U.S. Department of EnergyOffice of Scientific and Technical Information
Deep-Learning-derived Boundary Layer Height from Meteorological Data over the SGP, GOAMAZON, CACTI
Abstract
The planetary boundary-layer (PBL) height (PBLH) is an important parameter for various meteorological and climate studies. This study presents a multi-structure deep neural network (DNN) model, designed to estimate PBLH by integrating morning temperature profiles with surface meteorological observations. The DNN model is developed by leveraging a rich data set of PBLH derived from long-standing radiosonde records and augmented with high-resolution micropulse lidar and Doppler lidar observations. We access the performance of the DNN with an ensemble of 10 members, each featuring distinct hidden layer structures, which collectively yield a robust 27-year PBLH data set over the Southern Great Plains from 1994 to 2020. The influence of various meteorological factors on PBLH is rigorously analyzed through the importance test. Moreover, the DNN model's accuracy is evaluated against radiosonde observations and juxtaposed with conventional remote-sensing methodologies, including Doppler lidar, ceilometer, Raman lidar, and micropulse lidar. The DNN model exhibits reliable performance across diverse conditions and demonstrates lower biases relative to remote-sensing methods. In addition, the DNN model, originally trained over a plain region, demonstrates remarkable adaptability when applied to the heterogeneous terrains and climates encountered during the GoAmazon (tropical rainforest) and CACTI (middle-latitude mountain) campaigns. These findings demonstrate the effectiveness ofmore »
- Publication Date:
- Other Number(s):
- ARM0832
- DOE Contract Number:
- AC05-00OR22725
- Research Org.:
- Atmospheric Radiation Measurement (ARM) Archive, Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (US); ARM Data Center, Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
- Sponsoring Org.:
- USDOE Office of Science (SC), Biological and Environmental Research (BER)
- Collaborations:
- PNNL, BNL, ANL, ORNL
- Subject:
- 54 ENVIRONMENTAL SCIENCES; pblh
- OSTI Identifier:
- 2344988
- DOI:
- https://doi.org/10.5439/2344988
Citation Formats
SU, Tianning. Deep-Learning-derived Boundary Layer Height from Meteorological Data over the SGP, GOAMAZON, CACTI. United States: N. p., 2024.
Web. doi:10.5439/2344988.
SU, Tianning. Deep-Learning-derived Boundary Layer Height from Meteorological Data over the SGP, GOAMAZON, CACTI. United States. doi:https://doi.org/10.5439/2344988
SU, Tianning. 2024.
"Deep-Learning-derived Boundary Layer Height from Meteorological Data over the SGP, GOAMAZON, CACTI". United States. doi:https://doi.org/10.5439/2344988. https://www.osti.gov/servlets/purl/2344988. Pub date:Thu May 02 00:00:00 EDT 2024
@article{osti_2344988,
title = {Deep-Learning-derived Boundary Layer Height from Meteorological Data over the SGP, GOAMAZON, CACTI},
author = {SU, Tianning},
abstractNote = {The planetary boundary-layer (PBL) height (PBLH) is an important parameter for various meteorological and climate studies. This study presents a multi-structure deep neural network (DNN) model, designed to estimate PBLH by integrating morning temperature profiles with surface meteorological observations. The DNN model is developed by leveraging a rich data set of PBLH derived from long-standing radiosonde records and augmented with high-resolution micropulse lidar and Doppler lidar observations. We access the performance of the DNN with an ensemble of 10 members, each featuring distinct hidden layer structures, which collectively yield a robust 27-year PBLH data set over the Southern Great Plains from 1994 to 2020. The influence of various meteorological factors on PBLH is rigorously analyzed through the importance test. Moreover, the DNN model's accuracy is evaluated against radiosonde observations and juxtaposed with conventional remote-sensing methodologies, including Doppler lidar, ceilometer, Raman lidar, and micropulse lidar. The DNN model exhibits reliable performance across diverse conditions and demonstrates lower biases relative to remote-sensing methods. In addition, the DNN model, originally trained over a plain region, demonstrates remarkable adaptability when applied to the heterogeneous terrains and climates encountered during the GoAmazon (tropical rainforest) and CACTI (middle-latitude mountain) campaigns. These findings demonstrate the effectiveness of deep learning models in estimating PBLH, enhancing our understanding of boundary-layer dynamics with implications for enhancing the representation of PBL in weather forecasting and climate modeling.},
doi = {10.5439/2344988},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Thu May 02 00:00:00 EDT 2024},
month = {Thu May 02 00:00:00 EDT 2024}
}