (C-GLOPS), as well as the lake mix-layer temperature (LMLT) from the European Centers for Medium-Range Weather Forecast (ECMWF) ERA5 Land Reanalysis. GloboLakes, C-GLOPS, and ERA5 Land have overall comparable performance with Pearson correlations (R) of 0.87, 0.92 and 0.88 in comparison with in-situ measurements. LSWT products exhibit negative median biases of −0.27 K (GloboLakes) and −0.31 K (C-GLOPS), whereas the median bias of LMLT is 1.56 K. When mapped from their respective native resolutions to a common 9 km Equal-Area Scalable Earth (EASE) Grid 2.0 projection, similar relative performance was observed. LMLT and LSWT data are closer in performance over the 9 km grid cells that exhibit a small range of lake cover fractions (0.05–0.5). Despite comparable relative performance, ERA5 Land shows great advantages in spatial coverage and temporal resolution. In summary, an integrated evaluation on data accuracy, long-term availability, global coverage, temporal resolution, and regular forward processing with modest data latency led us to conclude that LMLT from the ERA5 Land Reanalysis product represents the most optimal path for use in the development of a long-term soil moisture product.

Runze Zhang↗

DOE OSTI2022

Neural Network-Based Electric Vehicle Range Prediction for Smart Charging Optimization

Range prediction is a standard feature in most modern road vehicles, allowing drivers to make informed decisions about when to refuel. Most vehicles make range predictions through data- or model-driven means, monitoring the average fuel consumption rate or using a tuned vehicle model to predict fuel consumption. The uncertainty of future driving conditions makes the range prediction problem challenging, particularly for less pervasive battery electric vehicles (BEV). Most contemporary machine learning-based methods attempt to forecast the battery SOC discharge profile to predict vehicle range. In this work, we propose a novel approach using two recurrent neural networks (RNNs) to predict the remaining range of BEVs and the minimum charge required to safely complete a trip. Each RNN has two outputs that can be used for statistical analysis to account for uncertainties; the first loss function leads to mean and variance estimation (MVE), while the second results in bounded interval estimation (BIE). These outputs of the proposed RNNs are then used to predict the probability of a vehicle completing a given trip without charging, or if charging is needed, the remaining range and minimum charging required to finish the trip with high probability. Training data was generated using a low-order physics model to estimate vehicle energy consumption from historical drive cycle data collected from medium-duty last-mile delivery vehicles. Here, the proposed method demonstrated high accuracy in the presence of day-to-day route variability, with the root-mean-square error (RMSE) below 6% for both RNN models.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

NASA NTRS2001

The Vertical Error Characteristics of GOES-derived Winds: Description and Impact on Numerical Weather Prediction

Errors in the height assignment of some satellite-derived winds exist because the satellites sense radiation emitted from a finite layer of the atmosphere rather than a specific level. Potential problems in data assimilation may arise because the motion of a measured layer is often represented by a single-level value. In this research, cloud and water vapor motion winds that are derived from the Geostationary Operational Environmental Satellites (GOES winds) are compared to collocated rawinsonde observations (RAOBs). An important aspect of this work is that in addition to comparisons at each assigned height, the GOES winds are compared to the entire profile of the collocated RAOB data to determine the vertical error characteristics of the GOES winds. The impact of these results on numerical weather prediction is then investigated. The comparisons at individual vector height assignments indicate that the error of the GOES winds range from approx. 3 to 10 m/s and generally increase with height. However, if taken as a percentage of the total wind speed, accuracy is better at upper levels. As expected, comparisons with the entire profile of the collocated RAOBs indicate that clear-air water vapor winds represent deeper layers than do either infrared or water vapor cloud-tracked winds. This is because in cloud-free regions the signal from water vapor features may result from emittance over a thicker layer. To further investigate characteristics of the clear-air water vapor winds, they are stratified into two categories that are dependent on the depth of the layer represented by the vector. It is found that if the vertical gradient of moisture is smooth and uniform from near the height assignment upwards, the clear-air water vapor wind tends to represent a relatively deep layer. The information from the comparisons is then used in numerical model simulations of two separate events to determine the forecast impacts. Four simulations are performed for each case: 1) A control simulation that assimilates no satellite wind data, 2) assimilation of all GOES winds according to their assigned single level height, 3) assimilation of all GOES winds spread over multiple levels, and 4) assimilation of all GOES winds spread over multiple levels, but with variations in the vertical influence of clear-air water vapor winds based on the moisture profile in the model. In the first case, a strong mid-latitude cyclone is present and the use of the satellite data results in improved storm tracks during the initial approx. 36 h forecast period. This is because the satellite data improves the analysis of the environment into which the storm progresses. Statistics for mean wind vector and height differences show that, with the exception of the height field at later times in the first case, the use of GOES winds improves the simulation with time. The simulation results suggest that it is beneficial to spread the GOES wind information over multiple levels, particularly when the moisture profile is used to define the vertical influence.

Rao, P. Anil↗

DOE OSTI2022

Multimodal sensor fusion framework for residential building occupancy detection

For several years now, smart building energy systems have been a research area of intensive activity. In light of the increasing need for sustainable buildings and energy systems, this trend motivates an increasing need for a solution to reduce carbon dioxide emissions and improve energy efficiency. This work proposes a high-performing and transferable occupancy detection framework that combines sensor data from different data modalities, including time series environmental data (temperature, humidity, and illuminance), image data, and acoustic energy data using ensemble method. To draw out the best prediction performance in each modality, the proposed framework was developed, including various models that were designed to learn the occupancy patterns reflected in the physical data streams. To tackle the time series environmental data, we designed two variants of an occupancy detection spatiotemporal pattern network (Occ-STPN) that performs both feature level and decision level fusion, respectively. We also propose a new metric; the fading memory mean square error (FMMSE), that provides a fair evaluation and penalization of delayed occupancy predictions. Multiple open-sourced datasets, including the Electricity Consumption and Occupancy and the University of California, Irvine's (UCI) building occupancy detection dataset, along with our own real data collected from six different houses, were used to validate the algorithms' performance. The experimental results presented herein break down the performance for each sensing modality, and a detailed analysis of the performance is also discussed.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

NASA NTRS2018

The Effects of Uncertainty in Initial CME Input Parameters on Deflection, Rotation, Bz, and Arrival Time Predictions

Understanding the effects of coronal mass ejections (CMEs) requires knowing if and when they will impact and their properties upon impact. Of particular importance is the strength of a CME's southward magnetic field component (Bz ). Kay et al. (2013, https://doi:10.1088/0004-637X/775/1/5, 2015, https://doi:10.1088/948 0004-637X/805/2/168) have shown that the simplified analytic model Forecasting a CME's Altered Trajectory (ForeCAT) can reproduce the deflection and rotation of CMEs. Kay, Gopalswamy, Reinard, and Opher (2017, https://doi.org/10.3847/1538-4357/835/2/117) introduced ForeCAT In situ Data Observer, which uses ForeCAT results to simulate magnetic field profiles. ForeCAT In situ Data Observer reproduces the in situ observations on roughly hourly time scales, suggesting that these models could be extremely useful for predictions of Bz . However, as with all models, both models are sensitive to their input parameters, which may not be precisely known for predictions. We explore this sensitivity using ensembles having small changes in the initial latitude, longitude, and orientation of the erupting CME. We explore the effects of different background magnetic field models and find that the changes in deflection and rotation resulting from the uncertainty in the initial parameters tend to exceed the changes from dierent magnetic backgrounds. The range in the in situ proles tends to scale with the range in the deflection and rotation. We also consider a simple arrival time model using ForeCAT results and nd an average absolute error of only 3 hr. We show that an uncertainty in the CME position of 8.1 ± 6.9 leads to variations of 6 hr in the arrival time. This measure depends strongly on the location of impact within the CME with the arrival time changing less for impacts near the nose.

Kay, Christina D.↗

DOE OSTI2021

Monitoring agroecosystem productivity and phenology at a national scale: A metric assessment framework

Effective measurement of seasonal variations in the timing and amount of production is critical to managing spatially heterogeneous agroecosystems in a changing climate. Although numerous technologies for such measurements are available, their relationships to one another at a continental extent are unknown. Using data collected from across the Long-Term Agroecosystem Research (LTAR) network and other networks, we investigated correlations among key metrics representing primary production, phenology, and carbon fluxes in croplands, grazing lands, and crop-grazing integrated systems across the continental U.S. Metrics we examined included gross primary productivity (GPP) estimated from eddy covariance (EC) towers and modelled from the Landsat satellite, Landsat NDVI, and vegetation greenness (Green Chromatic Coordinate, GCC) from tower-mounted PhenoCams for 2017 and 2018. Overall, our analysis compared production dynamics estimated from three independent ground and remote platforms using data for 34 agricultural sites constituting 51 site-years of co-located time series. Pairwise sensor comparisons across all four metrics revealed stronger correlation and lower root mean square error (RMSE) between end of season (EOS) dates (Pearson R ranged from 0.6 to 0.7 and RMSE from 32.5 to 67.8) than start of season (SOS) dates (0.46 to 0.69 and 40.4 to 66.2). Overall, moderate to high correlations between SOS and EOS metrics complemented one another except at some lower productivity grazing land sites where estimating SOS can be challenging. Growing season length estimates derived from 16-day satellite GPP (179.1 days) were significantly longer than those from PhenoCam G CC (70.4 days, p adj < 0.0001) and EC GPP (79.6 days, p adj < 0.0001). Landscape heterogeneity did not explain differences in SOS and EOS estimates. Annual integrated estimates of productivity from EC GPP and PhenoCam G CC diverged from those estimated by Landsat GPP and NDVI at sites where annual production exceeds 1000 gC/m –2 yr –1 . Based on our results, we developed a “metric assessment framework” that articulates where and how metrics from satellite, eddy covariance and PhenoCams complement, diverge from, or are redundant with one another. The framework was designed to optimize instrumentation selection for monitoring, modeling, and forecasting ecosystem functioning with the ultimate goal of informing decision-making by land managers, policy-makers, and industry leaders working at multiple scales.

54 ENVIRONMENTAL SCIENCES↗