Send the link below via email or IMCopy
Present to your audienceStart remote presentation
- Invited audience members will follow you as you navigate and present
- People invited to a presentation do not need a Prezi account
- This link expires 10 minutes after you close the presentation
- A maximum of 30 users can follow your presentation
- Learn more about this feature in our knowledge base article
Do you really want to delete this prezi?
Neither you, nor the coeditors you shared it with will be able to recover it again.
Make your likes visible on Facebook?
You can change this under Settings & Account at any time.
Transcript of Untitled Prezi
2) Finnish Meteorological Institute, Kuopio, Finland.
3) Finnish Meteorological Institute, Helsinki, Finland. Based on the reference data from the RT-model, what kind of instantaneous ADRE should we expect from the satellite based method over the study area? MODIS AOD Instantaneous ADRE The essential question is how well the satellite based method produces the aerosol-free flux? Can we predict in which kind of conditions the satellite based method is working well or not so well? based on the RT simulations we expect to obtain mostly negative ADRE
positive ADRE possible in the north western part of the study area Modeled aerosol-free SW TOA-flux Aerosol-free TOA-flux from satellite based fitting Aerosol-free TOA-flux anomaly
Satellite method - RT model In 58% of the cases the satellite based aerosol-free flux is within 10Wm-² from the modeled value.
23% of the cases satellite based aerosol-free flux is more than 10Wm-² larger than model
19% of the cases satellite based aerosol-free flux is more than 10Wm-² lower.
In summer the number of "extreme differences" is relatively largest. correlation coefficient between AOD and TOA flux vs. aerosol-free flux difference (satellite fit - model) Dynamic AOD range vs. aerosol-free flux difference Data used in this study Remote sensing data Reference data CERES-SSF product (In this study Terra)
combines CERES broad band observations with coincident aerosol and cloud observations from the higher-resolution MODIS instrument.
includes also meteorological information from Global Modeling and Assimilation Office (GMAO), e.g. amount of precipitable water.
Nadir resolution is 20 km.
MODIS surface albedo (MCD43C3)
used in the RT-simulations CERES observations Unknown Aerosol radiative effect (ADRE) at TOA is defined as the difference of the fluxes without and with aerosols present in the atmosphere: ADRE < 0, cooling
ADRE > 0, warming ADRE includes contribution from both natural and antropogenic aerosols
In this work ADRE is studied within the short wave region for a cloud-free sky.
Approaches used to derive ADRE:
Radiative transfer calculations
Radiative transfer calculations coupled with remote sensing observations
Remote sensing observations Motivation for this study To get an estimate of ADRE over eastern China.
In the previous publications the method is described well, but not studied in detail.
Cases of positive ADRE
Is this method working over some surface/ with some aerosol type/loading better than other and is there some parameter indicating that?
How would the results compare with reference data e.g. from radiative transfer model? The aerosol-free flux can not be obtained from the satellite measurements.
The estimate for aerosol-free flux is obtained by establishing a linear relationship between coincident broad band flux and AOD observations. The satellite based method has been used
over ocean e.g. by:
Loeb and Manalo-Smith, 2005; ADRE over global oceans.
Zhao et al., 2008; Derivation of component aerosol direct radiative forcing over ocean, combined study of satellite based ADRE and GOCART model.
Christopher, 2011; comparison of satellite based and RT-derived ADRE over global oceans.
over land e.g. by:
Patadia et al., 2008; Global ADRE over land.
Sena et al., 2013; The impact of deforestation in the Amazonian atmospheric radiative balance.
over land and ocean:
Feng and Christopher, 2013; Satellite and surface-based remote sensing of Southeast Asian aerosols and their radiative effects- It is assumed that the aerosol type does not change systematically.
Linear regression between the TOA flux and AOD ( < 2.0), extrapolate to AOD=0
Criteria for a successful regression e.g.:
observations are flagged cloud-free (based on higher resolution MODIS)
Number of observations/month ≥ 10
correlation coefficient ≥ |0.2| Data and Methods The LibRadtran radiative transfer code
for "theoretical study"
to model aerosol-free fluxes for reference
reference fluxes for normalization AERONET inversion data
ADRE, TOA-fluxes with and without aerosols TOA flux vs. SZA TOA flux vs. AOD for weakly and strongly absorbing aerosols the coincident cloud-free TOA-flux - AOD observations are collected in each 0.5 deg. grid cell over one month
To reduce the noise in the flux observations, the CERES fluxes are normalized to a fixed SZA, precipitable water content and day-of-year before the linear regression.
Reference fluxes are simulated with LibRadtran. Absorbing aerosols Weakly absorbing aerosols NOTE that the ADRE obtained with this method is instantaneous ADRE MODIS black-sky albedo Model simulations Model simulations March 2009 March 2009 October 2009 October 2009 March 2009 March 2009 March 2009 October 2009 October 2009 (Model) October 2009 (satellite method) Satellite aerosol-free flux - modeled flux All grid cells March-October 2009 The smallest difference to the model is obtained when correlation is greater than 0.6, however outliers still exist.
In cases of negative correlation (-> ADRE > 0) satellite estimate for aerosol-free flux is higher than from model. MODIS black sky albedo vs. aerosol-free flux difference When albedo is about 0.11-0.16, the difference between satellite and model is mostly within 10Wm-².
The correlation between aerosol-free flux difference and albedo is -0.47
Over bright surfaces the satellite based aerosol-free flux is smaller than from model. References Loeb and Manalo-Smith (2005). Top-Of-Atmosphere Direct Radiative Effect of Aerosols over Global Oceans from Merged CERES and MODIS observations. J. Climate, vol. 18, 3506-3526.
Zhao et al. (2008). Derivation of component aerosol direct radiative forcing at the top of the atmosphere for clear-sky oceans. J. Quant. Spectrosc. Radiat. Transfer, 109, 1162-1186.
Christopher (2011). Satellite remote sensing methods for estimating clear Sky shortwave Top of Atmosphere fluxes used for aerosol studies over the global oceans. Remote Sens. Environ., 115, 3002-3006.
Patadia et al. (2008). First observational estimates of global clear sky shortwave aerosol direct radiative effect over land. Geophys. Res. Lett., vol. 35.
Sena et al. (2013). Spatial variability of the direct radiative forcing of biomass burning aerosols and the effects of land use change in Amazonia. Atmos. Chem. Phys., 13, 1261-1275.
Feng and Christopher (2013). Satellite and surface-based remote sensing of Southeast Asian aerosols and their radiative effects. Atmos. Res., vol. 122, 544–554.
Sundström et al. (2013). Estimating the direct aerosol radiative effect over Estern China using multi-sensor satellite remote sensing measurements. In preparation. The case of positive ADRE, true of method artifact? Dynamic AOD range = Max(AOD) - Min(AOD) TOA flux vs. AOD and MODIS SW black-sky albedo Over bright surface dynamic AOD range is small During the summer months positive ADRE was obtained in unexpected places.
The AOD-TOA flux correlation was highly negative (-0.5), even after the normalization.
According to model simulations, the surface is so dark that the aerosols should have SSA ~ 0.7 to produce positive ADRE.
The TOA-flux values are considerably high with relatively low AOD.
Possible explanations: subvisual clouds, change in aerosol type or both. The direct aerosol radiative effect has been studied over Eastern China using satellite remote sensing based approach
The focus has been studying the method.
Results indicate that the normalization of TOA fluxes especially to a fixed SZA before actual fitting increases the correlation and decreases the difference of obtained ADRE to the reference data.
The aerosol free flux obtained from the satellite fitting method has similar spatial pattern than what is obtained from the RT model.
58% of all the cases are within 10 Wm-² from the modeled values.
The positive ADRE during summer months is most probably a method artifact, caused by subvisual clouds, change in aerosol type or both. The estimate for aerosol-free flux Normalization of CERES TOA-fluxes Correlation: 0.7
(satellite method - model) 0.5x0.5 degree grid Study period:
March-October 2009 Study area Note, color scale in this figure denotes albedo!