Universidade Federal de Santa Maria
Ci. e Nat., Santa Maria v.42, Special Edition: Micrometeorologia, e38, 2020
DOI:10.5902/2179460X53221
ISSN 2179-460X
Received: 17/08/20 Accepted: 17/08/20 Published: 28/08/20
Special Edition
Analysis of the simulation of the PALM model for the Convective Boundary Layer in the Amazon (GOAMAZON 2014/5)
Análise da simulação do modelo PALM para a Camada Limite Convectiva na Amazônia (GOAMAZON 2014/5)
Rayonil Gomes Carneiro I
Camilla Kassar Borges II
Alice Henkes III
Gilberto Fisch IV
I Instituto Nacional de Pesquisas Espaciais, São José dos Campos, Brazil. E-mail: rayonilcarneiro@gmail.com.
II Universidade Federal de Campina Grande, Campina Grande, Brazil. E-mail: camillakassar@gmail.com.
III Instituto Nacional de Pesquisas Espaciais, São José dos Campos, Brazil. E-mail: alicehenkes@gmail.com.
IV Universidade de Taubaté, Taubaté, Brazil. E-mail: gilberto.fisch@unitau.br.
ABSTRACT
The present work had the objective to evaluate the development of the convective boundary layer in the Amazon region simulated by a high resolution Large Eddy Simulation model (named PALM model), for days representative for rainy and dry seasons. The study used data from the GOAmazon Project 2014/2015 (Green Ocean Amazon). Using data from radiosondes and Ceilometer as truth values, they were compared with the simulations performed through the PALM model. The results showed that, in general, the convective boundary layer cycle for the Amazon region was well represented by PALM model. It´s outputs has showed an overestimation of ≈ 35 m in a rainy day and an underestimation of ≈ 20 m in a dry day, both in development phase of the convective layer at late morning. It was also observed that the latent heat flux profile was higher than the sensible heat in the atmosphere, because it is a region with a lot of humidity, with the boundary layer responding rapidly to the maximum surface forcing.
Keywords: Planetary boundary layer; Turbulent flows; Large eddy simulation.
RESUMO
O presente trabalho teve como objetivo avaliar o desenvolvimento da camada limite convectiva na região da Amazônia simulada pelo modelo de alta resolução de Large Eddy Simulation, denominado de PALM, para dias característicos das estações chuvosa e seca da região. O estudo utilizou dados proveniente do Projeto GOAmazon 2014/15 (Green Ocean Amazon), sendo que medidas obtidas por radiossondas e ceilometer foram utilizadas para comparar com as simulações realizadas. Os resultados demonstraram que, de forma geral, o ciclo da camada limite convectiva para a região Amazônica foi bem representado pelo PALM. Este apresentou superestimativa de ≈ 35 m no dia chuvoso e subestimativa de ≈ 20 m no dia seco, ambas na fase de desenvolvimento da camada convectiva ao final da manhã. Foi observado que o perfil de fluxo de calor latente era maior que o calor sensível na atmosfera, por se tratar de uma região com muita umidade, com a camada limite respondendo rapidamente ao máximo forçamento da superfície.
Palavras-chave: Camada limite planetária; Fluxos turbulentos; Modelo LES.
1 INTRODUCTION
The Planetary Boundary Layer (PBL) is the layer adjacent to the surface, where turbulent heat and momentum transport occurs throughout the order of an hour and with vertical resolution about 1 km, it has atmospheric processes due to the thermal and mechanical convection during daytime and suppressed turbulent conditions that persist and evolve during the nighttime. The daily cycle of PBL varies according to the heating of the Earth's surface according to the incident solar radiation, which consists usually of a convective phase during the day (Convective Boundary Layer - CBL) and a stable phase at night (Nocturnal Boundary Layer - NBL). Therefore, studies of atmospheric processes in the lower troposphere have an important impact on society and the environment, since they occur within the PBL which is where most of the people live.
One of the fundamental parameters in PBL studies is the determination of its height, since its depth is a measure of the intensity of the surface-atmosphere interaction, in addition to being an important parameter for atmospheric modeling (NEVES; FISCH, 2015). However, the logistics for PBL height observations are often complicated. Consequently, methods to estimate its height from numerical simulations are important, and with the advancement of computational resources it has allowed improvements in modeling to solve the different scales of turbulence.
Therefore, this work has the objective to evaluate the applicability of a large scale numerical vortex model, called PALM (PArallelized Large-Eddy Simulation Model), developed by the University of Hannover in Germany and described in Raasch and Schröter (2001) and Maronga et al. (2020), in CBL simulations comparing with observational data (radiosonde and ceilometer), in an one day representative of the rainy and dry seasons in the Central Amazon region.
2 Material and Methods
2.1 Study area
The experimental site called T3 (3º 12’6.0” S, 60º 36’0.0” W) was located in the city of Manacapuru, in Amazonas state (Figure 1), and was part of the Observations and Modeling of the Green Ocean Amazon (GoAmazon 2014/5) project carried out in 2014 and 2015. This site consists of an area of small vegetation (pasture) surrounded by native forest (MARTIN et al., 2016) and is located southwest of the city of Manaus (AM).
Figure 1 – Location of the T3 experimental site in the municipality of Manacapuru-AM, Brazil
Source: Adapted by by Martin et al. (2016).
2.2 Micrometeorological data
During the GOAmazon´s campaign, many instruments were used to observe the atmospheric. In this study, the data obtained from radiosonde (RS) launched at 02, 08, 11, 14 and 20 Local Time (LT) were used and the height of PBL were computed from profiles of the wind components u and v, potential temperature (θ) and specific humidity (q). Measurements from the Ceilometer sensor were also used, which provides direct measurements of the height of the PBL, with a temporal resolution of 16 s (more details in Shukla et al. (2014), Carneiro and Fisch (2020)).
2.3 PALM model
The PALM model used in this work was implemented by Raasch and Schröter (2001) and updated by Maronga et al. (2020), being a version of the LES (Large Eddy Simulation) model in which the calculations are parallelized, and based on the Navier-Stokes equations, assuming non-hydrostatic hypotheses, incompressible fluid, and the Boussinesq's approach. The PALM model is a very useful tool in several geophysical applications including studies of the convective and night boundary layer, pollutant dispersion modeling, among others (LETZEL; RAASCH, 2003; KANDA et al., 2004; RAASCH; FRANKE, 2011; MARONGA; RAASCH, 2013; MARONGA et al., 2015; NEVES; FISCH; RAASCH, 2018; CARNEIRO et al. 2020).
The simulations were carried out using a domain of 10 km x 10 km horizontally and approximately 5 km in the vertical, with a grid spacing of 50 m, both in horizontal and vertical scales. The model simulated 12 hours (comprising the period from 8 am to 7 pm), with a spin-up time of 1 hour, which is the time it takes to create the first turbulent vortexes and achieve statistical equilibrium given the initial conditions. The simulations were carried out with the initialization through the profiles of potential temperature (θ), specific humidity (q) and the wind components (u and v) made by the 8 LT radiosonde, for a typical day of the rainy season (March 21, 2014) and dry season (September 26, 2014).
3 Results and Discussion
Figure 2A shows the time evolution of the
vertical profile of the sensible heat fluxes obtained by
simulating the PALM model for the rainy season. It was noticed positive values
above 10 W m−2 from 10 LT onwards, starting the CBL growth. The
maximum sensible heat flux occurred between 12 and 15 LT with values slightly
above 100 W m−2 and reaching a depth of 600 m. This height was
estimated as a function of the level at which the sensitive heat flow changes
from positive to negative values. The presence of a negative
layer (downward
heat transport) was verified, representing the entrainment flux above the CBL,
around -50 W m−2. From the 17 LT onward it was found that the (
) becomes
negative, inducing the formation of a stable layer.
Figure 2 – The time evolution of the profiles
A) sensible heat flux (W m−2) and B) latent
heat flux (w’q’) (W m−2) obtained by
PALM for March 21, 2014 (rainy season)
Figure 2B shows the time evolution of the
vertical profile of the latent heat flux , in which it
showed high positive values (above 200 W m−2) starting at 09 LT, the
maximum flux were detected after 12 LT, ≈ 400 W m−2. As it is a day
related to the rainy season, when there is great water/moisture availability
for the evapotranspiration processes, the fluxes
was intense
until the end of the day (16 LT).
For the characteristic day of the dry season,
it was verified that the (Figure 3A)
showed positive surface values above 10 W m−2 at the beginning of
the simulated period (09 LT). At 11 HL the
showed a higher
emission surface with a maximum intensity greater than 150 W m−2 between 14 and 15 LT, presenting a depth of aproximately 1200 m. The entrainment
flux was approximately -45 W m−2 and it is very clearly since 12 LT
until 17 LT According to Neves et al. (2018), the physical reason for the maximum
values to occur at this time (14 LT) is due to the presence of more intense
vertical movements during in the afternoon of the CBL, carrying more heat from
the surface to inside the layer, together with the negative fluxes of the
convective penetration movements of the free atmosphere into the layer.
Figure 3 – The time evolution of the profiles
A) sensible heat flux (W m−2) and B) latent
heat flux
(W m−2) obtained by PALM for September 26, 2014
(dry season)
The for the dry
season (Figure 3B) began to show positive values after 10 LT in which the
estimated maximum greater than 300 W m−2, between 14 and 16 LT,
extending over depth is around 1200 m. Near sunset, (17-18 LT), the fluxes were
very reduced at 19 HL.
The temporal evolution of the PBL depth for the two simulated days obtained through the PALM simulations was compared with the heights measured through the ceilometer and the radiosonde. On the representative day of the rainy season (Figure 4A) it was possible to verify that the NBL (between 00 and 06 LT) showed oscillate in its depth, from 150 m (between 00 and 02 LT) to 450 m (05 LT). Whereas, for the day of the dry season (Figure 4B) the NBL showed a slight increase in its heights from 150 m (00 LT) to 250 m (between 03 to 06 LT). This information served as input data for the numerical simulations.
Figure 4 – Height of the Planetary Boundary Layer (m) for days A) March 21, 2014 (rainy season) and B) September 26, 2014 (dry season)
The first simulated time steps for the rainy season showed an average growth rate of ≈75.5 m h−1 (between 08 and 09 LT) and 150 m h−1 between 09 and 10 LT, similar values to the observations from the ceilometer data, of 72 m h−1 (between 08 and 09 LT) and 125 m h−1 (between 09 and 10 LT). The observed values are in accordance with the results obtained by Carneiro et al. (2016). They described a low growth rate of CBL in the early hours of the morning, as it is the end of the erosion of the NBL, which was estimated to occur at 9 LT. It was also noticed that the PALM presented a small overestimation of the depth of the CBL increase (≈ 35 m) between 09 and 10 LT. However, in the dry season the average hourly growth rate calculated for the ceilometer and PALM simulation was ≈ 200 m h−1 (08 and 10 LT), with an underestimate of the depth of the CBL in formation, at about 20 m (10 and 11 LT). In general, during the phase of the rapid development of CBL, the PALM outputs presented a satisfactory performance in comparison with the observations in situ, a result also found by KAUFMANN and FISCH (2016) and CARNEIRO et al. (2020).
From the moment the CBL is well established in the rainy season (between 11 and 16 LT), it was found that the simulations had a maximum depth of 1,275 m at 14 LT. The results found were satisfactory in comparison to the estimate made by the RS of 14 LT (1,300 m) and by the ceilometer measurements (1,240 m). The results of PALM simulations in the dry season were also in line with observational data, with a maximum of 1,525 m at 14 LT, while the height estimate by RS had a maximum of 1,590 m and the ceilometer equal to 1,530 m. These satisfactory results from PALM agree with those obtained by MARONGA and RAASCH (2013), HEINZE et al. (2017), even though they are carried out under different conditions than the Amazon.
After 16 HL a period in which the incidence of global solar radiation decreases, on both dates, there is a decline in the depth of the CBL. In that, PALM represented adequately in this interval, however, at the end of the simulation (18 and 19 LT), an underestimation was found, around 30 and 50 m for the rainy and dry day, respectively.
The PALM simulations of the CBL heights in relation to Ceilometer data proved to be satisfactory for the different days analyzed, since they exhibited a strong correlation denoted by R2 values greater than 0.90 (Figure 5).The Figure 5A showed a smooth overestimation of CBL heights for rainy day and, the dry day (Figure 5B) showed a smooth underestimated.
Figure 5 – Scatterplots of PALM in relation to Ceilometer data from PBL height to rainy day (A) and, dry (B)
4 Conclusion
In general, the PALM model proved to be efficient in simulating the cycle of the convective boundary layer for the Amazon region, which presented a small overestimation (of approximately 35 m) in the growth of CLC for the rainy day and a small underestimation. (Approximately 20 m) for the dry day. During the phases in which CBL is well established (between 11 and 16 LT) and in the decay of convection (between 17 and 18 LT), the PALM model showed better results for the dry day compared to the rainy day, due to the greater cloudiness in the rainy day.
It was also found that the growth of CBL in the PALM model quickly responded to the surface fluxes for both situations, which demonstrates it’s powerful for turbulence analysis.
Acknowledgments
The authors thanks, the Institutional supports were to provide by the National Institute of Space Research (INPE), the National Institute of Amazonian Research (INPA), and Amazonas State University (UEA). The authors acknowledge the Brazilian National Council for Scientific and Technological Development (CNPq) to support the financial and thank the GoAmazon project group for providing the data available for this study.
References
CARNEIRO, R. G.; FISCH, F.; KAUFMANN, T. Determinação da Altura da Camada Limite Planetária na Floresta Amazônica Utilizando um Ceilometer. Ciência e Natura, v. 38, p. 460–466, https://doi.org/10.5902/2179460X20265, 2016.
CARNEIRO, R. G.; FISCH, G. Observational analysis of the daily cycle of the planetary boundary layer in the central Amazon during a non-El Niño year and El Niño year (GoAmazon project 2014/5), Atmospheric Chemistry and Physics, v. 20, p. 5547–5558, https://doi.org/10.5194/acp-20-5547-2020, 2020.
CARNEIRO, R. G.; FISCH, G.; BORGES, C. K.; HENKES, A. Erosion of the nocturnal boundary layer in the central Amazon during the dry season, Acta Amazonica, v. 50, p. 80–89, https://doi.org/10.1590/1809-4392201804453, 2020.
HEINZE, R.; MOSELEY, C.; BÖSKE, L. N.; MUPPA, S. K.; MAURER, V.; RAASCH, S.; STEVENS, B. Evaluation of large-eddy simulations forced with mesoscale model output for a multi-week period during a measurement campaign. Atmospheric Chemistry and Physics, v. 17, p. 7083–7109, https://doi.org/10.5194/acp-17-7083-2017, 2017.
KANDA, M.; INAGAKI, A.; LETZEL, M.; RAASCH, S.; WATANABE, T. LES study of the energy imbalance problem with eddy covariance fluxes. Boundary-Layer Meteorology, v. 110, p. 381–404, https://doi.org/10.1023/B:BOUN.0000007225.45548.7a, 2004.
KAUFMANN, T.; FISCH, G. Validação do modelo LES PALM por meio de dados de radiosondagens e de aeronave coletados durante o Experimento GoAmazon. Ciência e Natura, v. 38, p. 38–40, https://doi.org/10.5902/2179460X19831, 2016.
LETZEL, M. O.; RAASCH, S. Large Eddy Simulation of Thermally Induced Oscillations in the Convective Boundary Layer. Journal of the Atmospheric Sciences, v. 60, p. 2328–2341, https://doi.org/10.1175/1520-0469(2003)060<2328:LESOTI>2.0.CO;2, 2003.
MARONGA, B.; Banzhaf, S.; Burmeister, C.; Esch, T.; Forkel, R.; Fröhlich, D.; et al. Overview of the PALM model system 6.0. Geoscientific Model Development, v. 13, p. 1335–1372, https://doi.org/10.5194/gmd-13-1335-2020, 2020.
MARONGA, B.; GRYSCHKA, M.; HEINZE, R.; HOFFMANN, F.; KANANI-SüHRING, F.; KECK, M.; KETELSEN, K.; LETZEL, M. O.; SüHRING, M.; RAASCH, S. The Parallelized Large-Eddy Simulation Model (PALM) version 4.0 for atmospheric and oceanic flows: model formulation, recent developments, and future perspectives. Geoscientific Model Development, v. 8, p. 2515–2551, https://doi.org/10.5194/gmd-8-2515-2015, 2015.
MARONGA, B.; RAASCH, S. Large-eddy simulations of surface heterogeneity effects on the convective boundary layer during the LITFASS-2003 Experiment. Boundary-Layer Meteorology, v. 46, p. 17–44, https://doi.org/10.1007/s10546-012-9748-z, 2013.
MARTIN, S.; ARTAXO, P.; MACHADO, L. A. T.; MANZI, A. O.; SOUZA, R. A. F.; SCHUMACHER, C.; WANG, J.; ANDREAE, M. O.; BARBOSA, H. M. J.; FAN, J.; FISCH, G.; GOLDSTEIN, A. H.; GUENTHER, A.; JIMENEZ, J. L.; POSCHL U. ANDSILVA DIAS, M. A.; SMITH, J. N.; WENDISEH, M. Introduction: Observations and Modeling of the Green Ocean Amazon (GoAmazon 2014/5). Atmospheric Chemistry Physics, v. 16, p. 4785–4797, https://doi.org/doi:10.5194/acp-16-4785-2016, 2016.
NEVES, T.; FISCH, G.; RAASCH, S. Local Convection and Turbulence in the Amazonia Using Large Eddy Simulation Model. Atmosphere, v. 9, p. 1–13, https://doi.org/10.3390/atmos9100399, 2018.
NEVES, T. T. de A. T.; FISCH, G. The Daily Cycle of the Atmospheric Boundary Layer Heights over Pasture Site in Amazonia. American Journal of Environmental Engineering, v. 5, p. 39–44, https://doi.org/ doi: 10.5923/s.ajee.201501.06, 2015.
RAASCH, S.; FRANKE, T. Structure and formation of dust devil-like vortices in the atmospheric boundary layer: A high-resolution numerical study. Journal of Geophysical Research Atmospheres, v. 116, p. 1–16, https://doi.org/10.1029/2011JD016010, 2011.
RAASCH, S.; SCHRÖTER, M. PALM – A large-eddy simulation model performing on massively parallel computers. Meteorologische Zeitschrift, v. 10, p. 363–372, https://doi.org/10.1127/0941-2948/2001/0010-0363, 2001.
SHUKLA, K. K.; PHANIKUMAR, D. V.; NEWSOM, R. K.; KUMAR, K. N.; RATNAM, M. V.; NAJA, M.; SINGH, N. Estimation of the mixing layer height over a high-altitude site in Central Himalayan region by using Doppler lidar. Journal of Atmospheric and Solar-Terrestrial Physics, v. 109, p. 48–53, https://doi.org/10.1016/j.jastp.2014.01.006, 2014.