Search for: All records

Abstract Vertical profiles of temperature and dewpoint are useful in predicting deep convection that leads to severe weather, which threatens property and lives. Currently, forecasters rely on observations from radiosonde launches and numerical weather prediction (NWP) models. Radiosonde observations are, however, temporally and spatially sparse, and NWP models contain inherent errors that influence short-term predictions of high impact events. This work explores using machine learning (ML) to postprocess NWP model forecasts, combining them with satellite data to improve vertical profiles of temperature and dewpoint. We focus on different ML architectures, loss functions, and input features to optimize predictions. Because we are predicting vertical profiles at 256 levels in the atmosphere, this work provides a unique perspective at using ML for 1D tasks. Compared to baseline profiles from the Rapid Refresh (RAP), ML predictions offer the largest improvement for dewpoint, particularly in the middle and upper atmosphere. Temperature improvements are modest, but CAPE values are improved by up to 40%. Feature importance analyses indicate that the ML models are primarily improving incoming RAP biases. While additional model and satellite data offer some improvement to the predictions, architecture choice is more important than feature selection in fine-tuning the results. Our proposed deep residual U-Net performs the best by leveraging spatial context from the input RAP profiles; however, the results are remarkably robust across model architecture. Further, uncertainty estimates for every level are well calibrated and can provide useful information to forecasters. 

Abstract The detection of multilayer clouds in the atmosphere can be particularly challenging from passive visible and infrared imaging radiometers since cloud boundary information is limited primarily to the topmost cloud layer. Yet detection of low clouds in the atmosphere is important for a number of applications, including aviation nowcasting and general weather forecasting. In this work, we develop pixel-based machine learning–based methods of detecting low clouds, with a focus on improving detection in multilayer cloud situations and specific attention given to improving the Cloud Cover Layers (CCL) product, which assigns cloudiness in a scene into vertical bins. The random forest (RF) and neural network (NN) implementations use inputs from a variety of sources, including GOES Advanced Baseline Imager (ABI) visible radiances, infrared brightness temperatures, auxiliary information about the underlying surface, and relative humidity (which holds some utility as a cloud proxy). Training and independent validation enlists near-global, actively sensed cloud boundaries from the radar and lidar systems on board theCloudSatandCALIPSOsatellites. We find that the RF and NN models have similar performances. The probability of detection (PoD) of low cloud increases from 0.685 to 0.815 when using the RF technique instead of the CCL methodology, while the false alarm ratio decreases. The improved PoD of low cloud is particularly notable for scenes that appear to be cirrus from an ABI perspective, increasing from 0.183 to 0.686. Various extensions of the model are discussed, including a nighttime-only algorithm and expansion to other satellite sensors. Significance StatementUsing satellites to detect the heights of clouds in the atmosphere is important for a variety of weather applications, including aviation weather forecasting. However, detecting low clouds can be challenging if there are other clouds above them. To address this, we have developed machine learning–based models that can be used with passive satellite instruments. These models use satellite observations at visible and infrared wavelengths, an estimate of relative humidity in the atmosphere, and geographic and surface-type information to predict whether low clouds are present. Our results show that these models have significant skill at predicting low clouds, even in the presence of higher cloud layers. 

Abstract Neural networks (NN) have become an important tool for prediction tasks—both regression and classification—in environmental science. Since many environmental-science problems involve life-or-death decisions and policy making, it is crucial to provide not only predictions but also an estimate of the uncertainty in the predictions. Until recently, very few tools were available to provide uncertainty quantification (UQ) for NN predictions. However, in recent years the computer-science field has developed numerous UQ approaches, and several research groups are exploring how to apply these approaches in environmental science. We provide an accessible introduction to six of these UQ approaches, then focus on tools for the next step, namely, to answer the question:Once we obtain an uncertainty estimate (using any approach), how do we know whether it is good or bad?To answer this question, we highlight four evaluation graphics and eight evaluation scores that are well suited for evaluating and comparing uncertainty estimates (NN based or otherwise) for environmental-science applications. We demonstrate the UQ approaches and UQ-evaluation methods for two real-world problems: 1) estimating vertical profiles of atmospheric dewpoint (a regression task) and 2) predicting convection over Taiwan based onHimawari-8satellite imagery (a classification task). We also provide Jupyter notebooks with Python code for implementing the UQ approaches and UQ-evaluation methods discussed herein. This article provides the environmental-science community with the knowledge and tools to start incorporating the large number of emerging UQ methods into their research. Significance StatementNeural networks are used for many environmental-science applications, some involving life-or-death decision-making. In recent years new methods have been developed to provide much-needed uncertainty estimates for NN predictions. We seek to accelerate the adoption of these methods in the environmental-science community with an accessible introduction to 1) methods for computing uncertainty estimates in NN predictions and 2) methods for evaluating such estimates. 

Abstract. The uptake of carbonyl sulfide (COS) by terrestrial plants is linked tophotosynthetic uptake of CO2 as these gases partly share the sameuptake pathway. Applying COS as a photosynthesis tracer in models requires anaccurate representation of biosphere COS fluxes, but these models have notbeen extensively evaluated against field observations of COS fluxes. In thispaper, the COS flux as simulated by the Simple Biosphere Model, version 4(SiB4), is updated with the latest mechanistic insights and evaluated with siteobservations from different biomes: one evergreen needleleaf forest, twodeciduous broadleaf forests, three grasslands, and two crop fields spread overEurope and North America. We improved SiB4 in several ways to improve itsrepresentation of COS. To account for the effect of atmospheric COS molefractions on COS biosphere uptake, we replaced the fixed atmospheric COS molefraction boundary condition originally used in SiB4 with spatially andtemporally varying COS mole fraction fields. Seasonal amplitudes of COS molefractions are ∼50–200 ppt at the investigated sites with aminimum mole fraction in the late growing season. Incorporating seasonalvariability into the model reduces COS uptake rates in the late growingseason, allowing better agreement with observations. We also replaced theempirical soil COS uptake model in SiB4 with a mechanistic model thatrepresents both uptake and production of COS in soils, which improves thematch with observations over agricultural fields and fertilized grasslandsoils. The improved version of SiB4 was capable of simulating the diurnal andseasonal variation in COS fluxes in the boreal, temperate, and Mediterraneanregion. Nonetheless, the daytime vegetation COS flux is underestimated onaverage by 8±27 %, albeit with large variability across sites. On aglobal scale, our model modifications decreased the modeled COS terrestrialbiosphere sink from 922 Gg S yr−1 in the original SiB4 to753 Gg S yr−1 in the updated version. The largest decrease influxes was driven by lower atmospheric COS mole fractions over regions withhigh productivity, which highlights the importance of accounting forvariations in atmospheric COS mole fractions. The change to a different soilmodel, on the other hand, had a relatively small effect on the globalbiosphere COS sink. The secondary role of the modeled soil component in theglobal COS budget supports the use of COS as a global photosynthesis tracer. Amore accurate representation of COS uptake in SiB4 should allow for improvedapplication of atmospheric COS as a tracer of local- to global-scaleterrestrial photosynthesis.