Vertically resolved separation of dust and other aerosol types by a new lidar depolarization method

Tao Luo; Zhien Wang; Richard A. Ferrare; Chris A. Hostetler; Renmin Yuan; Damao Zhang

doi:10.1364/OE.23.014095

1. Introduction

Dust aerosols are well known for the role in modulating the climate system at local and global scales, with great importance for many aspects of atmospheric science including radiative transfer, cloud microphysics, atmospheric chemistry, oceanic biogeochemical processes, and air quality [1–8]. However, current models still have large uncertainties in simulating dust optical depth, vertical extinction profiles and seasonal variations [9,10], which stresses the need for better dust observations to improve the dust related processes in the models [10,11]. A challenging part of dust observations is that dust aerosols are usually mixed with the other types of aerosols [12–15], and over 50% of the dust is polluted [12]. Mistreating the polluted dusts as pure dust could result in as much as ~200% overestimation of dust loading [15]. Therefore, a reliable height-resolved dust partition approach is needed to provide more accurate dust loading estimations, and thus to improve our understanding of the mixing of these climate-relevant aerosol components, the long-range transport of the dust, and the impact of aerosols on regional climate [13,16,17].

Lidar linear depolarization is a useful measurement to separate dust from the other types of aerosols [14,18–26]. The total depolarization ratio can be interpreted as the linear combination of particle depolarization ratio of dust and no-dust aerosols [14,20,25]. Therefore, the dust backscattering coefficients can be separated from the retrieved aerosol backscattering coefficients with the retrieved particle depolarization, and the assumed dust and non-dust depolarization ratios [i.e., 14,23]. To retrieve the particle depolarization ratios, the backscatter ratio (the ratio of the total backscatter coefficient to the molecular component) must be retrieved with the prior lidar ratio (S), i.e., vertically consistent S base on the other studies [23], or retrieving the height-resolved S with an additional measurement such as a Raman channel or High Spectral Resolution Lidar (HSRL) [14,27]. Groß et al. [20] presented the relationships of particle linear depolarization ratio and S in term of dust fraction (ratio of dust extinction and total aerosol extinction). The mixing lines from these relationships with selected dust fractions can describe the measured linear depolarization ratio–S relationship from HSRL [21]. Burton et al. [25] further simplified the mixing rules in [20] and developed a comprehensive and unified set of rules for characterizing the external mixing of S, backscatter color ratio, and depolarization ratio, and thus to retrieve the dust fraction from these intensive parameters measured by HSRL. It is also possible to separate multi-component aerosols with multiple channel lidar measurements together with T-matrix (for dust) and Mie scattering (for non-dust) calculations by assuming dust and non-dust aerosol size distributions and complex refractive index [27,28]. However, these ground-based measurements mainly provide local aerosol information.

The Cloud-Aerosol Lidar with Orthogonal Polarization (CALIOP) on board the Cloud-Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) satellite provide unique global measurements of aerosol vertical distributions, which is an unprecedented data set to study the dust generation and three-dimensional (3D) transport from regional to global scale [11,28–32]. Efforts [15,33] were made to separate the dust extinctions from polluted dust aerosols by using the CALIPSO level-2 product with the method in [14]. However, applying the ground-based methods in literatures as mentioned above to the CALIPSO observations suffers several difficulties mainly caused by the limited capabilities of CALIOP. There is no additional measurement to provide prior height-resolved S information for these global measurements. Therefore, the aerosol backscattering and particle depolarization ratio can only be retrieved with an assumed S relating to the aerosol types. However, it is challenging to subtype certain aerosols, i.e., polluted continental and smoke, because their optical properties are similar in CALIOP observations [12]. Another problem is CALIPSO level-2 product treats the polluted dust only as dust mixing with smoke with S of 65 sr [8], which is not proper when dust mixing with other types of aerosols such as sea salt [21,25]. Moreover, the dust mask in the CALIPSO level 2 can miss significant thin dust layer cases and may misclassify clouds as dust [32].

This paper aims to provide an improved method to separate dust aerosol extinction from the other types of aerosols with CALIPSO Level-1 data. The theoretical framework of lidar retrieval for external mixing of dust with other aerosols is presented in the section 2. The reliabilities of new methodology were fully tested in the section 3 under even wrong subtyping conditions. Then the new method was applied to CALIPSO data with a retrieval example presented in the section 4. Conclusions are briefed in the section 5.

2. Theory

For dust external mixed with non-dust aerosols, the single-scattering lidar equation [21],

β_{T o t a l}^{'} (z) = β_{⊥}^{'} (z) + β_{∥}^{'} (z) = (β_{a} (z) + β_{m} (z)) e^{- 2 \int_{0}^{z} (S_{a} β_{a} (z') + S_{m} β_{m} (z')) d z'}

can be rewritten as

β_{T o t a l}^{'} (z) = (β_{d} (z) + β_{n d} (z) + β_{m} (z)) e^{- 2 \int_{0}^{z} (S_{d} β_{d} (z') + S_{n d} β_{n d} (z') + S_{m} β_{m} (z')) d z'}

Here,

β_{T o t a l}^{'}

is the total attenuated backscattering;

β_{∥}^{'}

is the parallel component of the attenuated backscattering;

β_{⊥}^{'}

is the cross-polarized component of the attenuated backscattering; β_a, β_d, β_nd and β_m are volume backscattering coefficients for mixed aerosol, dust, non-dust and molecules respectively; S_a, S_d, S_nd and S_m are lidar ratios for mixed aerosol, dust, non-dust and molecules respectively.

Similar to [25], we can defined

δ_{p}^{'} (z) = \frac{β_{a, ⊥} (z)}{β_{a, ∥} (z) + β_{a, ⊥} (z)} = \frac{δ_{p} (z)}{δ_{p} (z) + 1} = \frac{δ_{d}^{'} (z) β_{d} (z) + δ_{n d}^{'} (z) β_{n d} (z)}{β_{d} (z) + β_{n d} (z)}

Here,

δ_{p} (z) = β_{a, ⊥} (z) / β_{a, ∥} (z)

; β_a,⊥ and β_a,|| are the cross-polarized and parallel components of the mixed aerosol backscattering coefficients;

δ_{d}^{'} (z) = δ_{d} (z) / (δ_{d} (z) + 1)

;

δ_{n d}^{'} (z) = δ_{n d} (z) / (δ_{n d} (z) + 1)

;

δ_{m}^{'} (z) = δ_{m} (z) / (δ_{m} (z) + 1)

; δ_d, δ_nd and δ_m are depolarization ratios for dust, non-dust and molecules respectively.

Form (3), we can get

β_{n d} (z) = c β_{d} (z)

c (z) = \frac{δ_{p}^{'} (z) - δ_{d}^{'} (z)}{δ_{n d}^{'} (z) - δ_{d}^{'} (z)}

Then, substituting Eq. (4) into Eq. (2), Eq. (2) can be reformed as

β_{T o t a l}^{'} = (β_{a} (z) + β_{m} (z)) e^{- 2 \int_{0}^{z} (S_{e f f} β_{a} (z^{'}) + S_{m} β_{m} (z^{'})) d z^{'}}

β_{a} (z) = (1 + c (z)) β_{d} (z)

S_{e f f} (z) = \frac{S_{d} + c (z) S_{n d}}{1 + c (z)}

Here, S_eff is the effective S for the mixed aerosol. Equations (6)-(8) can also be derived by substituting Eqs. (15) and (27) in [25] into Eq. (1) to cancel the dust fraction item.

We represented the similar derivation in [14,20,25] with the form of the lidar equation to solve it in the following. Because there are no height-resolved intensive parameters of mixed aerosol available from other supplement measurements for CALIPSO, the assumption of the consistent optical properties (S_d, δ_d, S_nd and δ_nd) for dust and non-dust aerosols within the mixed aerosol layer was used to make Eq. (6) solvable. Then, iterative steps are designed to solve Eq. (6) as [34] with the separation of the dust and non-dust aerosol contributions during the retrieval stage, as summarized as followings:

Step 1: Solving Eq. (1) with S of 65 sr to estimate the β_a. Then the δ_p can be estimate as

δ_{p} (z) = \frac{β_{⊥} (z) - δ_{m} β_{m} e^{- 2 \int_{0}^{z} (S β_{a} (z^{'}) + S_{m} β_{m} (z^{'})) d z^{'}}}{β_{∥} (z) - β_{m} e^{- 2 \int_{0}^{z} (S β_{a} (z^{'}) + S_{m} β_{m} (z^{'})) d z^{'}}}

Step 2: Calculating c and S_eff by Eq. (5) and (8 $)$ . If δ_p is close to δ_nd at certain height range, the S_eff = S_nd and β_nd = β_a at this height range in the solutions; if δ_p is larger than δ_d at certain altitude range, then the S_eff = S_d and β_d = β_a at this altitude range in the solutions.

Step 3: Solving Eq. (6) with S_eff to retrieve β_a. Then δ_p is re-calculated with S_eff and retrieved β_a, and then go back to the step 2.

Step 4: Repeating step 2 to 3 until δ_p converges. Usually 2-3 iterations are enough. Then β_d and β_nd are calculated by using Eq. (7).

The pure dust and non-dust aerosols optical properties can be chosen based on other studies [i.e 20.] after subtyping the mixed aerosols. Due to the limitations of CALIPSO as mentioned in the section 1, sensitivity experiments were designed to examine the impact of these parameters on dust extinction retrievals especially under the wrong subtyping conditions in the next section.

3. Sensitivity experiments

For experiments in this section, the input signals were calculated according to

\begin{matrix} β_{T o t a l}^{'} = (β_{d} (z) + β_{n d} (z) + β_{m} (z)) e^{- 2 \int_{0}^{z} (S_{d} β_{d} (z^{'}) + S_{n d_i n i} β_{n d} (z^{'}) + S_{m} β_{m} (z^{'})) d z^{'}} \\ β_{⊥}^{'} = (δ_{d}^{'} β_{d} (z) + δ_{n d}^{'} β_{n d} (z) + δ_{m}^{'} (z)) e^{- 2 \int_{0}^{z} (S_{d} β_{d} (z^{'}) + S_{n d_i n i} β_{n d} (z^{'}) + S_{m} β_{m} (z^{'})) d z^{'}} \end{matrix}

Here, $β_{d} (z) = β_{d_n o r m} (z) τ_{d} / S_{d}$ and $β_{n d} (z) = β_{n d_n o r m} (z) τ_{n d} / S_{n d_i n i}$ are calculated based on the normalized dust and non-dust backscattering coefficients (as shown Fig. 1, $β_{d_n o r m} = β_{d_i n p u t} / \int_{0}^{z t} β_{d_i n p u t} (z^{'}) d z^{'}$ and $β_{n d_n o r m} = β_{n d_i n p u t} / \int_{0}^{z t} β_{n d_i n p u t} (z^{'}) d z^{'}$ , where β_{d_input} and β_{nd_input} are selected dust and non-dust backscattering coefficients from typical boundary aerosol layers); S_{nd_ini} is the S_nd used for calculating input signals; zt is the aerosol layer top; τ_d and τ_nd are input dust and non-dust optical depth.

Fig. 1 The normalized backscattering coefficients of dust and non-dust aerosol used for the sensitive experiments. Those profiles are used to simulate the mixing of an evaluated dust layer with a boundary layer non-dust aerosol.

Download Full Size | PDF

Based on the fact that the clean continental and polluted continental aerosols have similar lidar ratios and depolarization ratios as the sea salt and smoke respectively, only dust mixing with sea salt and smoke were considered here, denoted as SS and SM respectively. Four sets of experiments were designed, which are: (1) SS-SS: dust mixing with sea salt treated correctly; (2) SM-SM: dust mixing with smoke treated correctly; (3) SS-SM: dust mixing with sea salt wrongly treated as dust mixing with smoke; (4) SM-SS: dust mixing with smoke wrongly treated as dust mixing with sea salt. Two selected mixing situations and full tests with wide mixing situations were performed. The parameters used for input and new two-component retrievals (Sim2) were summarized in Table 1, in which, the δ_d and δ_nd were chosen based on the mean value from CALIPSO level 2 products to be more preventative over the global. The retrievals by the general single-component method (Sim1) with S of 65 sr for polluted dust [12] were also performed, following with the dust partition from the Sim1 retrieved δ_p and β_a by Eq. (7). The subscript of “_ret” denotes the parameter used in retrieval in the Table 1 and the following.

Table 1. The input (upper panel) and retrieval (lower panel) parameters used in the simulation experiments. The sea salt S is from [39]. Other parameters are based on [12] and used in the CALIPSO level 2 products.

View Table | View all tables in this article

Retrieved profiles of β_d, β_a and δ_p from the two selected mixing situations for SS and SM by Sim2 and Sim1 were shown in the Fig. 2. The δ_{d_ret} and δ_{nd_ret} in these experiments were set as their true values to focus on the influence of S_{nd_ret} and the retrieval methods on dust backscattering coefficients retrievals. As shown in Fig. 2, when correct dust and non-dust information is used in the retrievals (SS-SS and SM-SM), the two-component method retrieves the true β_d and β_nd as expected (indicating the true value in Fig. 2 as red line). If the non-dust aerosol type is misclassified (SS-SM and SM-SS, the blue lines), the two-component method overestimates (SS-SM) or underestimates (SM-SS) β_nd as expected. However, the retrieved β_d is still very close to the true values, as indicated by the layer means of the absolute relative errors (the absolute difference between retrieval and the true value divided by the true value) given in the Table 2. It indicates that depolarization information provides a strong constraint to obtain the reliable dust retrievals and to compensate the error due to wrong lidar ratio to a certain extent. Misclassifying the mixed sea salt as smoke will result in relatively larger errors than those by misclassifying the mixed smoke as sea salt. For Sim1 (the green lines), the single-components method can only provide the total backscattering coefficients close to the true value when the used S is close to the S_eff. However, the performance of separation based on single-component method is worse than that of the two-component method. Furthermore, Fig. 2 clearly shows that the retrieved δ_p is not very sensitive to the choice of retrieval method because the molecular depolarization ratio is very small.

Fig. 2 Simulation results for the SS experiments (a1-b3) and SM experiments (c1-d3). The first panel (a1-a3) shows SS experiment of the weak dust mixing case. The second panel (b1-b3) shows the SS experiment of moderated dust mixing case. The third panel (c1-c3) shows SM experiment of the weak dust mixing case. The forth panel (d1-d3) shows the SS experiment of moderated dust mixing case. The true value is the same as SS-SS or SM-SM (the red lines).

Download Full Size | PDF

Table 2. Summaries of the layer mean of the absolute relative errors in β_d for the selected (upper panel).

View Table | View all tables in this article

The performances of the two-component method were then fully tested under wide mixing situations. The conclusions are similar to those from selected mixing situation experiments. Therefore, only the layer mean of the absolute relative errors in β_d from the sensitive tests with retrieval parameters of δ_{d_ret} = 0.25, δ_{nd_ret} = 0.04 were shown in Fig. 3. The full results are summarized in Table 3. In Sim2, δ_{d_ret} is key source of uncertainties in the estimations of dust backscattering with resulted error up to ~26%; δ_{nd_ret} is less important but with resulted additional error up to ~11% in some worse conditions; the misclassification of non-dust aerosol type generally results in additional errors less than ~12%. The Sim2 have better performances than Sim1.

Fig. 3 The layer mean of the absolute relative errors in β_d from full sensitive tests with retrieval parameters of δ_{d_ret} = 0.25, δ_{nd_ret} = 0.04, and others as in Table 1.

Download Full Size | PDF

Table 3. Summaries of the layer mean of the absolute relative errors in β_d for full (lower panel) experiments. The errors in β_d for full tests are averaged for each experiment with assigned δ_{d_ret} and δ_{nd_ret} (i. e. over each plot in Fig. 3).

View Table | View all tables in this article

Therefore, if only the dust extinction profiles were needed, information from δ_p and δ_d can be used as a strong constrains to provide reasonable dust retrievals even without prior information of the non-dust aerosols. The new two-component method can resolve the dust aerosols in the mixed aerosols in the lidar equation inversion stage and is better than doing the partitions based on lidar inversion products as in the former studies [15,33].

4. Applications to CALIPSO measurements

A dust mixed with sea salt case measured by CALIOP as well as the NASA Langley airborne HSRL [35] at 04:00:00 on August 24 2010 was used to evaluate the new two-component methods, as shown in Fig. 4. This long-distance transported and elevated dust layer subsided into the marine boundary layer and mixed with the sea salt, as indicated by the HSRL measured δ_p [Fig. 4(a2)] and CALIOP retrieved δ_p [Fig. 4(b2)]. δ_d = 0.3, S_d = 43 sr, δ_nd = 0.075 and S_nd = 24 sr were used in the following retrievals based on the HSRL measured $δ_{p}^{'} - S$ mean relationship (red dot-line in Fig. 5(a)), which agree well with the parameterized one by Eq. (8) (dashed-dot line in Fig. 5(a)). For this roomHSRL, β_aero can be reliably determined; thus dust and sea salt contributions (β_dust and β_seasalt) to β_aero can be determined with measured δ_p, δ_d = 0.3 and δ_nd = 0.075 based on Eq. (7), shown as in Figs. 4(a5)-4(a6) respectively. However, we also can retrieve these by applying the proposed method to the part of HSRL measurements. Firstly, the aerosol mask was identified with the measured δ_p>2* δ_m and then the two-component method was applied to HSRL measured aerosol $β_{T o t a l}^{'}$ and $β_{⊥}^{'}$ to retrieve β_a, β_d and β_nd with the aerosol layer top determined based on aerosol masks. The retrieved β_a in Fig. 4(a4) accords well with the β_aero in Fig. 4(a3), with the mean difference of −2 × 10⁻⁴ sr⁻¹km⁻¹ and the mean absolute relative difference of 7.5%. The retrieved $β_{d}$ (not shown here) has the mean difference is −0.7 × 10⁻⁴ sr⁻¹km⁻¹ and the mean absolute relative difference is 9%, as compared to β_dust.

Fig. 4 A dust mixed with sea salt case observed by CALIPSO and under-fly HSRL on 2010/08/24/. (a1-a6) are from HSRL measurements or derived properties: (a1-a3) are the measured $β_{T o t a l}^{'}$ , δ_p, and aerosol backscattering coefficients β_aero; (a4) is retrieved β_a with the new two component method; (a5) and (a6) are β_dust and β_seaosalt resolved from β_aero with measured δ_p and Eq. (7). (b1-b5) are CALIPSO measured or derived properties: (b1) measured $β_{T o t a l}^{'}$ ; and (b2 – b5) are retrieved δ_p, β_a, β_d and β_nd with the new two component method.

Download Full Size | PDF

Fig. 5 (a) The $δ_{p}^{'} - S$ relationship from HSRL measurements overlaid with the measured mean $δ_{p}^{'} - S$ relationship (the red dot-line) and the predicted S_eff by Eq. (8) (black dash line); (b) the mean profiles from HSRL and CALIPSO measurements and retrieval; (c) comparison between CALIPSO retrieved β_d and HSRL derived β_dust; (d) comparison between CALIPSO and HSRL measured $β_{T o t a l}^{'}$ .

Download Full Size | PDF

Then, the two-component method was applied to CALIOP observations with the input parameters of δ_d = 0.3, S_d = 43 sr, δ_nd = 0.075 and S_nd = 24 sr. The dust mask was determined with the new dust layer detection method to provide improved optically thin dust detections, and with improved cloud detections, especially for optical thin ice clouds, by combining CALIPSO lidar and CloudSat observations as detailed in [32]. The molecular backscattering was calculated by ECMWF-AUX data sets. A 30-km moving smoothing was used to deal with its low SNR, so as to better retrieve the thin aerosol layers and to make the solution more stable with less noise influence [32]. The δ_p, β_d and β_nd were retrieved by the new two-component method using horizontal 30km moving smoothing for dusty profiles only, as presented in Figs. 4(b3-b5).

The CALIOP retrieved β_d and β_nd [Figs. 4(b4)-4(b5)] show very similar patterns to those from HSRL [Figs. 4(a5)-4(a6)]. The retrieved sea salt aerosol layer is mainly located within boundary layer with top lower than ~2km, and the dust layer shows a strong elevated layer and a subsidence layer. Compared to HSRL, the CALIOP retrievals (when height >0.5km) have the mean difference of −2.13 × 10⁻⁴ sr⁻¹km⁻¹ and the mean absolute relative difference of 17% in β_a and of −1 × 10⁻⁴ sr⁻¹km⁻¹ and 28% in β_d. The differences should be mainly inherited from the uncertainties in CALIOP measurements. As shown in Fig. 5(c)-5(d), both β_d and $β_{T o t a l}^{'}$ are underestimated as compared to HRSL. The mean difference and the mean absolute relative difference (when height >0.5km) for CALIOP measured $β_{T o t a l}^{'}$ are −6 × 10⁻⁴ sr⁻¹km⁻¹ and 31% (before 30km average), respectively. Those differences could be resulted by the poor SNR in daytime CALIOP observations, as can be seen in measured $β_{T o t a l}^{'}$ [Fig. 4(b1)], or the calibration problem. The nighttime CALIOP observations have better SNR and thus a better performance of the new two-component method can be expected. Another potential issue is that the HSRL and CALIOP are slightly temporally and spatially mismatched. The HSRL measurements started at the time when CALIPSO overpassed and then flew along the CALIPSO track. However, these two observations did not measure exactly the same aerosol and clouds, especially in the lower level where the aerosols and clouds are more variable due to the impacts of small scale boundary layer processes, as indicated in measured $β_{T o t a l}^{'}$ [Fig. 4(a1)-4(b1)] and relatively larger differences in the lower 1km as shown in Fig. 5(b). Considering these, the differences between HSRL and CALIOP retrievals are reasonable, which supports that the new two-component method can separate the dust layer well from the mixed aerosols.

Thirdly, sensitivity tests were performed to show the uncertainties in CALIOP retrievals with the referee of retrieved β_d in Fig. 4(b4). One selected parameter was changed at each time, and the mean differences and the mean absolute relative differences were calculated and summarized in the Table 4. Moreover, those sensitive tests show that the mean absolute relative differences are less than 2% in δ_p compared to the retrieved δ_p in Fig. 4(b2). These sensitivity tests and those in the last section both indicated that the miss subtyping of mixed non-dust aerosols (such as sea salt, smoke, clean continental and polluted continental) and using wrong S have minor impacts. The δ_d and δ_nd are the two most important parameters controlling the retrieval accuracies. Fortunately, these two parameters can be derived from δ_p statistics regionally. Therefore, the new two-component method can correctly separate the dust from mixed aerosols and can provide reliable dust retrievals.

Table 4. Summaries of errors of β_d in CALIPSO sensitivity tests.

View Table | View all tables in this article

5. Conclusions and discussions

This paper presents a new two-component retrieval method to vertically resolve the dust aerosols in the mixed aerosols based on the lidar total and perpendicular channel measurements of the aerosols. The theoretical framework of external mixing of the dust and non-dust aerosol was developed and shows that the lidar ratio of the external mixing aerosols can be predicted and the contributions of irregular shape aerosols with large depolarization ratio and spherical shape aerosols with small depolarization ratio can be separated. A new two-component methodology was designed to resolve the dust contribution from the mixed aerosol in the lidar equation inversion stage, rather than doing partition based on lidar inversion products.

The detailed sensitivity tests showed that the particle depolarization ratio can be reliably determined and can be used as a strong constrain to resolve the dust extinctions. The new two-component method showed better performances than the single-component method in retrieving dust information even without prior information of the non-dust aerosols. The new method not only is suitable for ground-based or airborne measurements, but also can be applied for CALIPSO lidar measurements, which has difficulties in classifying the aerosol types in some situations.

Then, the new two-component method was implemented using the CALIOP measurements and compared with airborne HSRL. Applying the two-component retrieval method to HSRL data shows that the new methodology can well describe the lidar ratio of the mixed aerosols, and can retrieve total and dust aerosol backscattering coefficients with uncertainties less than 9%. Comparisons of co-incident CALIOP and HSRL retrieved show that the improved two-component method retrieves reliable dust backscattering with the CALIOP measurements. Further sensitivity tests based on this case show that the new method could correctly resolve dust and non-dust aerosols backscattering coefficient profiles within the mixed aerosols with uncertainties less than ~18%. However, as demonstrated by [36–38], it is important to consider multiple scattering effects in CALIOP measurements for optically thick aerosol layers with the choice of effective S. In this study, the impacts of multiple scattering are simply considered with an effective S of 40 sr in the section 3, which can be used in less dense layers where the multiple scattering is relatively insignificant with the dust extinction small than 1 km⁻¹ [36,37].

The new method will be applied to the new global dust database as built in [32] to provide better estimations of the global dust loading from CALIOP observations, which can be used to better understand the global dust generation and transportation processes and to evaluate and improve model simulations.

Acknowledgments

The CALIPSO data set was obtained from the NASA Langley Research Center Atmospheric Science Data Center (eosweb.larc.nasa.gov). The CloudSat 2B-GEOPROF and ECMWF-AUX products were obtained from CloudSat Data Processing Center (cloudsat.cira.colostate.edu). The HSRL data was obtained from NASA (science.larc.nasa.gov/hsrl/). This research was funded by NASA grant NNX13AQ41G and a contract from NASA/JPL.

References and links

1. J. Overpeck, D. Rind, A. Lacis, and R. Healy, “Possible role of dust induced regional warming in abrupt climate change during the last glacial period,” Nature 384(6608), 447–449 (1996). [CrossRef]

2. I. Tegen, A. A. Lacis, and I. Fung, “The influence on climate forcing of mineral aerosols from disturbed soils,” Nature 380(6573), 419–422 (1996). [CrossRef]

3. areP. Alpert, Y. J. Kaufman, Y. Shay-El, D. Tanre, A. da Silva, S. Schubert, and J. H. Joseph, “Quantification of dust-forced heating of the lower troposphere,” Nature 395(6700), 367–370 (1998). [CrossRef]

4. S. Kinne, M. Schulz, C. Textor, S. Guibert, Y. Balkanski, S. E. Bauer, T. Berntsen, T. F. Berglen, O. Boucher, M. Chin, W. Collins, F. Dentener, T. Diehl, R. Easter, J. Feichter, D. Fillmore, S. Ghan, P. Ginoux, S. Gong, A. Grini, J. Hendricks, M. Herzog, L. Horowitz, I. Isaksen, T. Iversen, A. Kirkevåg, S. Kloster, D. Koch, J. E. Kristjansson, M. Krol, A. Lauer, J. F. Lamarque, G. Lesins, X. Liu, U. Lohmann, V. Montanaro, G. Myhre, J. Penner, G. Pitari, S. Reddy, O. Seland, P. Stier, T. Takemura, and X. Tie, “An AeroCom initial assessment – optical properties in aerosol component modules of global models,” Atmos. Chem. Phys. 6(7), 1815–1834 (2006). [CrossRef]

5. IPCC, “Climate change 2013: the physical science basis (contribution of working group i to the fifth assessment report of the intergovernmental panel on climate change),” in: Stocker TF, Qin D, Plattner G-K, Tignor M, Allen SK, Boschung J, Nauels A, Xia Y, Bex V, Midgley PM, eds. Cambridge University, p1535 (2013).

6. J.-P. Chaboureau, E. Richard, J.-P. Pinty, C. Flamant, P. Di Girolamo, C. Kiemle, A. Behrendt, H. Chepfer, M. Chiriaco, and V. Wulfmeyer, “Long-range transport of saharan dust and its radiative impact on precipitation forecast: a case study during the convective and orographically-induced precipitation study (COPS),” Q. J. R. Meteorol. Soc. 137(S1), 236–251 (2011). [CrossRef]

7. D. Zhang, Z. Wang, A. Heymsfield, J. Fan, D. Liu, and M. Zhao, “Quantifying the impact of dust on heterogeneous ice generation in midlevel supercooled stratiform clouds,” Geophys. Res. Lett. 39(18), L18805 (2012). [CrossRef]

8. J. Fan, L. R. Leung, P. J. DeMott, J. M. Comstock, B. Singh, D. Rosenfeld, J. M. Tomlinson, A. White, K. A. Prather, P. Minnis, J. K. Ayers, and Q. Min, “Aerosol impacts on california winter clouds and precipitation during calwater 2011: local pollution versus long-range transported dust,” Atmos. Chem. Phys. 14(1), 81–101 (2014). [CrossRef]

9. A. T. Evan, C. Flamant, S. Fiedler, and O. Doherty, “An analysis of aerolian dust in climate models,” Geophys. Res. Lett. 41(16), 5996–6001 (2014). [CrossRef]

10. D. Kim, M. Chin, H. Yu, T. Diehl, Q. Tan, R. A. Kahn, K. Tsigaridis, S. E. Bauer, T. Takemura, L. Pozzoli, N. Bellouin, M. Schulz, S. Peyridieu, A. Chédin, and B. Koffi, “Sources, sinks, and transatlantic transport of north African dust aerosol: a multimodel analysis and comparison with remote sensing data,” J. Geophys. Res. 119, 6259–6277 (2014).

11. I. Uno, K. Eguchi, K. Yumimoto, T. Takemura, A. Shimizu, M. Uematsu, Z. Liu, Z. Wang, Y. Hara, and N. Sugimoto, “Asian dust transported one full circuit around the globe,” Nat. Geosci. 2(8), 557–560 (2009). [CrossRef]

12. H. Ali, “Omar, D. M. Winker, M. A. Vaughan, Y. Hu, C. R. Trepte, R. A. Ferrare, K.-P. Lee, C. A. Hostetler, C. Kittaka, R. R. Rogers, R. E. Kuehn, and Z. Liu, “The CALIPSO automated aerosol classification and lidar ratio selection algorithm,” J. Atmos. Ocean. Technol. 26, 1994–2014 (2009).

13. A. Petzold, A. Veira, S. Mund, M. Esselborn, C. Kiemle, B. Weinzierl, T. Hamburger, G. Ehret, K. Lieke, and K. Kandler, “Mixing of mineral dust with urban pollution aerosol over Dakar (Senegal): impact on dust physico-chemical and radiative properties,” Tellus 63(4), 619–634 (2011). [CrossRef]

14. M. Tesche, A. Ansmann, D. Müller, D. Althausen, R. Engelmann, V. Freudenthaler, and S. Groß, “Vertically resolved separation of dust and smoke over Cape Verde using multiwavelength raman and polarization lidars during Saharan Mineral Dust Experiment 2008,” J. Geophys. Res. 114(D13), D13202 (2009). [CrossRef]

15. E. Giannakaki, D. Balis, and V. Amiridis, “Vertical resolved separation of aerosol types using CALIPSO Level-2 Product,” Proc. SPIE 8182, 81820T (2011). [CrossRef]

16. B. T. Johnson, B. Heese, S. A. McFarlane, P. Chazette, A. Jones, and N. Bellouin, “Vertical distribution and radiative effects of mineral dust and biomass burning aerosol over West Africa during DABEX,” J. Geophys. Res. 113, D00C12 (2008).

17. G. Myhre, C. R. Hoyle, T. F. Berglen, B. T. Johnson, and J. M. Haywood, “Modeling of the solar radiative impact of biomass burning aerosols during the Dust and Biomass-Burning Experiment (DABEX),” J. Geophys. Res. 113, D00C16 (2008). [CrossRef]

18. A. Ansmann, M. Tesche, P. Seifert, S. Groß, V. Freudenthaler, A. Apituley, K. M. Wilson, I. Serikov, H. Linné, B. Heinold, A. Hiebsch, F. Schnell, J. Schmidt, I. Mattis, U. Wandinger, and M. Wiegner, “Ash and fine-mode particle mass profiles from EARLINET-AERONET observations over central Europe after the eruptions of the Eyjafjallajökull volcano in 2010,” J. Geophys. Res. 116, D00U02 (2011). [CrossRef]

19. D. Müller, I. Veselovskii, A. Kolgotin, M. Tesche, A. Ansmann, and O. Dubovik, “Vertical profiles of pure dust and mixed smoke-dust plumes inferred from inversion of multiwavelength Raman/polarization lidar data and comparison to AERONET retrievals and in situ observations,” Appl. Opt. 52(14), 3178–3202 (2013). [CrossRef] [PubMed]

20. S. Groß, M. Tesche, V. Freudenthaler, C. Toledano, M. Wiegner, A. Ansmann, D. Althausen, and M. Seefeldner, “Characterization of Saharan dust, marine aerosols and mixtures of biomass-burning aerosols and dust by means of multi-wavelength depolarization and Raman lidar measurements during SAMUM 2,” Tellus 63 B, 706–724 (2011).

21. S. Groß, M. Esselborn, B. Weinzierl, M. Wirth, A. Fix, and A. Petzold, “Aerosol classification by airborne high spectral resolution lidar observations,” Atmos. Chem. Phys. 13(5), 2487–2505 (2013). [CrossRef]

22. M. Tesche and A. Ausmann, “D. Mu ̈ller, D. Althausen, I. Mattisand co-authors, “Vertical profiling of Saharan dust with Raman Lidars and airborne HSRL in southern Morocco during SAMUM,” Tellus 61B, 144–164 (2008).

23. M. Esselborn, M. Wirth, A. Fix, M. Tesche, and G. Ehret, “Airborne high spectral resolution lidar for measuring aerosol extinction and backscatter coefficients,” Appl. Opt. 47(3), 346–358 (2008). [CrossRef] [PubMed]

24. A. Shimizu, N. Sugimoto, I. Matsui, K. Arao, I. Uno, T. Murayama, N. Kagawa, K. Aoki, A. Uchiyama, and A. Yamazaki, “Continuous observations of Asian dust and other aerosols by polarization lidars in China and Japan during ACE-Asia,” J. Geophys. Res. 109(D19), D19S17 (2004). [CrossRef]

25. S. P. Burton, M. A. Vaughan, R. A. Ferrare, and C. A. Hostetler, “Separating mixtures of aerosol types in airborne High Spectral Resolution Lidar data,” Atmos. Meas. Tech. 7(2), 419–436 (2014). [CrossRef]

26. V. Freudenthaler, M. Esselborn, M. Wiegner, B. Heese, M. Tesche, A. Ansmann, D. Müller, D. Althausen, M. Wirth, A. Fix, G. Ehret, P. Knippertz, C. Toledano, J. Gasteiger, M. Garhammer, and M. Seefeldner, “Depolarization ratio profiling at several wavelengths in pure Saharan dust during SAMUM 2006,” Tellus 61(1), 165–179 (2009). [CrossRef]

27. T. Nishizawa, H. Okamoto, N. Sugimoto, I. Matsui, A. Shimizu, and K. Aoki, “An algorithm that retrieves aerosol properties from dual wavelength polarized lidar measurements,” J. Geophys. Res. 112(D6), D06212 (2007). [CrossRef]

28. T. Nishizawa, N. Sugimoto, I. Matsui, A. Shimizu, and H. Okamoto, “Algorithms to retrieve optical properties of three component aerosols from two-wavelength backscatter and one-wavelength polarization lidar measurements considering nonsphericity of dust,” J. Quant. Spectrosc. Radiat. Transf. 112(2), 254–267 (2011). [CrossRef]

29. D. Liu, Z. Wang, Z. Liu, D. Winker, and C. Trepte, “A height resolved global view of dust aerosols from the first year CALIPSO lidar measurements,” J. Geophys. Res. 113(D16), D16214 (2008). [CrossRef]

30. D. Liu, Y. Wang, Z. Wang, and J. Zhou, “The three-dimensional structure of transatlantic African dust transport: a new perspective from CALIPSO lidar measurements,” Adv. Meteorol. 2012, 850704 (2012). [CrossRef]

31. Z. Liu, D. Liu, J. Huang, M. Vaughan, I. Uno, N. Sugimoto, C. Kittaka, C. Trepte, Z. Wang, C. Hostetler, and D. Winker, “Airborne dust distributions over the Tibetan Plateau and surrounding areas derived from the first year of CALIPSO lidar observations,” Atmos. Chem. Phys. 8(16), 5045–5060 (2008). [CrossRef]

32. T. Luo, Z. Wang, D. Zhang, X. Liu, Y. Wang, and R. Yuan, “Global dust distribution from improved thin dust layer detection using A-train satellite lidar observations,” Geophys. Res. Lett. 42(2), 620–628 (2015). [CrossRef]

33. V. Amiridis, U. Wandinger, E. Marinou, E. Giannakaki, A. Tsekeri, S. Basart, S. Kazadzis, A. Gkikas, M. Taylor, J. Baldasano, and A. Ansmann, “Optimizing CALIPSO Saharan dust retrievals,” Atmos. Chem. Phys. 13(23), 12089–12106 (2013). [CrossRef]

34. F. G. Fernald, “Analysis of atmospheric lidar observations: some comments,” Appl. Opt. 23(5), 652–653 (1984). [CrossRef] [PubMed]

35. J. W. Hair, C. A. Hostetler, A. L. Cook, D. B. Harper, R. A. Ferrare, T. L. Mack, W. Welch, L. R. Izquierdo, and F. E. Hovis, “Airborne high spectral resolution lidar for profiling aerosol optical properties,” Appl. Opt. 47(36), 6734–6752 (2008). [CrossRef] [PubMed]

36. Z. Liu, D. Winker, A. Omar, M. Vaughan, C. Trepte, Y. Hu, K. Powell, W. Sun, and B. Lin, “Effective lidar ratios of dense dust layers over North Africa derived from the CALIOP measurements,” J. Quant. Spectrosc. Radiat. Transf. 112(2), 204–213 (2011). [CrossRef]

37. U. Wandinger, M. Tesche, P. Seifert, A. Ansmann, D. Müller, and D. Althausen, “Size matters: influence of multiple scattering on CALIPSO light-extinction profiling in desert dust,” Geophys. Res. Lett. 37(10), L10801 (2010). [CrossRef]

38. D. Winker, “Accounting for multiple scattering in retrievals from space lidar,” Proc. SPIE 5059, 128–139 (2003). [CrossRef]

39. T. Luo, R. Yuan, and Z. Wang, “On factors controlling marine boundary layer aerosol optical depth,” J. Geophys. Res. 119(6), 3321–3334 (2014).

Input parameters			SS				SM
Molecular			S_m = 8*π/3 sr, δ _m = 0.0036
Dust			S_d = 40 sr, δ _d = 0.25
Non-Dust			S_{nd_ini} = 25 sr, δ_d = 0.04				S_{nd_ini} = 70 sr, δ_d = 0.04
Mixing situations	Weak		τ_d = 0.05, τ_nd = 0.2				τ_d = 0.05, τ_nd = 0.4
	Moderated		τ_d = 0.15, τ_nd = 0.15
	Full		τ_d varies from 0.01 to 0.5; τ_nd varies from 0.01 to 0.2				τ_d and τ_nd vary from 0.01 to 0.5
Retrievals				SS-SS	SS-SM	SM-SM		SM-SS	Sim1(SS) Sim1(SM)
S_d				40 sr					S = 65 sr
S_{nd_ret}				25 sr	70 sr	70 sr		25 sr
Depolarization ratios		Weak		δ_{d_ret} = 0.25, δ_{nd_ret} = 0.04
		Moderated
		Full		δ_{d_ret} = δ_d - Δ_d, δ_d, and δ_d + Δ_d with δ_d = 0.25 and Δ_d = 0.05; δ_{nd_ret} = δ_nd - Δ_nd, δ_nd and δ_nd + Δ_nd with δ_nd = 0.04 and Δ_nd = 0.01.

	SS-SM	SM-SS	Sim1(SS)	Sim1(SM)
Weak	10%	8%	29%	18%
Moderated	14%	5%	41%	42%

	SS-SS			SS-SM			SM-SM			SM-SS
δ_{nd_ret} δ_{d_ret}	0.03	0.04	0.05	0.03	0.04	0.05	0.03	0.04	0.05	0.03	0.04	0.05
0.2	34%	25%	22%	40%	29%	25%	22%	17%	15%	18%	14%	13%
0.25	8%	0%	8%	20%	11%	12%	6%	0%	6%	8%	4%	13%
0.3	15%	17%	25%	18%	15%	20%	11%	13%	19%	21%	24%	29%
	Sim1(SS)			Sim1(SM)
δ_{nd_ret} δ_{d_ret}	0.03	0.04	0.05	0.03	0.04	0.05
0.2	55%	44%	40%	45%	37%	36%
0.25	29%	23%	24%	18%	16%	19%
0.3	25%	24%	28%	18%	21%	26%

Selected parameter	S_d (sr)		δ_d		S_nd (sr)	δ_nd
Selected value	40	50	0.2	0.3	70	0.06	0.09
Mean difference (sr⁻¹km⁻¹)	−4.9 × 10⁻⁷	−1.4 × 10⁻⁶	1.7 × 10⁻⁴	−1.3 × 10⁻⁴	3.8 × 10⁻⁵	5 × 10⁻⁵	1.2 × 10⁻⁵
Mean absolute relative difference	<1%	<1%	18%	14%	5.4%	7.7%	5.1%

Input parameters			SS				SM
Molecular			S_m = 8*π/3 sr, δ _m = 0.0036
Dust			S_d = 40 sr, δ _d = 0.25
Non-Dust			S_{nd_ini} = 25 sr, δ_d = 0.04				S_{nd_ini} = 70 sr, δ_d = 0.04
Mixing situations	Weak		τ_d = 0.05, τ_nd = 0.2				τ_d = 0.05, τ_nd = 0.4
	Moderated		τ_d = 0.15, τ_nd = 0.15
	Full		τ_d varies from 0.01 to 0.5; τ_nd varies from 0.01 to 0.2				τ_d and τ_nd vary from 0.01 to 0.5
Retrievals				SS-SS	SS-SM	SM-SM		SM-SS	Sim1(SS) Sim1(SM)
S_d				40 sr					S = 65 sr
S_{nd_ret}				25 sr	70 sr	70 sr		25 sr
Depolarization ratios		Weak		δ_{d_ret} = 0.25, δ_{nd_ret} = 0.04
		Moderated
		Full		δ_{d_ret} = δ_d - Δ_d, δ_d, and δ_d + Δ_d with δ_d = 0.25 and Δ_d = 0.05; δ_{nd_ret} = δ_nd - Δ_nd, δ_nd and δ_nd + Δ_nd with δ_nd = 0.04 and Δ_nd = 0.01.

Vertically resolved separation of dust and other aerosol types by a new lidar depolarization method

Abstract

1. Introduction

2. Theory

3. Sensitivity experiments

4. Applications to CALIPSO measurements

5. Conclusions and discussions

Acknowledgments

References and links

Cited By

Figures (5)

Tables (4)

Equations (10)

Optics Express