BACKGROUND: DESCRIPTION OF THE SAHARAN AIR LAYER (SAL).
Heavy concentrations of airborne aerosols within the lower troposphere are responsible for poor air quality at both global and regional scales, which in turn can translate into unhealthy respiratory conditions that impact the population, primarily children and the elderly. In this study, the focus is on recurring pulses of Saharan dust emanating from northern Africa that propagate westward across the Atlantic basin and eventually impact the populace within the greater Caribbean, northern South America, Central America, Gulf of Mexico, and southeastern United States (Dunion and Velden 2004; Evan et al. 2009; Karyampudi and Carlson 1988; Prospero and Carlson 1972). Figure 1 provides an overview of the Saharan air layer (SAL) impact region that is of special concern to the National Weather Service in San Juan, Puerto Rico (NWS-PR). For this paper, other interest areas include the West Indies’ Guadeloupe and Barbados islands; these islands contain ground-based instruments that profile and measure the initial arrival of the dust.
Coverage area for monitoring of the SAL passage into the western tropical Atlantic basin. The inset contains the various sectors (outlined by lines and curves) that comprise the NWS-PR area of responsibility (AOR). Additionally, Guadeloupe and Barbados islands contain in situ instruments that NRL-MMD applies to the SAL-WEB.
Citation: Bulletin of the American Meteorological Society 99, 2; 10.1175/BAMS-D-16-0212.1
From a public health perspective, asthma is a significant problem in Puerto Rico and is greatly exacerbated by increased concentrations of airborne particles associated with Saharan dust (Gyan et al. 2005; Prospero 2012; Prospero and Mayol-Bracero 2013; Prospero and Diaz 2016). Monitoring of regional dust transport from satellite sensors, in situ aerosol measuring instruments, and airborne dust atmospheric modeling has improved dramatically over the past few decades, and improvements toward effective operational forecasting are now possible. In particular, the NWS-PR and the public it serves stand to benefit from these advancements that would result in more accurate and timely warnings of impending hazardous air quality.
Figure 2 depicts a three-dimensional perspective of the SAL’s structure and flow pattern. The SAL is a year-round phenomenon with the potential for maintaining its atmospheric sounding characteristics (i.e., relatively hot, dry, elevated mixed layer, strong easterly jet, and decoupled from its surroundings) for several thousand kilometers downwind (Dunion 2011). The SAL has been proposed as a factor in mitigating tropical storm cyclogenesis by cooling the sea surface temperature, suppressing low-level convection, and enhancing midlevel vertical shear (Dunion and Velden 2004; Evan and Mukhopadhyay 2010). There is a growing field of research related to the association of the SAL with possibly dangerous environmental conditions for populated islands particularly within the greater Caribbean region. The SAL air mass can supply the necessary atmospheric conditions for wildfire outbreaks, but more importantly degrade air quality, worsening human respiratory conditions (Akinbami and Moorman 2011; Lara et al. 2006). For this article, we focus on SAL air masses containing heavy concentrations of dust that are eventually transported to the greater Caribbean.
Three-dimensional conceptual view of the Saharan air layer (SAL) with the perspective of looking westward from its source in North Africa. The flow pattern is westward toward the Caribbean and is associated with an easterly wave; part of the SAL also turns anticyclonically northward toward the mid-Atlantic basin. The SAL forms within the convection over the hot desert and semiarid terrain in North Africa, extending from the surface to the 500-hPa level. The SAL is bounded on the south by the intertropical convergence zone (ITCZ) with an associated midlevel easterly jet (MLEJ, red arrow). The SAL development and transport often occurs behind easterly waves. [Figure and portions of caption courtesy of Karyampudi and Carlson (1988).]
Citation: Bulletin of the American Meteorological Society 99, 2; 10.1175/BAMS-D-16-0212.1
Figure 3 displays a generalized two-dimensional isentropic depiction of SAL airmass transport. The nature of SAL initiation and strengthening is highly dependent upon terrain, as well as climatological, meteorological, and anthropogenic dynamics (Knippertz and Todd 2012; Prospero and Mayol-Bracero 2013). Viewing from right to left, the first panel depicts the initialization of the SAL over northeast Africa (Sahel and Sahara regions during tinitial) with strong intense desert heating at the surface that generates strong sensible heat, associated turbulent flux, and convection at the surface. Surface dust is often scoured by surface winds, and then lifted upward to heights reaching 500 hPa. As the SAL eventually propagates westward across northwest Africa and into the Atlantic basin, the SAL air mass follows the isentropic contours. Its base becomes decoupled from the surface; just offshore, the vertical layer extends between 850 and 500 hPa. The leading portion of the SAL layer typically takes 6–7 days to travel from the African coast to Barbados, where the water vapor mixing ratio is conserved at very dry values (∼5%–10%) and the descent rate of the layer is estimated at 7 hPa (day)−1. The associated dust throughout the SAL airstream is maintained and well mixed in the vertical as a result of weak turbulent mixing. By the time the leading edge of the SAL reaches the Caribbean (t+7 days), the lower portion of the SAL penetrates well into the marine boundary layer. At this time, much of the dust is impacting the surface.
Vertical profile of the SAL air mass being transported via convection and turbulent mixing from its hot desert source (right side: Sahel/Saharan region) westward to the NW Africa coast, across the north tropical Atlantic basin, and finally through the Caribbean islands. The top portion describes the typical transit time of approximately 7 days. The color shading within the SAL layer represents the transition from coarse and large dust particles (red shades) to finer and more diffuse particles farther west (yellow shades). The vertical brown curved arrows depict larger dust particles settling to the surface. Isentropic contours are annotated in blue, with associated theta labels. The marine boundary layer is shown sloping upward from east to west. Cumulus clouds are prevalent throughout the maritime tropical Atlantic basin and scavenge aerosol particles from the SAL layer.
Citation: Bulletin of the American Meteorological Society 99, 2; 10.1175/BAMS-D-16-0212.1
Easterly trade winds transport the elevated SAL air mass across the tropical Atlantic basin (tinitial to t+4 days), where its statically stable warm and dry air mass is in sharp contrast to the cooler and moist marine layer. The SAL air mass and associated dust retain their isentropic characteristics. Eventually, the larger aerosols settle to the surface as depicted by the arrows in Fig. 3. By time t+7 days, the SAL air mass has traveled several thousand kilometers upon reaching the greater Caribbean region. Additionally, the airmass structure and composition have changed: the top of the SAL air mass tends to decrease very gradually, while the base becomes ill-defined and blended with the marine boundary layer. During this time, the dust in the layer is dominated by smaller particles. Formation of clouds and associated precipitation can significantly alter the SAL layer, most importantly by scavenging dust along its path (Engelstaedter et al. 2009). There is more interaction of particles within the SAL with the surface, which, from a health perspective, increases impacts over the Caribbean islands.
ROLE OF THE NAVAL RESEARCH LABORATORY MARINE METEOROLOGY DIVISION.
The U.S. Naval Research Laboratory Marine Meteorology Division (NRL-MMD), located in Monterey, California, is sponsored by the National Oceanic and Atmospheric Administration (NOAA) to conduct this research. Part of the focus is studying SAL environmental and dynamical characteristics. The primary goal of this project is supporting NWS-PR with state-of-the-art decision aids that would supplement the Advanced Weather Interactive Processing System (AWIPS-II) product suite in monitoring and predicting SAL events over their area of responsibility (AOR) (see Fig. 1). NRL-MMD is well-equipped to provide a variety of satellite-derived products for a variety of applications, including the monitoring of the SAL events as they approach the Caribbean. NRL-MMD is also training forecasters on the use (or application) of NRL-MMD products that are not yet integrated into the AWIPS-II product network. Additionally, the remote sensing and modeling staff within NRL-MMD provide dust modeling to help assess SAL characteristics and predict its arrival over the entire Caribbean domain, particularly the West Indies and Puerto Rico. In addition to NOAA, NRL-MMD also collaborates with the National Aeronautics and Space Administration (NASA), the Cooperative Institute for Meteorological Satellite Studies (CIMSS), and the Cooperative Institute for Research in the Atmosphere (CIRA) in receiving and providing datasets related to SAL monitoring.
Besides supporting the NWS-PR, NRL-MMD collaborates in SAL research and development with the Caribbean Institute for Meteorology and Hydrology (CIMH; http://cimh.edu.bb/) and the Caribbean Aerosol-Health Network (CAHN). CIMH, located in Barbados, is a training, research, and investigative organization whose mission is to provide general meteorological and hydrological resources to 16 countries based throughout the greater Caribbean. CAHN is a consortium of international agencies that are actively working to understand the impacts of the SAL on the air quality, health, climate, weather, and ecosystems in the Caribbean region. Their future plans include global expansion into regions that often suffer from health-related ailments associated with high concentrations of harmful pollutants.
NRL-MMD SAL WEBSITE (SAL-WEB).
Figure 4 is an illustration of the NRL-MMD SAL website (SAL-WEB; www.nrlmry.navy.mil/SAL.html), a publicly accessible global resource hosting a variety of products that serve as a repository for SAL monitoring and research. Table 1 lists the suite of products within SAL-WEB that are customized to monitor the SAL.
The NRL-MMD SAL website. (top) Home page that covers the entire tropical Atlantic region of the SAL project. (bottom) Subdomains that focus on regions of interest of the SAL project—namely the Caribbean, Puerto Rico, and West Indies.
Citation: Bulletin of the American Meteorological Society 99, 2; 10.1175/BAMS-D-16-0212.1
Suite of products provided by the NRL-MMD SAL website (SAL-WEB). Within the environmental products section is DEBRA, the Dynamic Enhanced Background Reduction Algorithm, which represents a novel approach toward viewing elevated dust.
Currently, SAL-WEB contains imagery, model output, and surface-based instruments with a focus on monitoring SAL events. The overarching goal is to provide as many forecasting tools as possible toward characterizing and predicting the impact of the SAL to the forecaster. Figure 4 describes the various domains available, including the entire north tropical Atlantic basin, along with the encompassed subdomains within the Caribbean. The user-friendly characteristics of this website consist of near-real-time product accessibility, which is an important component in daily operations. The website also features feedback options where developers are quick to respond to user needs or address technical problems. Currently, NRL-MMD and NWS-PR conduct ongoing communications to help in the development and refinement of products and solidify strategic planning toward more efficient operational applications.
SENSORS AND MEASURING INSTRUMENTS TO MONITOR THE SAL.
Remote sensing observations.
SAL-WEB includes a variety of tools to help the forecaster in both monitoring and predicting SAL events. The primary tools are satellite-derived image products that provide both a “quick look” of current environmental events as well as tracking SAL features. Satellite imagery is one of the most feasible resources for monitoring dust outbreaks since dust provides an easily identifiable tracer. The total precipitable water (TPW; http://tropic.ssec.wisc.edu/real-time/mimic-tpw/prodDesc/) product also allows users to track an equally important aspect of SAL, that of elevated dry layers. At ground level, numerous instruments can provide onsite depictions of dust and thermodynamics related to the passing of SAL events. SAL-WEB hosts the NASA Micropulse Lidar Network (MPLNET) located in Barbados (Campbell et al. 2015; Campbell et al. 2016a; Campbell et al. 2016b), which provides continuous vertical profiles of aerosol content.
In situ observations.
NASA’s Aerosol Robotic Network (AERONET; Holben et al. 1998) stations situated about the eastern Caribbean regions (Puerto Rico, West Indies islands) provide a validation of aerosol measurements to be compared against satellite-derived and model-derived products. The Tapered Element Oscillating Microbalance (TEOM) sites provide an assessment of surface-based aerosol extinction measurements, which are not only an important source for validation but also provide the public with much-needed resources in preparing for dust-related respiratory ailments that plague primarily the very young and elderly. SAL-WEB does not directly host AERONET or TEOM instruments because of their easily accessible online resources. By far, the most definitive measurement of SAL’s depth and thermodynamic aspects are sounding profiles. Carlson (2016) conducted a 5-yr study that focused on the SAL structure and dynamics using atmospheric soundings throughout the path from Africa to Miami, Florida. Additionally, Dunion (2011) conducted an 8-yr sounding study using Caribbean rawinsonde data to compare SAL to the moist tropical air masses associated with the Jordan mean sounding profile.
DUST MODELING OF SAL EVENTS.
The above-mentioned SAL monitoring instruments are crucial to monitoring SAL characteristics from a near-real-time perspective, but provide a somewhat limited perspective. The responsibility of the NWS-PR is to assess and predict the onset, strength, and arrival of an SAL event over Puerto Rico, the U.S. Virgin Islands, and surrounding marine areas. To this end, numerical weather prediction models specializing in dust transport provide the essential guidance to alert the populace of hazardous air quality conditions in a timely manner.
Dust forecast modeling has faced great challenges across oceanic basins, such as the tropical Atlantic basin, mainly because of sparse in situ instruments to constrain initial conditions. There are a number of dust models applied to assess and predict dust conditions over the tropical Atlantic; SAL-WEB hosts the Navy Aerosol Analysis and Prediction System (NAAPS) model output for assessment of SAL conditions. NAAPS (Lynch et al. 2016; Westphal et al. 1988) has been run quasi-operationally at NRL-MMD since 1998, and became the world’s first operational global aerosol model in 2006 with implementation at the Fleet Numerical Meteorology and Oceanography Center (FNMOC). The Navy Atmospheric Variational Data Assimilation System for aerosol optical thickness (NAVDAS-AOT; Zhang et al. 2008) was operationally implemented in 2010. The system assimilates quality-assured and quality-controlled (QA/QC) two-dimensional Moderate Resolution Imaging Spectroradiometer (MODIS) AOT at 550 nm (Zhang and Reid 2006; Hyer et al. 2011) by default, but it can also assimilate other satellite aerosol optical depth (AOD) products after QA/QC, including Visible Infrared Imaging Radiometer Suite (VIIRS) AOT, for example.
In its current operational configuration, NAAPS provides 6-day forecasts four times a day at ⅓° spatial resolution and 42 vertical levels driven by Navy Global Environmental Model (NAVGEM) meteorology (Hogan et al. 2014). NAAPS characterizes anthropogenic and biogenic fine (ABF), dust, smoke, and sea salt aerosols. The NAAPS also serves as a research and development resource. Details describing the development and validation of the research version of NAAPS and the AOD reanalysis product can be found in Lynch et al. (2016). Updates to the operational model can be found at the NRL aerosol website (www.nrlmry.navy.mil/aerosol/).
From the research and development perspective, scientists are constantly providing upgrades in the NAAPS data assimilation (DA) and dynamic processing; in this case, NAAPS is one of the first dust models to implement the highly anticipated capabilities of the VIIRS AOD for forecast initialization. Preliminary results show that this new application (scheduled to become operational during early 2018) will yield significant improvement in overall dust prediction and assessments. Additional information on the NAAPS AOD plots can be found at SAL-WEB. NAAPS is also part of the International Cooperative for Aerosol Prediction (ICAP) suite of operational global aerosol models, whose overarching mission is to develop a global multimodel aerosol forecasting ensemble intended for eventual operational and basic research use (Sessions et al. 2015).
One of the most important aspects of this project is providing the NWS-PR with improvements in their forecasting skill through better dust model prediction across their AOR. NRL-MMD is in the process of validating the performance of NAAPS by incorporating VIIRS AOD into the NAAPS DA process. Currently, only the MODIS AOD is applied for operations. The following two sections will describe both the qualitative and quantitative aspects of this effort.
Qualitative comparison of Operational NAAPS with the addition of VIIRS AOD: Case study during 23–28 June 2014.
A large SAL outbreak occurred during 23–28 June 2014, as illustrated in Fig. 5. The leading edge of the SAL, annotated in yellow, shows the SAL covering the eastern Atlantic on 23 June, approaching Barbados on 24 June, and then reaching Puerto Rico farther west on 25 June. The sun glint patterns in bright northwest (NW)–southeast (SE)-oriented linear gray shades often obscure the dust within the SAL, but the SAL is still easily distinguishable from this feature. The regions of interest are Puerto Rico (PR) and Barbados (BB). This case represents a strong SAL event that eventually moves into the Gulf of Mexico region.
VIIRS-derived true color products providing a daily sequence (23–28 Jun 2014) of the SAL event propagating from NW Africa westward to the greater Caribbean. The bold dashed yellow arcs depict the leading edge of the SAL; PR and BB denote the positions of Puerto Rico and Barbados, respectively. For each panel, the linear NNW–SSE-oriented features of enhanced radiances across the open water represent sun glint.
Citation: Bulletin of the American Meteorological Society 99, 2; 10.1175/BAMS-D-16-0212.1
Figure 6 compares the NAAPS dust model (AOD) output between the current operational version that uses MODIS aerosol optical thickness (AOT) for DA in the top-left panel and the addition of VIIRS AOT in the top-right panel. The AERONET plot during this time period (bottom panel) provides validation. For this case, the addition of VIIRS into NAAPS shows an improvement in the measured AOD over Puerto Rico at ∼1800 UTC, where the AOD values ranging from 0.6 to 0.8 (500-nm AOT wavelength) correspond well with the La Parguera, Puerto Rico, AERONET site in the bottom panel. Contrast this to the operational NAAPS, where the contours about Puerto Rico (0.2–0.3) range significantly lower than the AERONET measurements during this time. Although it is only one case, this example reveals the significance of incorporating VIIRS into the NAAPS model.
Comparing NAAPS with the data assimilated (top left) MODIS AOD vs (top right) NAAPS MODIS + VIIRS AOD on 26 Jun 2014. Annotated PR represents Puerto Rico location. (bottom) AERONET AOD plot over La Parguera in SW Puerto Rico. Vertical red dashed line indicates the corresponding time (1800 UTC) with the upper-model outputs.
Citation: Bulletin of the American Meteorological Society 99, 2; 10.1175/BAMS-D-16-0212.1
Quantitative assessment of Operational NAAPS with the addition of VIIRS AOD.
NRL-MMD is also proceeding to incorporate VIIRS AOD with the existing MODIS AOD data into the NAAPS DA. To quantitatively assess the impact of Suomi National Polar-orbiting Partnership VIIRS data on operational aerosol predictions, NRL obtained a 3-month dataset for May–July 2015 of the NOAA Enterprise Aerosol product (NOAA Enterprise) (Zhang et al. 2016). Since July 2017, NOAA has been producing the Enterprise Aerosol product operationally as the replacement for the at-launch VIIRS aerosol product currently provided by the Interface Data Processing System (IDPS) (Jackson et al. 2013; Huang et al. 2016). The NOAA Enterprise product incorporates significant improvements to the over-ocean aerosol retrieval, and a novel method for retrieval of aerosol properties over land. Comparison of a 2-yr dataset from NOAA Enterprise to sun photometer measurements of AOD from AERONET (Holben et al. 1998) showed substantial improvement over the original IDPS product (Zhang et al. 2016).
NRL-MMD also processed and compared NAAPS AOD output with the combination of MODIS and the Enterprise VIIRS AOD datasets. This study applied the most selective quality assurance filtering (QA = high), and aggregated the 750-m retrievals to 1° bins, producing a set of gridded observations for every 6 h of input aerosol data. NAAPS (Lynch et al. 2016) was then used to generate AOD analysis for two separate runs:
Assimilation of MODIS AOD only. This run is most similar to the operational NAAPS model, except that the operational NAAPS runs at a resolution of ⅓° and this experimental run used a resolution of 1°.
Assimilation of MODIS + VIIRS. Gridded MODIS and VIIRS observations were combined, with both observations assimilated in cases where both satellites reported AOD.
Figure 7 shows the results using a scatterplot and a corresponding graphical mapping for a more in-depth view. NAAPS AOD output was compared to AERONET level 2.0 data for a 6-week period from 15 June to 31 July 2015. For the scatterplot panel on the right, the global rmse for the VIIRS + MODIS simulation was 0.11, compared with 0.12 for the MODIS-only simulation. The global correlation between NAAPS and AERONET was R2 = 0.74 for VIIRS + MODIS, compared with R2 = 0.68 for MODIS only. Results were also calculated for each individual AERONET station, as shown in Fig. 7. Each point on the map and scatterplot represents a single AERONET station. Symbols on the map are sized proportional to the correlation coefficient R2 between NAAPS and AERONET data at each site. From the map and scatterplot it is clear that VIIRS + MODIS (blue symbols) has greater correlation to AERONET compared with MODIS only (red symbols) over much of the globe, especially North America. Correlation coefficient R2 was greater for VIIRS + MODIS than MODIS only at 132 of the 208 AERONET stations with at least 100 level 2.0 AOD retrievals during the evaluation period.
NAAPS testing of MODIS-only AOD compared to MODIS + VIIRS Enterprise AOD over the global span of AERONET sites. Results cover a 3-month period from May to Jul 2015. (left) The graphical map displays qualitative correlations over each AERONET site (marked by small gray x’s). Size (diameter) of circles is proportional to the values of R2. Gray circles indicate little difference between both NAAPS output types. Color circles indicate significant departures between both NAAPS sets. Outer red rings (as seen in most locations) indicate that the addition of Enterprise VIIRS with MODIS AOD into NAAPS outperforms MODIS-only AOD. Outer blue rings (such as in southern Africa) indicate that MODIS AOD into NAAPS processing outperforms VIIRS + MODIS AOD. (right) The scatterplot provides a quantitative perspective of the correlations.
Citation: Bulletin of the American Meteorological Society 99, 2; 10.1175/BAMS-D-16-0212.1
To summarize this section, the addition of VIIRS via NOAA Enterprise into the NAAPS DA showed clear benefits in AOD analysis accuracy, which should translate to improved medium-range forecasting of aerosol events. Even better results can likely be achieved with additional filtering. Operational assimilation of NOAA Enterprise in NAAPS is planned during early 2018.
SUMMARY.
The persistent occurrence of African dust over Puerto Rico and surrounding Caribbean islands has become a prominent focus area of research due to asthma and other significant public health problems. From the NWS-PR perspective, monitoring dust-laden SAL events in the Caribbean is essential in alerting the public about poor air quality conditions for a specific period. A precise forecast will provide sufficient lead time that allows state agencies and the general public to take the necessary actions to minimize the health impacts associated with SAL events. Aircraft and maritime planning operations will also benefit from an accurate prediction of the magnitude of the SAL event. Using SAL-WEB, NRL-MMD is supplementing the resources currently available on AWIPS-II for NWS-PR and expanding the forecasting capabilities of CIHM. Currently, SAL-WEB contains products from 11 satellite sensors, 3 dust models, ground-based lidar, and an assortment of TEOM instruments. A focal research aspect is improving the NAAPS dust model through the data assimilation of the VIIRS AOD. Preliminary results indicate improvements in prediction and 3D characteristics of approaching SAL events, including the atmospheric parameters related to thermodynamic assessment and predictability.
This project was motivated in part by discussions at a symposium on airborne dust and its impacts on human health on 19–21 May 2015 in Miami (Prospero and Diaz 2016) to discuss how both scientific and health communities can better combine and coordinate efforts in studying SAL impacts on human health. The overarching goal is to better educate and prepare the Caribbean populace in mitigating exposure to SAL’s harmful dusty environment. More recently, another workshop was organized by CAHN in San Juan on 24–25 April 2016 to improve collaboration and communication between the Caribbean countries.
Current and near-term plans for NRL-MMD include ongoing interactions with NWS-PR, CIMH, and CAHN agencies to share environmental resources, with the goal of hosting a more comprehensive set of SAL-related sensing products into SAL-WEB. The website is publicly accessible; public feedback is encouraged via the SAL-WEB (www.nrlmry.navy.mil/SAL.html) feedback tab.
ACKNOWLEDGMENTS
The work was supported by the NOAA/NESDIS Joint Polar Satellite System (JPSS) Office and the Center for Satellite Applications and Research (STAR) under NA13AANAG0343. The authors are also greatly appreciative to both Toby Carlson (professor emeritus at The Pennsylvania State University) and Ernesto Rodriguez (forecaster at the NWS-PR) for their in-depth and insightful reviews of this article.
FOR FURTHER READING
Akinbami, L., and J. E. Moorman, 2011: Asthma prevalence, health care use, and mortality: United States, 2005–2009. Natl. Health Statistics Rep. 32, 15 pp.
Campbell, J. R., M. A. Vaughan, M. Oo, R. E. Holz, J. R. Lewis, and E. J. Welton, 2015: Distinguishing cirrus cloud presence in autonomous lidar measurements. Atmos. Meas. Tech., 8, 435–449, https://doi.org/10.5194/amt-8-435-2015.
Campbell, J. R., and Coauthors, 2016a: Applying advanced ground-based remote sensing in the Southeast Asian Maritime Continent to characterize regional proficiencies in smoke transport modeling. J. Appl. Meteor. Climatol., 55, 3–22, https://doi.org/10.1175/JAMC-D-15-0083.1.
Campbell, J. R., S. Lolli, J. R. Lewis, Y. Gu, and E. J. Welton, 2016b: Daytime cirrus cloud top-of-the-atmosphere radiative forcing properties at a midlatitude site and their global consequences. J. Appl. Meteor. Climatol., 55, 1667–1679, https://doi.org/10.1175/JAMC-D-15-0217.1.
Carlson, T. N., 2016: The Saharan elevated mixed layer and its aerosol optical depth. Open Atmos. Sci. J., 10, 26–38, https://doi.org/10.2174/1874282301610010026.
Dunion, J. P., 2011: Rewriting the climatology of the tropical north Altlantic and Caribbean sea atmosphere. J. Climate, 24, 893–908, https://doi.org/10.1175/2010JCLI3496.1.
Dunion, J. P., and C. S. Velden, 2004: The impact of the Saharan Air Layer on Atlantic tropical cyclone activity. Bull. Amer. Meteor. Soc., 85, 353–365, https://doi.org/10.1175/BAMS-85-3-353.
Engelstaedter, S., R. Washington, and N. Mahowald, 2009: Impact of changes in atmospheric conditions in modulating summer dust concentration at Barbados: A back-trajectory analysis. J. Geophys. Res., 114, https://doi.org/10.1029/2008JD011180.
Evan, A. T., and S. Mukhopadhyay, 2010: African dust over the northern tropical Atlantic: 1955–2008. J. Appl. Meteor. Climatol., 49, 2213–2229, https://doi.org/10.1175/2010JAMC2485.1.
Evan, A. T., D. J. Vimont, A. K. Heidinger, J. P. Kossin, and R. Bennartz, 2009: The role of aerosols in the evolution of tropical North Atlantic Ocean temperature anomalies. Science, 324, 778–781, https://doi.org/10.1126/science.1167404.
Gyan, K., and Coauthors, 2005: African dust clouds are associated with increased paediatric asthma accident and emergency admissions on the Caribbean island of Trinidad. Int. J. Biometeorol., 49, 371–376, https://doi.org/10.1007/s00484-005-0257-3.
Hogan, T., and Coauthors, 2014: The Navy Global Environmental Model. Oceanogr., 27, 116–125, https://doi.org/10.5670/oceanog.2014.73.
Holben, B. N., and Coauthors, 1998: AERONET—A federated instrument network and data archive for aerosol characterization. Remote Sens. Environ., 66, 1–16, https://doi.org/10.1016/S0034-4257(98)00031-5.
Huang, J., and Coauthors, 2016: Validation and expected error estimation of Suomi-NPP VIIRS aerosol optical thickness and Ångström exponent with AERONET. J. Geophys. Res., 121, 7139–7160, https://doi.org/10.1002/2016JD024834.
Hyer, E. J., J. S. Reid, and J. Zhang, 2011: An over-land aerosol optical depth data set for data assimilation by filtering, correction, and aggregation of MODIS Collection 5 optical depth retrievals. Atmos. Meas. Tech., 4, 379–408, https://doi.org/10.5194/amt-4-379-2011.
Jackson, J. M., H. Q. Liu, I. Laszlo, S. Kondragunta, L. A. Remer, J. F. Huang, and H. C. Huang, 2013: Suomi-NPP VIIRS aerosol algorithms and data products. J. Geophys. Res., 118, 12 673–12 689, https://doi.org/10.1002/2013JD020449.
Karyampudi, V. M., and T. N. Carlson, 1988: Analysis and numerical simulations of the Saharan air layer and its effect on easterly wave disturbances. J. Atmos. Sci., 45, 3102–3136, https://doi.org/10.1175/1520-0469(1988)045<3102:AANSOT>2.0.CO;2.
Knippertz, P., and M. C. Todd, 2012: Mineral dust aerosols over the Sahara: Meteorological controls on emission and transport and implications for modeling. Rev. Geophys., 50, https://doi.org/10.1029/2011RG000362.
Lara, M., L. Akinbami, G. Flores, and H. Morgenstern, 2006: Heterogeneity of childhood asthma among Hispanic children: Puerto Rican children bear a disproportionate burden. Pediatrics, 117, 43–53, https://doi.org/10.1542/peds.2004-1714.
Lynch, P., and Coauthors, 2016: An 11-year global gridded aerosol optical thickness reanalysis (v1.0) for atmospheric and climate sciences. Geosci. Model Dev., 9, 1489–1522, https://doi.org/10.5194/gmd-9-1489-2016.
Prospero, J. M., 1999: Assessing the impact of advected African dust on air quality and health in the eastern United States. Hum. Ecol. Risk Assess. Int. J., 5, 471–479, https://doi.org/10.1080/10807039.1999.10518872.
Prospero, J. M., and T. N. Carlson, 1972: Vertical and areal distribution of Saharan dust over the western equatorial north Atlantic Ocean. J. Geophys. Res., 77, 5255–5265, https://doi.org/10.1029/JC077i027p05255.
Prospero, J. M., and H. Diaz, 2016: The impact of African dust on air quality in the Caribbean basin. Eos, Trans. Amer. Geophys. Union, 97, https://doi.org/10.1029/2016EO043831.
Prospero, J. M., and O. L. Mayol-Bracero, 2013: Understanding the transport and impact of African dust on the Caribbean basin. Bull. Amer. Meteor. Soc., 94, 1329–1337, https://doi.org/10.1175/BAMS-D-12-00142.1.
Sessions, W. R., and Coauthors, 2015: Development towards a global operational aerosol consensus: Basic climatological characteristics of the International Cooperative for Aerosol Prediction Multi-Model Ensemble (ICAP-MME). Atmos. Chem. Phys., 15, 335–362, https://doi.org/10.5194/acp-15-335-2015.
Westphal, D. L., O. B. Toon, and T. N. Carlson, 1988: A case study of mobilization and transport of Saharan dust. J. Atmos. Sci., 45, 2145–2175, https://doi.org/10.1175/1520-0469(1988)045<2145:ACSOMA>2.0.CO;2.
Zhang, H., S. Kondragunta, I. Laszlo, H. Liu, L. A. Remer, J. Huang, S. Superczynski, and P. Ciren, 2016: An enhanced VIIRS aerosol optical thickness (AOT) retrieval algorithm over land using a global surface reflectance ratio database. J. Geophys. Res., 121, 10 717–10 738, https://doi.org/10.1002/2016JD024859.
Zhang, J., and J. S. Reid, 2006: MODIS aerosol product analysis for data assimilation: Assessment of over-ocean level 2 aerosol optical thickness retrievals. J. Geophys. Res., 111, D22207, https://doi.org/10.1029/2005JD006898.
Zhang, J., J. S. Reid, D. L. Westphal, N. L. Baker, and E. J. Hyer, 2008: A system for operational aerosol optical depth data assimilation over global oceans. J. Geophys. Res., 113, D10208, https://doi.org/10.1029/2007JD009065.
