Search for: All records

Abstract This study employs an explainable machine learning (ML) framework (XGBoost‐SHapley Additive exPlanations analysis) to investigate controlling factors on cloud liquid water path (LWP) using EPCAPE observations near the California coast. Aerosols are found to be the dominant factor explaining LWP variability, surpassing meteorological factors (MFs). By isolating aerosol effects from meteorological influences, the ML reveals a negative linear relationship between LWP and cloud droplet number concentration (N_d) in log space, likely driven by entrainment drying via evaporation‐entrainment feedback. This aligns with the negative regime of the inverted‐V relationship reported in previous studies, while no positive LWP responses are found due to a limited number of precipitating cases in EPCAPE. Furthermore, the sensitivity of LWP toN_dshows a non‐linear dependence on MFs like moisture contrast between surface and free troposphere and lower‐tropospheric stability. This occurs due to the interplay between the MFs' direct effects on entrainment drying and indirect effects through LWP adjustments. 

Abstract. Remote sensing measurements have been widely used to estimate the planetary boundary layer height (PBLHT). Each remote sensing approach offers unique strengths and faces different limitations. In this study, we use machine learning (ML) methods to produce a best-estimate PBLHT (PBLHT-BE-ML) by integrating four PBLHT estimates derived from remote sensing measurements at the Department of Energy (DOE) Atmospheric Radiation Measurement (ARM) Southern Great Plains (SGP) observatory. Three ML models – random forest (RF) classifier, RF regressor, and light gradient-boosting machine (LightGBM) – were trained on a dataset from 2017 to 2023 that included radiosonde, various remote sensing PBLHT estimates, and atmospheric meteorological conditions. Evaluations indicated that PBLHT-BE-ML from all three models improved alignment with the PBLHT derived from radiosonde data (PBLHT-SONDE), with LightGBM demonstrating the highest accuracy under both stable and unstable boundary layer conditions. Feature analysis revealed that the most influential input features at the SGP site were the PBLHT estimates derived from (a) potential temperature profiles retrieved using Raman lidar (RL) and atmospheric emitted radiance interferometer (AERI) measurements (PBLHT-THERMO), (b) vertical velocity variance profiles from Doppler lidar (PBLHT-DL), and (c) aerosol backscatter profiles from micropulse lidar (PBLHT-MPL). The trained models were then used to predict PBLHT-BE-ML at a temporal resolution of 10 min, effectively capturing the diurnal evolution of PBLHT and its significant seasonal variations, with the largest diurnal variation observed over summer at the SGP site. We applied these trained models to data from the ARM Eastern Pacific Cloud Aerosol Precipitation Experiment (EPCAPE) field campaign (EPC), where the PBLHT-BE-ML, particularly with the LightGBM model, demonstrated improved accuracy against PBLHT-SONDE. Analyses of model performance at both the SGP and EPC sites suggest that expanding the training dataset to include various surface types, such as ocean and ice-covered areas, could further enhance ML model performance for PBLHT estimation across varied geographic regions. 

Abstract Knowledge of the planetary boundary layer height (PBLH) is crucial for various applications in atmospheric and environmental sciences. Lidar measurements are frequently used to monitor the evolution of the PBLH, providing more frequent observations than traditional radiosonde‐based methods. However, lidar‐derived PBLH estimates have substantial uncertainties, contingent upon the retrieval algorithm used. In addressing this, we applied the Different Thermo‐Dynamic Stabilities (DTDS) algorithm to establish a PBLH data set at five separate Department of Energy's Atmospheric Radiation Measurement sites across the globe. Both the PBLH methodology and the products are subject to rigorous assessments in terms of their uncertainties and constraints, juxtaposing them with other products. The DTDS‐derived product consistently aligns with radiosonde PBLH estimates, with correlation coefficients exceeding 0.77 across all sites. This study delves into a detailed examination of the strengths and limitations of PBLH data sets with respect to both radiosonde‐derived and other lidar‐based estimates of the PBLH by exploring their respective errors and uncertainties. It is found that varying techniques and definitions can lead to diverse PBLH retrievals due to the inherent intricacy and variability of the boundary layer. Our DTDS‐derived PBLH data set outperforms existing products derived from ceilometer data, offering a more precise representation of the PBLH. This extensive data set paves the way for advanced studies and an improved understanding of boundary‐layer dynamics, with valuable applications in weather forecasting, climate modeling, and environmental studies. 

The Planetary Boundary Layer Height (PBLH) significantly impacts weather, climate, and air quality. Understanding the global diurnal variation of the PBLH is particularly challenging due to the necessity of extensive observations and suitable retrieval algorithms that can adapt to diverse thermodynamic and dynamic conditions. This study utilized data from the Cloud-Aerosol Transport System (CATS) to analyze the diurnal variation of PBLH in both continental and marine regions. By leveraging CATS data and a modified version of the Different Thermo-Dynamics Stability (DTDS) algorithm, along with machine learning denoising, the study determined the diurnal variation of the PBLH in continental mid-latitude and marine regions. The CATS DTDS-PBLH closely matches ground-based lidar and radiosonde measurements at the continental sites, with correlation coefficients above 0.6 and well-aligned diurnal variability, although slightly overestimated at nighttime. In contrast, PBLH at the marine site was consistently overestimated due to the viewing geometry of CATS and complex cloud structures. The study emphasizes the importance of integrating meteorological data with lidar signals for accurate and robust PBLH estimations, which are essential for effective boundary layer assessment from satellite observations. 

Aerosol-cloud interactions (ACIs) are vital for regulating Earth’s climate by influencing energy and water cycles. Yet, effects of ACI bear large uncertainties, evidenced by systematic discrepancies between observed and modeled estimates. This study quantifies a major bias in ACI determinations, stemming from conventional surface or space measurements that fail to capture aerosol at the cloud level unless the cloud is coupled with land surface. We introduce an advanced approach to determine radiative forcing of ACI by accounting for cloud-surface coupling. By integrating field observations, satellite data, and model simulations, this approach reveals a drastic alteration in aerosol vertical transport and ACI effects caused by cloud coupling. In coupled regimes, aerosols enhance cloud droplet number concentration across the boundary layer more homogeneously than in decoupled conditions, under which aerosols from the free atmosphere predominantly affect cloud properties, leading to marked cooling effects. Our findings spotlight cloud-surface coupling as a key factor for ACI quantification, hinting at potential underassessments in traditional estimates. 

Abstract Aerosols are important environmental factors that can influence deep convective clouds (DCCs) by serving as cloud condensation nuclei. Due to complications in DCC dynamics and microphysics, and aerosol size distribution and composition, understanding aerosol‐DCC interactions has been a daunting challenge. Recently, the convective invigoration mechanisms through enhancing latent heating in condensation and ice‐related processes that have been proposed in literature are debated for their significance qualitatively and quantitatively. A salient issue arising from these debates is the imperative need to clarify essential knowledge and methodologies in investigating aerosol impacts on deep convection. Here we have presented our view of key aspects on investigating and understanding these invigoration mechanisms as well as the aerosol and meteorological conditions under which these mechanisms may be significant based on new findings. For example, the condensational invigoration is most significant under a clean condition with an introduction of a large number of ultrafine particles, and the freezing‐induced invigoration can be significant in a clean condition with a large number of relatively large‐size particles being added. We have made practical recommendations on approaches for investigating aerosol impacts on convection with both modeling and observations. We note that the feedback induced by the invigoration via the enhanced latent heating to circulation and meteorology can be an important part of aerosol impacts but is very complicated and varies with different convective storm types. This is an important future direction for studying aerosol‐DCC interactions. 

Key Points A preexisting stratocumulus deck is more persistent when experiencing warm‐air advection than cold‐air advection This persistence is due to reduced entrainment drying as a result of decoupling, which outweighs decreased cloud‐base moisture transport The mechanism is more notable when free‐tropospheric humidity is higher 

Abstract. Lightning is affected by many factors, many of which are not routinely measured, well understood, or accounted for in physical models. Machine learning (ML) excels in exploring and revealing complex relationships between meteorological variables such as those measured at the South Great Plains (SGP) Atmospheric Radiation Measurement (ARM) site; a site that provides an unprecedented level of detail on atmospheric conditions and clouds. Several commonly used ML models have been applied to analyse the relationship between ARM data and lightning data from the Earth Networks Total Lightning Network (ENTLN) in order to identify important variables affecting lightning occurrence in the vicinity of the SGP site during the summers (June, July, August and September) of 2012 to 2020. Testing various ML models, we found that the Random Forest model is the best predictor among common classifiers. It predicted lightning occurrence with an accuracy of 76.9 % and an area under curve (AUC) of 0.850. Using this model, we further ranked the variables in terms of their effectiveness in predicting lightning and identified geometric cloud thickness, rain rate and convective available potential energy (CAPE) as the most effective predictors. The contrast in meteorological variables between no-lightning and frequent-lightning periods was examined on hours with CAPE values conducive to thunderstorm formation. Besides the variables considered for the ML models, surface variables such as equivalent potential temperature and mid-altitude variables such as minimum equivalent potential temperature have a large contrast between no-lightning and frequent-lightning hours. Finally, a notable positive relationship between intra-cloud (IC) flash fraction and the square root of CAPE was found suggesting that stronger updrafts increase the height of the electrification zone, resulting in fewer flashes reaching the surface and consequently a greater IC flash fraction. 

Abstract. Lightning is affected by many factors, many of which are not routinely measured, well understood, or accounted for in physical models. Several commonly used machine learning (ML) models have been applied to analyze the relationship between Atmospheric Radiation Measurement (ARM) data and lightning data from the Earth Networks Total Lightning Network (ENTLN) in order to identify important variables affecting lightning occurrence in the vicinity of the Southern Great Plains (SGP) ARM site during the summer months (June, July, August and September) of 2012 to 2020. Testing various ML models, we found that the random forest model is the best predictor among common classifiers. When convective clouds were detected, it predicts lightning occurrence with an accuracy of 76.9 % and an area under the curve (AUC) of 0.850. Using this model, we further ranked the variables in terms of their effectiveness in nowcasting lightning and identified geometric cloud thickness, rain rate and convective available potential energy (CAPE) as the most effective predictors. The contrast in meteorological variables between no-lightning and frequent-lightning periods was examined for hours with CAPE values conducive to thunderstorm formation. Besides the variables considered for the ML models, surface variables and mid-altitude variables (e.g., equivalent potential temperature and minimum equivalent potential temperature, respectively) have statistically significant contrasts between no-lightning and frequent-lightning hours. For example, the minimum equivalent potential temperature from 700 to 500 hPa is significantly lower during frequent-lightning hours compared with no-lightning hours. Finally, a notable positive relationship between the intracloud (IC) flash fraction and the square root of CAPE (CAPE) was found, suggesting that stronger updrafts increase the height of the electrification zone, resulting in fewer flashes reaching the surface and consequently a greater IC flash fraction.