Chapter 6

Description of the Individual Platforms

The required parameters (Table 3 and Table 4) will be estimated from a variety of instruments and platforms. The instruments under each platform are shown in Table 6a, Table 6b, Table 7a, Table 7b, Table 8a, Table 8b, and Table 9.

The composite cruise and flight tracks are shown in Figure 18a with Malé as the base of flight operations and in Figure 18b with Seychelles as the base. The figures show cruise tracks by the 2 research vessels (the Sonne and the Sagar Kanya) and flight tracks for the 5 aircraft (4 US and 1 Dutch). The flight mission types (described next) are shown in Figure 19.

Figure 18a. INDOEX IFP with Malé as the Base of Flight Operations. Figure 18b. INDOEX IFP with Seychelles as the Base of Flight Operations.

Figure 19a. Aerosol - Chemistry - Radiation Gradient Flights

Figure 19b. Upper Level Aerosol - Chemistry - Radiation Profiling Flights

Figure 19c. Lower Level Aerosol - Chemistry - Radiation Profiling Flights

The two ships would traverse all 5 regions of INDOEX (Figure 13) in the north-south direction between February 15 to April 1. In addition, the Sagar Kanya will sample the air along the Indian coast in the Arabian sea. The ship tracks will be designed to facilitate overflights to permit instrument comparisons and radiation column measurements.

The composite flight profiles and cruise tracks are based on an examination of diurnal, daily and weekly evolution of cloud patterns as gleaned from INSAT and ERBE data. Figures 20a and 20b show a sample of the diurnal and week-to-week variations of thermal radiation from clouds obtained from 1989 INSAT pixel level radiance data.

Jan 25, 1989
Jan 15-Feb. 25, 1989
Figure 20a. Thermal emission from clouds at 3-hour intervals during Jan. 25, 1989. The thermal radiation is expressed as equivalent black-body temperature (°K). The zonal band of convection located south of the equator persists throughout the diurnal cycle. Clouds with higher thermal emission in the Arabian Sea and Bay of Bengal (cirrus or mid-level clouds) do not change location. The thermal emission increases from these regions during the day, presumably as the cirrus clouds begin to dissipate. Panel a1 shows 0300 GMT, panel a2 shows 0600 GMT, panel a3 shows 0900 GMT, panel a4 shows 1200GMT, panel a5 shows 1500 GMT, and panel a6 shows 1800 GMT. Figure 20b. Weekly-averaged thermal emission from clouds over 6 weeks. The thermal radiation is expressed as equivalent black-body temperature (°K). A zonal band of convection located south of the equator is present throughout the 6-week period. The Arabian Sea and Bay of Bengal are characterized by high thermal emission (characteristic of clear skies or low clouds) during the latter 4 weeks. Panel b1 shows Jan. 15-21, panel b2 shows Jan. 22-29, panel b3 show Jan. 29-Feb. 4, panel b4 shows Feb. 5-11, panel b5 shows Feb 12-18, and panel b6 shows Feb. 19-25.

Details of individual flight tracks must be, for necessity, determined on the basis of safety considerations and specific day-to-day meteorology. Every effort will be made to overfly ship locations and coordinate with satellite overpasses, and whenever possible to stack three or four of the aircraft with the objective to maximize the benefits of the comparison of instrument calibrations and other characteristics.

6.1. Aircraft

(F.P.J. Valero, A. Clarke, A. Tuck, A. Heymsfield, J. Ogren, F. Arnold, M. Andreae)

Aircraft range and location of operations must be selected such that the ITCZ and beyond (further to the south by about 5š) can be reached (Figure 18), as well as locations over the Arabian Sea.

The mid- and high-level aircraft should have comparable speeds, to allow examination of cloud shortwave absorption and indirect aerosol radiative forcing. This also applies to the lower level aircraft. Performance specifications for the proposed aircraft appear in Table 10.

Table 10. Aircraft Performance

ER-2 WB-57 C-130 P-3 Citation II
True Air Speed, knots 400 400 290 320 400
Range of Operation, nmi 2900 2300 3000 3000 1350
Ceiling, km 21 20 8 8 12.5
Payload, kg 1180* 2000 5900 1000 1300

* plus 1550kg on wing pods

Flight Mission Types

Radiation - Aerosol - Chemistry Gradient/Profiling Flights

The flight patterns which satisfy the objectives, except those related to aerosol events over the Arabian Sea (described next), include a combination of stacked constant level flights, as shown schematically in Figure 19a (gradient flights), and in-and-out of cloud flights shown in Figure 19b and 19c (profiling flights). On outbound legs (gradient flights), the two low-level aircraft, the C-130 and P-3, will be stacked with one above the boundary layer inversion and the other close to the ocean surface (Figure 19a). This will allow measurements of the north-south (or east-west) gradients in vertically integrated (between the 2 aircraft altitudes) spectral optical depth, spectral and total fluxes, transmission, reflectance, absorption of solar radiation by the aerosol layer and clouds, and air chemistry. On the return legs (profiling flights), the upper level aircraft will profile the aerosol layers and sample clouds where present, while the lower level aircraft will maintain its position close to the surface.

For the mid- and high- level aircraft, we plan to follow similar flight patterns (Figures 19a and 19b) to measure the direct and indirect effects of aerosols and air chemistry, as well as cloud and clear sky solar absorption and microphysics. Dropsondes will be released from the WB-57. While the majority of the flights will cross the ITCZ, some will be flown over the Arabian Sea. These will again be paired missions with the aircraft stacked similar to the configuration shown for the outbound legs across the ITCZ (Figures 19a and 19b). A main difference will be that the upper aircraft of the low-level pair will fly as high as possible.

With respect to chemistry, the Citation will profile chemistry in selected regions, and the WB-57 and C-130 will establish aerosol/gas properties related to air mass types. The C-130 will profile and characterize lower inflow into the ITCZ while the WB-57 characterizes upper level outflow. Gradient flights will be oriented to complement the ship tracks. An example of the aircraft vertical stacking is shown in Figure 19a. Individual gradient flights will be linked to associated satellite and ship data in order to interpret gradients for the period under consideration.

Radiation - Aerosol Closure Flights

Figure 21. Diagrams of Radiative Closure Flights.
These flights will emphasize vertical profiles in the lower troposphere that link aerosol physical and chemical properties to their optical influence on radiative transfer (see C-130 section below). A primary target will be the continental outflow of aerosol into the INDOEX region, and flights will include profiles (Figure 21) in high, medium and low concentration regions within the aerosol gradient. This will allow determination of the range of radiative influence of continental emissions advected over marine regions and the key changes in characteristics as the "continental" aerosol column is transformed into a "clean" marine aerosol. The separate contributions of key continental aerosols (sulfate, soot, dusts) will be isolated and linked to changes in satellite radiances and optical depth.

The approach will follow suggested procedures [Penner et al., 1993] and expand upon the approach successfully employed on the 1992 Atlantic Stratocumulus Transition Experiment (ASTEX) [Clarke et al., 1995] and recently completed 1995 Southern Hemisphere Marine Aerosol Characterization Experiment (ACE-1).

A typical radiative closure vertical profile will take on the order of 1.5 - 2.5 hours depending upon the structure and variety in the aerosol levels that are encountered.

In order to best quantify radiative forcing and test and validate satellite aerosol retrieval algorithms, radiative closure flights would ideally be centered on a satellite overpass and repeated in regions of varying optical depth and with different column concentrations of sulfate, dust etc. With the constraint of limited flight hours and the large distances and ferry times expected for these missions, it is important to link the gradient flights and radiative profiles in a way that maximizes objectives.

Indirect Aerosol - Radiative Forcing

The sampling of low level clouds during the radiation-aerosol gradient/profiling flights will be used to explore the response of cloud albedo to enhanced cloud droplets, CCN and continental aerosol (e.g., sulfate). Legs will be flown below, in and above clouds where conditions are favorable as part of both "gradient" and "closure" transect flights. While the C-130 with its aerosols and radiation instrumentation flies legs below, inside, and just above clouds, the WB-57 and/or the ER-2 will map the radiation fields above clouds. Flights will emphasize characterization of cloud ensembles and contrast their microphysical and radiative properties between regions of marked differences in aerosol properties. In addition, dedicated flights will be carried out when optimum conditions of homogeneous stratus or stratocumulus are present across marked gradients in aerosol fields.

Ferry Flight Science

There is great potential value in using the flying time necessary for the aircraft to get to and return from the Indian Ocean, which would otherwise be wasted from a scientific point of view. Figure 9 shows the possible routes for air from the lower stratosphere to enter the upper tropical troposphere, while Figure 10 shows recent ER-2 evidence that it indeed does, with negative CFC-11 lapse rates below the tropical tropopause and values which are less than the free tropospheric values in either hemisphere. The 6 weeks of WB-57 flights when the aircraft is based in the Indian Ocean will explore routes A and C in Figure 9, the former in conjunction with the C-130 and the P-3. The only opportunity to explore route B will be on the ferry flights, when BN will be accessible because the WB-57 will ferry at 45,000 feet, the altitude of the subtropical jet stream core. Climatologies show 3 maxima in this jet during northern winter, with the 2 most prominent being over north Africa-Arabia and east Eurasia-Japan. These have never been investigated with in situ tracer instruments, and by ferrying westbound out across the Atlantic and westbound home across the Pacific, a great deal could be learned about tracer distributions at the core of the subtropical jet stream, and how effective a barrier it is to quasi-isentropic transfer between the upper tropical troposphere and the lower midlatitude stratosphere.

Payloads and Flight Hours

Payloads which satisfy the objectives for the proposed US aircraft are listed in Tables 6a and 6b. Complete radar information using a combination of tail and fuselage radars and Doppler radial velocities are also proposed for the P-3. The aircraft will be equipped to measure meteorological parameters.

The total flight hours and the allocation of the flight hours between the 3 flight missions (gradient, profile and closure) for each of the US aircraft are shown in Table 11.

Table 11. Flight Hour Allocation (Assumes Malé as the base)*

Type and Mission Total Hours Platforms INDOEX Regions
(see Table 2 for scientific objectives, Figure 13 for regions)

A. Gradient/Profiling Flights
(across regions 2,3,4 and 5)
* solar absorption in clouds
* gases and ITCZ chemistry
* aerosols and dust
65 ER-2
* establishes gradients across regions 2,3,4 and 5; profiles the chemistry, microphysics and optical properties
B. Aerosol-Radiation Closure Flights 20 ER-2
* chemistry and microphysics in regions 3 and 4 * indirect aerosol effects (if possible)

C. Dedicated Aerosol Indirect Forcing 15 C-130
* aerosol indirect effect in low clouds in regions 2 and 3

*(for Seychelles as base, add another 20 hours for the longer transit time)

Flight Operations

The simplistic latitude-height picture given in Figure 19 is not of course what will be experienced on any given day. The real atmosphere will contain three-dimensional circulation systems which will show individual variability and flow patterns. The ITCZ in the area of interest (50°E to 75°E) is marked by cyclonic disturbances centered at about 10ºS. It appears that air flowing from the north sweeps around the eastern and southern flanks of these cyclones, to as much as 5 degrees of latitude before flowing northward into the ITCZ, where it presumably can enter the deep cumulonimbus towers which are found in the cyclones. How far up from the surface does this flow pattern extend? Observations seem to indicate that it may be only 1.5 km deep over India, with the inversion rising in height to be 2.4 km near Gan (1šS, 73šE) and 4 km at 5šS, and disappearing within the cyclonic disturbances in the ITCZ containing the cumulonimbi. Above the inversion, the wind turns rapidly with height, yielding mean easterlies in the 50šE - 75šE sector in the March/April time period.

Since there is a need to survey at least 5š south of the ITCZ, the mean position of which during boreal winter in the Indian Ocean is about 10šS, it is clear for all aircraft that it will be necessary to base close to the equator. The 1000 nautical mile (17š) radii of action from Seychelles and Malé are shown in Figures 18a and 18b. Malé offers the best location for INDOEX because of their proximity to the areas of interest in the Arabian Sea and Indian Ocean. The Seychelles will require longer transit flights and refueling stops at Diego Garcia or Malé.

Weather Forecast

The reality of research aircraft flight planning is that it depends upon having the best possible 24-hour weather forecast and current satellite images, and basing the flight track upon it, combined with the knowledge gained from the initial flights.

Comparison of Satellite and Aircraft Radiometers

On approximately half of the flights envisioned for INDOEX, individual legs of ~100 km length and vertical profiles will be aligned so as to facilitate comparison of the radiances observed with the aircraft instruments and those obtained with instruments on the overflying polar orbiting satellites, in particular the AVHRRs on the NOAA satellites and Clouds and Earth Radiation Energy System (CERES) on TRMM. The polar orbiting satellite observations will be used to calibrate radiances obtained with the geostationary satellite.

6.1.1. The ER-2 and the WB-57

Scientific Objectives

* Identify, by direct atmospheric observations, the vertical structure of the upper tropospheric aerosols (size distribution and composition), cirrus microphysics, reactive gaseous species, and the radiation fluxes (both broad band and spectral).

* Evaluate the effect of cirrus on radiation fluxes over the equatorial Indian Ocean.

* Measure the north-south gradients of aerosols, ozone, radiation fluxes, cloud forcing and solar absorption along the Indian Ocean, across the ITCZ.

* Explore the microphysical factors contributing to the solar absorption and high albedo of widespread tropical cirrus layers.

* Examine the role of ITCZ in trace gas transport and document the differences in aerosols and cloud solar absorption between clouds subject to continental and pristine marine air.

Specific Tasks


* Determine the variation of upwelling, downwelling and net solar broadband (0.3-4 µm) and narrow band spectral fluxes (at least 6 bands) at the flight altitude of the ER-2 and WB-57 for clear skies and cloudy skies.

* Identify clear skies from the lidar, MODIS airborne simulator (MAS), narrow field-of-view longwave radiometer (NFOVR) and solar radiometer on the ER-2. Identify clear skies below the WB-57 from multi-channel cloud radiometer (MCR), NFOVR and solar radiometer on the WB-57. Use microphysics instruments to identify whether the WB-57 is within or below the cirrus.

* Estimate longwave and solar heating within the cirrus by differencing the net flux measured by the ER-2 to the net flux measured by the WB-57.

* Differentiate the broadband and 8-10 µm outgoing longwave radiation (OLR) between clear and cloudy skies to determine the greenhouse effect of cirrus (longwave cloud forcing) and that of ozone. From the ER-2 radiation flux observations, difference the cloudy-sky albedo from the clear-sky albedo to estimate the reduction in absorbed solar flux due to clouds. From similar estimates from the WB-57, separate the effects of cirrus on outgoing longwave and solar fluxes.

* Estimate the profile of longwave and solar heating rates within the mid to upper troposphere with WB-57 for model validation. The longwave cooling rates would be estimated with the spectral radiance between 15 to 100 µm obtained from the uplooking and downlooking, tropospheric airborne Fourier transform spectrometer (TAFTS) (Table 6a). This instrument would primarily measure cooling rates from CO2 and the rotation band of water vapor.

* Estimate the upwelling radiance at the tropopause. The high-resolution interferometer sounder (HIS) on ER-2 yields radiance in the wavelength range of 3.4 to 16.6 µm, with about 0.5 cm-1 resolution. With accurate (< 1 K) knowledge of the sea surface temperature, the spectral radiance can be used to estimate the greenhouse effect of ozone, methane and other anthropogenic gases and the infrared effect of aerosols. In addition, it can be used for vertical profiles of temperature (accuracy of 1K) with 1-3 km vertical resolution down to cloud top.

* Determine the scattering phase function for aerosols by placing the WB-57 in pylon turns over cloud-free but aerosol burdened regions and use the MCR to measure the anisotropy of the reflected radiances.

Water-vapor concentrations

* Measure the variation of water-vapor and methane concentration along the flight altitude of the ER-2 and the WB-57.

* Measure the profile of water vapor from 18 km to about 10 km from onboard instrument and below about 10 km with the WB-57's dropsondes.

Cloud properties

* Estimate the optical depth of the cirrus at selected wavelengths using the total direct-diffuse radiometer (TDDR) (Table 6a, Table 6b) on the WB-57. Determine the albedo of the cirrus from the ER-2 and the lower-level clouds from the WB-57. Albedo is defined as the ratio of reflected upward solar flux to downward solar flux.

* Estimate cloud-top altitudes with the lidar and NFOVR on the ER-2. Estimate cloud-base altitudes with the WB-57.

* Determine cloud-top temperature and estimate optical depth of thick cirrus (* > 1) and cirrus emissivity by retrieval from the ER-2 NFOVR data at 6.7 µm and 10.5 µm. These measurements will be taken only after flux-divergence-related objectives have been met and during missions dedicated to microphysical observations by the WB-57.

* Determine the scattering phase function for cirrus ice crystals by placing the WB-57 in pylon turns over isolated, single-layered, optically thin cirrus and use the MCR to measure the anisotropy of the reflected and emitted radiances.


* Measure the cirrus ice-particle-size spectra and characterize the ice crystal habits along the equatorial tracks. Collect data from the cirrus cloud base (or 25šC and lower temperatures) to as high as the WB-57 can penetrate, emphasizing the higher altitudes (colder temperatures).

* Simultaneously and at the same horizontal position, collect cloud brightness data and cloud structure with the ER-2. Retrievals of cloud structure and cloud microphysics will be undertaken using the lidar and MAS observations.

* Collect ice crystals with the WB-57 for later analyses of particle nucleus composition and size.

* Measure the winds and radar reflectivity structure with the P-3, coordinating measurements with the WB-57 and the ER-2.

Chemistry (NOAA component)

Measure the concentrations of the following gases:

* Ozone and NO/NO2/NOy [Ridley et al., 1994]. These gases are central to tropospheric chemistry, and will address issues connected with the questions raised under Objective 3. The NOy content may be affected by both continental sources (including many anthropogenic ones) and by lightning, the latter being obviously of potential importance in and near the ITCZ. The ozone content is clearly important because of its role as a greenhouse gas; understanding the time dependence in the flow will be important.

* CO. It is an excellent tracer of continental air, and also has a very steep decrease at the tropopause. It will be an invaluable component in understanding general airmass character and the ozone evolution in particular.

* CH4. High precision mixing ratio measurements of CH4 serve as a means of estimating what fraction of stratospheric air is present in an upper tropospheric air parcel.

* Nitrous oxide. It is produced at the surface and is lost only in the stratosphere, so it is a tracer for stratosphere - troposphere exchange. It also is the main source of nitrogen oxides in the stratosphere, and study of its correlation with NOy is very informative.

* Hydrocarbons. Hydrocarbons are very important for two reasons. One is the recent discovery of the importance of organic compounds containing the carboxylic acid group in individual aerosol particles, the other is their potential role in the chemistry controlling the speciation of reactive nitrogen and in the balance of OH and H2O.

* H2SO4 and SO2. Since a significant fraction of the aerosol in the upper tropical troposphere is thought to be sulfuric acid, it is clearly important to have measurement of H2SO4 vapor and its major gaseous precursor, SO2.

* CO2. When measured with high precision, it can be an informative tracer for northern hemisphere air, and for stratospheric air [Boering et al., 1995].

* Air Samples: Whole air measurements, in which air is pumped into special containers and subsequently analyzed in a laboratory, bring some additional chemical information with them. Although having lower time resolution (a few minutes) than continuous-working instruments, it can measure a large array of species with a spread of lifetimes ranging from days to decades, a virtue which will be useful when considering air samples from the lower troposphere to the tropopause in the ITCZ. It is of particular interest to see if species with lifetimes of 1-2 days against tropospheric chemical loss are transported to the upper tropical troposphere and lower stratosphere; methyl iodide is one example.

Aerosols (NOAA Component)

The aerosol payload on WB-57 (Table 6a) is largely motivated by recent work which, using the University of Denver suite of instruments [Wilson et al., 1993], has shown that the tropical upper troposphere is a source of new particles [Brock et al., 1995], which confirms the earlier observations and hypothesis by Clarke et al (1993) that convective outflow regions in the free troposphere were a favored region for new particle production.

* Measure the size distribution of aerosols from about 0.008 to 2 µm. The focused cavity aerosol spectrometer (FCAS) measures size distributions in the 0.08 to 2 µm diameter range. The condensation nucleus counter (CNC II) measures the mixing ratio of particles larger than 0.008 µm, and by using one heated and one unheated channel, can determine the fraction which is volatile and is presumably therefore sulfuric acid. Combining the FCAS and CNC II data permits study of the temporal evolution of the aerosol, including the new particles.

* Measure the chemical content of new particles: The multi-sample aerosol collecting system (MACS), which can collect particles down to 0.01 µm diameter on up to 24 electron microscope grids for subsequent laboratory analysis. These samples have shown the sulfur content of new particles formed in the upper troposphere [Sheridan et al., 1994], and gradients of chemical composition across the tropopause in particles characterized as crustal, metallic, marine, carbon rich (soot) and carbon rich (other). These measurements are needed to understand the effect upon the aerosol of mixing marine and continental air masses.

* Measure the refractive index and size distribution of larger size particles. The size distribution needs to be covered out to larger diameters than the 2 µm maximum provided by FCAS, and this can be done by the multi-angle aerosol scattering probe (MASP), which can measure size distributions in the 0.3 to 20 µm range. It can also obtain the refractive index of the particles, which is valuable directly in radiative calculations and which may also provide some information on chemical composition.

* Measure the aerosol chemical composition. The ability to measure the chemical composition of individual aerosol particles in real time has recently been demonstrated [Murphy and Thomson, 1995] by the particle aerosol laser mass spectrometer (PALMS). Such observations should prove to be fundamentally important to questions about whether aerosols are externally or internally mixed, and what the range of chemical compositions is in the aerosol present in the upper tropical troposphere.

6.1.2. The C-130

The C-130 will operate in the low to mid-troposphere in support each of the primary INDOEX objectives. The instrumentation emphasizes radiation, aerosol physical, chemical and optical measurements along with several gas phase species expected to be important to understanding the properties, formation and evolution of the aerosol. The C-130 will be the primary platform for addressing the in situ direct radiative forcing of the aerosol and linking these measurements to the contributions due to sulfate, dust and other species. It will also provide in situ aerosol, CCN and cloud droplet measurements that will be linked to radiative measurements aboard the ER-2 and WB-57 required to understand the indirect forcing issues associated with low stratus and stratocumulus clouds. Another primary function of the C-130 will be to work in coordination with the ER-2 and WB-57 to acquire radiative measurements below the cirrus outflow. It will also profile and characterize the low level inflow region of the ITCZ in coordination with the WB-57 as part of a combined effort to document the role of the ITCZ in the transport and processing of marine boundary layer air.

Scientific Objectives

* Quantify the direct radiative effects of aerosol types present over the Indian Ocean.

* Determine the contribution and relative importance of sulfates, dust, sea-salt, soot and black carbon to the direct aerosol forcing.

* Explore the indirect radiative effect of aerosol on cloud optical properties through measurement of the radiative properties of clouds in diverse regions and their relationship to measured aerosol properties, CCN and droplet spectra.

* Measure the direct and total solar radiation reaching the sea surface in broad band and about 6 visible and near IR bands.

* Provide measurement of aerosol physics, chemistry and related gaseous species appropriate to evaluation of gradients, vertical profiles and transformations occurring on local and regional scales.

Specific Tasks


* Determine the variation of upwelling, downwelling and net solar broadband (0.3-4 µm) and narrow band spectral fluxes (at least 6 bands) flux at the flight altitude of the C-130 for clear skies and cloudy skies.

* Identify clear skies from the lidar, NFOVR and solar radiometers. Identify clear skies below the WB-57 from NFOVR and the solar radiometer on the WB-57.

* Estimate longwave and solar heating by differencing the net flux measured by the ER-2, the net flux measured by the WB-57 and the net flux measured by the C-130.

* Differentiate the broadband and 8-10 µm OLR between clear and cloudy skies to determine the greenhouse effect of cirrus (longwave cloud forcing) and that of ozone. From the ER-2 radiation flux observations, difference the cloudy-sky albedo from the clear-sky albedo to estimate the reduction in absorbed solar flux due to clouds. From similar estimates from the WB-57 and C-130, separate the effects of clouds on outgoing longwave and solar fluxes

* Estimate the profile of longwave and solar heating rates within the mid to low troposphere for model validation.

* Measure altitude profiles of radiative fluxes and optical depth (with the TDDR) to relate to simultaneously measured aerosol properties and achieve radiative closure.


* Determine aerosol size, concentration and chemistry from 0.003 to 20 µm at all altitudes and during vertical profiles.

* Measure the aerosol optical properties: scattering coefficient and absorption coefficients that determine aerosol single scatter albedo, the mass scattering coefficient, and the humidity dependence of these coefficients.

* Measure the dynamics of the size distribution between 0.003 and 0.3 µm to allow assessment of nucleation events, cloud processing and aerosol modification during air mass evolution.

* Integrate these measurements both horizontally and vertically with those of key gas-phase species to establish regional characteristics suitable for satellite intercomparisons and model development.

* Measure the CCN and cloud droplet spectra in conjunction with radiative characterization of stratus and stratocumulus.

Aerosol Radiative Effects, Physics and Chemistry

Specific capabilities afforded by the C-130 instrumentation listed in Table 6a include: the aerosol size distribution from 0.003-20 µm; in situ and column optical properties and size resolved mass scattering coefficient; evidence of new particle production; internal/external mixing; long range transport of continental aerosol (e.g., soot); modification due to precipitation or clouds; the total number distribution of sea salt; optical extinction. The rapid gas phase and aerosol measurements, in conjunction with lidar instrumentation will allow us to map their features and the evolution of aerosol field. Moreover, the vertical profiles from takeoff and landings at Malé for each mission will allow assessment of the column representativeness of similar surface measurements being made there as part of the INDOEX program.

* Dynamics of the aerosol distribution will be studied with a variety of techniques, including a modified 256 channel laser optical particle counter (OPC) [0.15< r < 7.0 µm] [Clarke, 1991].

* Two TSI 3760 CN counters provide total CN, refractory CN (@ 300šC) and volatile CN (by difference) as a real time indicator (1 Hz) for air mass variability [Clarke, 1993]. The distribution of aerosol in smaller sizes (0.007< r <0.35 µm) is measured by a radial differential mobility analyzer (RDMA) [Zhang, 1995]. This instrument also has a thermal decomposition system similar to the OPC that can address the internal vs. external mixing of the aerosol.

* A continuous CCN counter will be used to obtain the CCN supersaturation spectra that can be compared to the RDMA observations.

* Recently it was found that nucleated particles in the 0.003-0.01 µm range can be detected by difference between counts from the TSI 3025 (ultrafine condensation nucleus counter) and the TSI 3010 counter.

* The simultaneous measurements of radiation fields and aerosol optical properties will be used to quantitatively link aerosol to radiative effects. A continuous light absorption photometer (CLAP) provides a real-time index of the aerosol absorption coefficient (usually soot carbon from combustion) for optical models and provides a direct means of stratifying the data with regard to clean and polluted conditions.

* Two new TSI 3563 3-wavelength nephelometers will provide the continuous integrated total scattering and backscattering coefficient both in a heated (low relative humidity, RH) mode and at a high controlled RH mode in order to measure the increase in scattering of the aerosol due to water uptake, providing direct validation of "dry" and "wet" calculated scattering coefficients. A Berner impactor (size cut at 1.0 µm) will be alternately cycled upstream of the nephelometer so that the optical properties of the total and the submicrometer aerosol can alternately be characterized including their RH dependence.

Related measurements include bulk filter samples collected for major anion and cation analysis (e.g., SO4=, NO3-, Cl-, Na+, etc.) by ion chromatography. These will be size resolved and require about 30 min. sample time. Major gas phase species (e.g., SO2, DMS, O3) will be measured to allow assessment of the evolution and relationship of these species to aerosol properties during transport. Characterization of the aerosol fields for our flight missions will be provided by the NCAR scanning aerosol backscatter lidar (SABL), and it will allow extension of our in situ measurement to associated and representative spatial scales.

The above radiation and aerosol instrumentation will provide a continuous characterization of the radiative and in situ aerosol fields at all times during the aircraft missions. The combination of instruments have been used to distinguish between marine, continentally polluted, continental dusts and "clean" regions of the troposphere and their corresponding radiative layering. The optical measurements can yield the aerosol extinction, single scatter albedo and the asymmetry parameter, all of which are required for calculating the radiative flux perturbation from an aerosol layer. The dual nephelometer measurements of RH-dependence of aerosol light scattering are required to model aerosol differential and column optical depth for varying ambient RH.

The bulk aerosol samples and gas phase measurements will allow the interpretation of other data in terms of specific chemical species important to aerosol properties and evolution. These instruments are flight ready for INDOEX and have been recently deployed during the 1995 Southern Oxidant Study (NOAA P-3), the 1995 ACE-1 (NCAR C-130) and the ASTEX (NCAR Electra).

6.2. Surface Observations

(J. Prospero, V. Ramanathan, G. Shaw and E. Woodbridge)

Scientific Objectives

* Characterize the physical and chemical properties of aerosols that are important to the radiative properties of the atmosphere and to climate.

* Investigate the processes that affect these properties; and assess the relative importance of natural and human sources.

* Estimate the magnitude of solar absorption in tropical atmosphere through a full annual cycle and understand the physics and chemistry behind the absorption.

Specific Tasks

* Determine the seasonal variation in aerosols, radiation fluxes, cloud properties and gases, across the full south-west summer monsoon and the NE winter monsoon.

* Use high spectral resolution solar radiance measurements between 0.3 to 3 µm and compare observed radiances with computed radiances to identify the causes for the excess absorption in clear and cloudy skies.

* Establish the link between the polluted continental air, aerosol composition, its optical properties and radiation fluxes.

* Establish the link between deep convection and vertical distribution of ozone and water vapor.

* Establish the seasonal variation of the boundary layer and the depth of the NE monsoon inversion.

The surface-based experimental program has two primary components:

A continuous sampling program. The long term surface observations of INDOEX are designed to build a climatology of the Indian Ocean region in terms of aerosol and cloud radiative forcings, the impact of anthropogenic emissions from the Indian subcontinent on those forcings, and the regional or global extent of the anthropogenic perturbations. This effort will develop a climatological data set on a wide range of atmospheric chemical and physical properties. These activities will be largely focused on Malé; ancillary activities could take place on mainland India and on the southern island Reunion.

The envisioned three-year continuous sampling campaign with radiometers, aerosol collectors and a limited suite of chemical tracer measurements provides some means of estimating the seasonal and interannual variability in the observed fields. Such estimates are crucial to bounding the data that comes out of the intensive field phase (IFP). In addition, this continuous data provides a means of extending the IFP data to time scales of relevance to climate.

The IFP adds additional measurements that augment the continuous data set and complement the planned aircraft and ship programs. This intensive will take place concurrently with the main field phase of INDOEX, and will require the presence of trained scientists.

Site Selection Strategy

The nature of this investigation requires a surface observing site in the northern Indian Ocean subject to the influence of the northeast monsoon. Measurements of solar flux under clear and cloudy conditions, aerosol radiative properties and chemical composition, trace gases tied to anthropogenic emissions, and standard meteorological variables are essential at this site. The nature of these measurements and the scientific objectives dictate certain minimum requirements for this site. First, it should be far enough from major land masses to be representative of the remote marine environment. Second, the topography or "orography" of the site should have little impact on the regional and local atmospheric circulation and meteorology. Third, the population of the site should be small enough or far enough from the observing station to prevent recording of local anthropogenic emissions. Fourth, although the instrumentation planned for the long term observations is largely automated, there is still a need for trained personnel to perform daily, weekly or monthly maintenance of some instruments. Lastly, regularly scheduled air service to the site is an important part of maintaining the quality of the observations while reducing the overall cost of the program.

Republic of the Maldives

Ideally the northern hemisphere site would be off of the south western coast of India. The Maldives archipelago is a double chain of 19 coral atolls, comprised of roughly 1200 islands. No island is larger than 13 km2 and none rise more than ~3 m above sea level. The capital of the Republic of the Maldives, Malé, located on the eastern arm of the archipelago (4.2šN, 73.5šE), has a population of 55,000 (1991 est.) on an island of 1.5 km2. The entire archipelago has a population of ~252,000 [Wells, 1948; IOA, 1976; United Nations Environmental Program (UNEP), 1986; US Dept. of State, 1990; Stanger and Majeed, 1994; World Almanac, 1995].

The Maldives are sufficiently far from India and Sri Lanka to be truly representative of the remote marine environment of the northern Indian Ocean, yet they are strongly influenced by the northeast monsoon [Stanger and Majeed, 1994]. The islands themselves have little effect on the general circulation or meteorology of the region, and the small population minimizes the anthropogenic impact on the air and water quality of the region. This makes the Maldives ideal for studying the influence of pollutants carried by the northeast monsoon on the chemical composition and radiative properties of aerosols and clouds.

Figure 22. Summary, four years of observations at Malé. (A.J. Wells, Weather, October 1948, pg. 310)
Figure 22 presents the monthly means from four years of surface observations at Malé [Wells, 1948]. In January and February, 80% of the observations report wind from the northeast quadrant, illustrating the strong influence of the northeast monsoon. By March a large shift to the northwest is underway, signaling the onset of the southwest monsoon, which is well established by early April. This is very consistent with the monthly flow patterns from the CAIO [1979].

Precipitation shows a negligible seasonal cycle at Malé, and small interannual variability [Stanger and Majeed, 1994]. Extreme precipitation events are associated with single, stationary convective cells, or with the spiraling feeder bands associated with strong storm systems or cyclones to the north or south of the Maldives [Stanger and Majeed, 1994].

In short, the Republic of the Maldives is the ideal location for studying the influence of the northeast monsoon and the associated anthropogenic emissions from the Indian subcontinent on aerosol and cloud radiative forcing.


Table 7a lists the measurements for the Continuous Protocol. Table 7b lists the Intensive Protocol. Both tables list the location of the operation, the principal investigators (PI), and the type of measurement (aerosols, gases, radiation, and meteorology). We recognize the need for observations in the southern Indian Ocean, south of the ITCZ, to assess the aerosol and cloud radiative forcing in a very clean, unperturbed environment. For these measurements we will leverage the existing efforts of other US and international programs in Reunion/Mauritius and Seychelles. These programs and the associated measurements are listed in Table 7a as part of the continuous protocol. Measurements that need support under this proposal are noted "Proposed for INDOEX." Measurement activities that support the objectives of this proposal but are being funded by other sources are noted "Ongoing programs." In addition, relevant observations are available at surface sites in India, Maldives and Mauritius (Table 12).

Table 12. Supporting Platforms/Observations

Observation Institution Location
Aerosol lidar VSSC, ISRO, IITM Trivandrum, Pune
MST Radar Tirupati
Dobson Ozone IMD New Delhi, Pune, Kodaikanal
Brewer Ozone IMD New Delhi, Kodaikanal
Ozone sondes IMD New Delhi, Pune, Trivandrum
Multi-WL Radiometers WMO (BAPMON)
Precip. Obs WMO (BAPMON)
T, H2O(g) profiles (merchant ships) Indian Ocean region

Meteorological observations Republic of Maldives Maldives
Meteorological observations Republic of Mauritius Mauritius


We will carry out an intensive aerosol measurement program that focuses on the chemical, physical, and radiative properties of aerosols over the Indian Ocean and their impact on the radiative balance in the marine atmosphere. We will make a suite of measurements that will allow closure for the direct forcing terms in the radiative energy balance equation - that is, we will seek to obtain internal consistency among the measured aerosol variables and the modeled terms.

Chemical measurements will focus on the principal aerosol components: nss-SO4=, NO3-, NH4+, BC, MSA, sea-salt, and mineral aerosol. Physical properties will include aerosol size distributions, aerosol light scatter and absorption, and the effects of relative humidity changes on these properties. Black carbon measurements will provide a good measure of anthropogenic impacts from energy-related combustion sources or from biomass burning (although, as previously stated, the latter is not expected to be important for India).

During INDOEX a large suite of measurements will be undertaken to relate the aerosol chemical, physical and radiative properties to the size distribution. These measurements involve size-segregated aerosol collection, optical particle counting and aerosol lidar.


The surface radiation measurement program will continuously monitor shortwave radiation, both broadband and in selected wavelengths including the near infrared, in such a way that cloud opacity and microphysics, and aerosol radiative effects, can be accurately inferred. Continuous automated monitoring of downwelling flux, both broadband and spectrally in the visible and some near-IR wavelengths, will serve to retrieve cloud scattering optical depth and aerosol phase function, and provide further insight into near-IR cloud absorption.

During the IFP, a Fourier transform spectroradiometer (FTS) based on the Bomem DA-8 Series Michelson interferometer will be added to the radiation instrumentation. Providing approximately 0.1 cm-1 spectral resolution from 0.3-4.0 µm, this instrument will be configured to measure direct beam radiance transmitted through clear and cloudy atmospheres. In addition, its high spectral resolution will allow insight into continuum absorption by trace gases throughout the near infrared.

Population statistics from the long-term observations, augmented by the extrapolation of high sensitivity measurements from the field intensives, will be used to assess spatial and interannual variability, and the validity of model predictions. The aerosol properties will be related to specific source regions and types based on additional statistical examinations of aerosol properties in conjunction with meteorological analyses. We will test our knowledge by comparing measured with computed radiative characteristics of marine air.

Our strategy follows that described in the Department of Energy (DOE) publication, "Quantifying and Minimizing Uncertainty of Climate Forcing by Anthropogenic Aerosols" [Penner et al., 1993]; we will incorporate many of the same measurements and strategies identified therein.


We will measure the gases CO and O3 as part of the continuous protocol. CO serves as the primary tracer of polluted continental air, carried to the island site by the NE monsoon. O3 photochemistry plays a central role in the tropospheric oxidation of many trace gases, and is central to maintaining the photochemical balance of HOx, NOx and Ox. The continuous protocol will aid in developing a climatology for these species and assessing the impact of continental air on the photochemistry of the remote marine troposphere.

While anthropogenic impacts are of major concern in this program, such impacts can only be properly assessed by considering natural sources as well. Of the major species studied in this program, only nss-SO4= has a substantial natural oceanic source, DMS, which is oxidized in the atmosphere to a variety of products including SO4= and MSA (methanesulfonate). MSA can be used to assess the natural oceanic nss-SO4= fraction. The DMS emissions vary with season and, depending on the time of year, the ocean can contribute a substantial (at times, major) fraction of the nss-SO4= aerosol even in regions that are normally dominated by pollutant SO4=. During the continuous protocol, DMS will be measured by whole-air sample at regular intervals.

During the IFP, additional measurements of SO2 and NOy/NOx will be made to estimate the anthropogenic contribution to marine aerosol nss-SO4= and bound the HOx/NOx/Ox photochemical system.

6.3. Ships and Buoy

(R. Dickerson, D. Kley, A.P. Mitra, D. Lubin, G. Shaw)

Sicentific Objectives

* Determine the effects of aerosols and low and high (cirrus) clouds on solar radiation fluxes at the surface over the Indian Ocean.

* Measure the latitudinal gradient in marine boundary layer (MBL) concentrations and properties of selected trace species.

* Explore the impact of continental emissions, deep convection, and MBL photochemistry on the composition of the lower and upper troposphere over the Indian Ocean.

* Study the transport and transformation of ozone, aerosols and their precursors.

Specific Tasks


* Measure the spectral radiance and irradiance using pyranometers and Fourier transform infrared radiometers (FTIR).

Vertical Profiling

* Probe the troposphere for standard dynamic and thermodynamic variables every 6 hr.

* Probe the troposphere for water vapor and ozone and the stratosphere for ozone every 12 hr.

* Probe the troposphere and stratosphere for ozone and water vapor, with frost-point hygrometers, twice a week.

* Measure ozone and aerosol backscatter profiles with lidar.

* Measure winds from the MBL to about 11.5 km altitude continuously with a Doppler wind profiler.

Gas-Phase Chemistry

* Measure in the MBL (on time scales of minutes to hours) ozone, its precursors and associated variables:

Aerosol Properties

* Measure (with filter samples) bulk aerosol composition every 12 to 24 hr.

* Measure (with impactors) size-segregated aerosol composition every 12 to 24 hr.

* Determine the size-number distribution of MBL aerosol.

* Determine the aerosol scattering and absorption coefficients.

* Measure total aerosol optical depth at several wavelengths.

* Count condensation nuclei and determine their basic composition.

* measure the precursors to particle formation: DMS, MSA, SO2.

Ships are the third leg of the tripod on which the INDOEX field experiment is based. They offer intensive continuous sampling in the MBL and remote sensing of the atmosphere on temporal and spatial scales compatible with the synoptic-scale meteorology of the region. The chemistry and physics of the MBL are key to INDOEX because much of the pollution from India is carried in the lower atmosphere, because convective clouds develop from warm moist MBL air, and because the remote MBL is a major sink for ozone.

The ship portion of the INDOEX program will rely heavily on collaboration with European colleagues. While the US will supply aircraft on which Europeans will participate as co-investigators, Germany will supply the ship on which Americans will participate as co-investigators. Although the likelihood of participation of this German ship in INDOEX is excellent, in the event that it is not available we will request one NSF operated US ship, such as the newly commissioned R/V Roger Revelle. Since we lack the resources to fund this entirely from our program, we will share this ship with oceanographers who have interests in the tropical Indian Ocean.

Ship Capabilities and Logistics

R/V Sonne: Operated by the BMFT, the German equivalent of the NSF, this is a state-of-the-art research vessel (Table 13). She is a converted fishing ship, refurbished for scientific work in 1993, with a complete on-board computer network and a full complement of basic oceanographic and meteorological instruments. D. Kley, ship coordinator, and P. J. Crutzen have made the formal request for the research vessel, and have estimated high probability of the proposal being approved. The timing is excellent, since the Sonne will be finishing a geophysical expedition in January 1998 with the last stop in Colombo, Sri Lanka. The BMFT Panel will allocate ship time in May 1996.

Table 13. Ship Capabilities and Logistics

Sonne Sagar Kanya
Size 97m 100.34 m
Scientific Crew 25* 31
Endurance 50 sea days 45 sea days
Range 14,000 n. miles 9999 n. miles
Speed 12.5 knots 14.2 knots
Laboratory space 475 m2 290 m2
Container space 7 spaces 2m x 6m

* at least half of whom must be German scientists

R/V Sagar Kanya: Operated by the National Institute of Oceanography, Goa, India, this ship conducted in January 1996 a pre-INDOEX cruise from the Arabian Sea directly south to the ITCZ (Table 13). The INDOEX part of this cruise consisted of chemistry, aerosol and radiation experiments, including four instruments from US investigators. The Indian institutions participating are: National Physical Laboratory, New Delhi; Physical Research Laboratory, Ahmedabad; and the Space Research Organization, Trivandrum. Dr. A.P. Mitra (the INDOEX PI from India) has contacted the Indian Ocean Development Board about participation in INDOEX and received a favorable response.

Sampling Strategy and Proposed Cruise Tracks

The ship will travel from regimes directly impacted by continental emissions to those impacted less strongly, and on to the relatively clean Southern Hemisphere (Figure 18a). The ship will follow the mean low-level air trajectories (streamlines) to the ITCZ (Track 1; Figure 18a). This portion of the track will monitor changes in atmospheric composition and properties as air is advected away from the source region. From the ITCZ, the track heads directly to the SH clean-air site, Reunion. The approach from the northeast avoids local contamination from Madagascar and its archipelago. Starting at about 15šN, a second track will cruise SSE parallel to and a few hundred km off the coast of India, to obtain a profile of emissions (Track 2; Figure 18a). Possibly the Sagar Kanya will focus on this leg closest to India to free the Sonne for more time in remote regions. An alternative return route (Track 3, Figure 18a) parallels the ITCZ in the SH to increase the amount of information collected on the dynamics and microphysics of convective clouds there. This track will produce nearly continuous information on the formation, growth, and dissipation of deep convective clouds, and may offer the greatest hope of discerning the impact of anthropogenic CCN on cloud development.

The ship will occasionally dock at the island locations of the surface sites and the aircraft operations base. During these stops, investigators can meet to make repairs, cross calibrate instruments, discuss results, and modify procedures as necessary.

From 15šN to 20šS along the tracks 1 and 2 (Figure 18a) is approximately 5500 km (3000 n. miles) and at 12.5 knots each leg would last about 10 days. With a day or two in port after alternate legs, four profiles of the Indian Ocean will be conducted in six weeks. These transects will provide latitude/altitude contour plots of ozone, water vapor, and aerosol back scatter. They will further provide complete latitude profiles of concentrations of trace species in the MBL and radiation received at the surface. Rawinsondes released from the ship would improve back trajectories and chemistry transport model (CTM) wind fields in this data-poor region of the world. The ship track is similar to aircraft flight tracks and offers opportunity for correlated measurements and intercomparisons with aircraft instruments.

Proposed Observations and Instruments

The following paragraphs describe the general characteristics of the measurements to be made to meet the INDOEX goals. A detailed list of instruments and measurements is given in Table 8a and Table 8b. The US component, for which NSF funding is requested, is carefully distinguished (Table 8a) from the separately-funded European component (Table 8b).

Basic Meteorological Observations

In addition to the standard surface measurements of meteorological variables, the Sonne will support extensive sonde launches, described below. We also request the Doppler wind profiler and radio acoustic sounding system (RASS) of the NCAR Integrated Sounding System (ISS). The Doppler wind profiler operates at 915 MHz, in conjunction with RASS, to obtain half-hourly virtual temperature and wind profiles. Vertical resolution of RASS profiles are 60 m from 112 m above ground level (AGL), up to a maximum of 1,553 m AGL. Vertical resolution of wind profiles are 238 m from 91 m AGL up to a maximum of 11,755 m AGL, depending upon atmospheric conditions.

The high-resolution ISS data sets are recorded on-site and later transmitted to NCAR for final processing and quality control. A subset of these data (one-hour RASS/wind profiles, mandatory/significant levels, half-hour surface averages) will be inserted onto the Global Telecommunication System (GTS) in real time for incorporation into forecast and analysis models.

Aerosol and Cloud Properties

The shipboard measurements of aerosol physical, chemical and radiative properties largely mirror those of the island site, with the added advantage of measuring the latitudinal dependence of these quantities. Bulk aerosol mass and composition will again focus on the principal aerosol components: nss-SO4=, NO3-, NH4+, BC, MSA, sea-salt, and mineral aerosol. In addition, size segregated aerosol composition will be measured with multi-stage impactors. These impactors collect aerosol particles in a range of aerodynamic diameters most highly resolved in the accumulation mode [Pszenny, 1992]. Physical properties will include aerosol size distributions, aerosol light scatter and absorption, and the effects of relative humidity changes on these properties. Black carbon measurements will provide a good measure of anthropogenic impacts from energy-related combustion sources or from biomass burning.

The physical characteristics of ambient aerosol will be monitored using the methods described in Section 6.1. The Institute for Tropospheric Research (ITL) will likewise investigate the properties of MBL aerosol with a suite of instruments (Table 8b). By employing a differential mobility particle sizer (DMPS), differential mobility analyzer (DMA), and an aerodynamic particle sizer (APS), they will determine the size-number distribution over the range of 3-10 nm diameter. Further, they will investigate the hygroscopic nature of the aerosol with two tandem differential mobility analyzers (TDMA) set to operate at different relative humidities.


Again complementing the island sites, the shipboard measurements of radiation will focus on net, direct and diffuse shortwave and longwave radiance and irradiance. This powerful set of measurements, both broadband and in narrow spectral regions, provides data critical to addressing the INDOEX objectives: determining aerosol optical depth and radiative forcing, the indirect effect of aerosols on cloud scattering and cloud optical depth, and the excess near-infrared (NIR) absorption by clouds. The most complex component will be a FTIR based on the Bomem MR-150 Series Michelson interferometer adapted for use at approximately 4 cm-1 spectral resolution throughout the near infrared (0.7-4.0 µm). This instrument will be configured to measure downwelling zenith radiance, with accurate radiometric calibration provided by on-board blackbodies and standard lamps. With the right combination of detectors, this instrument will cover both the 1.6 µm and 2.2 µm atmospheric windows (for cloud microphysical retrieval), and its high spectral resolution will provide insight into continuum absorption by trace gases throughout the near infrared.

Broadband and spectral irradiance (flux) measurements will be made with a standard suite of filtered and unfiltered pyranometers spanning the UV-NIR. Aerosol optical depth will be determined with a hand-held spectral sun photometer operating at 425, 500 and 790 nm. In addition, we will add another wavelength centered in a water absorption band to derive column water.

The ships will also continuously measure SST, which is key to understanding how aerosols and clouds regulate solar energy fluxes at the sea surface. The SST measurements will be used directly and as a means to calibrate the SST remotely sensed by the INDOEX aircraft and satellite platforms. In turn, the aircraft and satellite data plus SST from one buoy will be used to extend the INDOEX SST and solar radiation data set over a longer period of time and a greater area. Further, the solar and longwave flux measurements at the surface will be correlated with clouds (from satellites) and ER-2 measurements to examine the link between the water-vapor, aerosol optical depth, ozone amount and clouds and downward solar radiation at the surface, as well as the effect of cirrus on surface solar insolation.

The Max-Plank Institute for Meteorology (MPIM) will determine the heat flux and skin temperature of the ocean surface with an FTIR instrument. With these and other observations they will determine evaporation rate and the heat balance of the surface water with high accuracy around 5 Wm-2.

Ozone and Photochemistry

As in the IFP for the island stations, the shipboard measurements will include CO, O3, NO, NOx, NOy and SO2, among others (Table 8a, Table 8b). As noted, CO serves as the primary tracer of polluted continental air, and will be invaluable in assessing the extent to which this air impacts the ITCZ. O3 photochemistry plays a central role in the tropospheric oxidation of many trace gases, and is central to maintaining the photochemical balance of HOx, NOx and Ox. NOx is the principle driver of O3 photochemistry, and thus plays a major role in shaping the oxidative power of the troposphere and in anthropogenic alterations in the natural HOx/NOx/Ox photochemical balance. SO2 measurements provide a means to quantify the anthropogenic contribution to marine aerosol nss-SO4= and any observed perturbation in the radiative properties of the aerosol.

Because reactive nitrogen species are so critical to ozone production, and because their measurement in pptv levels is technically demanding, two groups will measure these species. This planned redundancy improves reliability, and the results will provide important intercomparisons of techniques. NO, for example, will be measured by O3 chemiluminescence (CL) using commercial (Tecan, MPIC) and specially-built (UMD/AOML) detectors. NOy will be measured by Au catalyzed CO reduction (MPIC) and conversion to NO on Mo 375šC (UMD/NOAA). By adding Teflon and nylon filters to inlets of matched NOy converters, we will learn about the speciation of NOy [Dickerson et al., 1995]. NO2 will be measured by TDLAS (MPIC), and photolytic conversion (UMD/NOAA).

Atmospheric profiles

The INDOEX aircraft will measure upward and downward shortwave and longwave fluxes at more or less fixed altitudes. In order to explain or model these measurements, however, it is necessary to follow the variation of temperature, water vapor, ozone and clouds vertically above the sea surface and horizontally along the flight tracks. As explained earlier, the WB-57 will drop a limited number of sondes, which will profile temperature and humidity from an altitude of 35,000­40,000 ft (10.5­12 km) down to the surface. The C-130 will complete intermittent temperature and humidity profiles from altitudes of 30 m - 3 km using aircraft-mounted instruments.

In order to intercalibrate and extend the aircraft profiles, the ship will deploy a series of sondes which are of a type characteristically more accurate than the dropsondes. These upsondes, about 190 "packages" total, will be deployed along the track with special emphasis on deployment at the time of aircraft overflights and at the island sites. The suggested sonde launch routine for the ship while under way includes:

* 120 separate Vaisala (pressure, temperature, and humidity) sondes, one every 60-100 km

* 12 frostpoint hygrometers, with attached Vaisala and ozone sondes, one every 600 km

* 60 separate ozone sondes, one every 500 km.

UV differential absorbing lidar (DIAL) and lidar will be operated by Brandenburg Technical University (BTU) to measure vertical profiles of ozone and aerosol backscatter, while cloud bases will be monitored with a ceilometer. Water vapor profiles will also be investigated with a combination of DIAL and FTIR instruments operated by the MPIM.

PROTEUS Buoy in the ITCZ

A PROTEUS (PROfile TElemetry of Upper ocean currentS) buoy will be moored at 5šS and 73šE and data collected for a minimum of 2 years, including the IFP period. The surface quantities to be measured include air temperature, relative humidity, wind speed and direction, shortwave radiation (direct plus diffuse) and rainfall. Sub-surface data include temperature profiles up to about 200 m. Data are internally recorded with hourly resolution and the daily means are transmitted real time. A ship is needed to deploy the buoy in 1997 and service it in 1998. The Institute of Oceanography at Goa will be contacted for deploying the buoy in 1997. One possibility is to deploy the buoy during the Sagar Kanya pre-INDOEX cruise. As mentioned earlier, Dr. A. P. Mitra has already received approval for this pre-INDOEX cruise.

The data from the PROTEUS mooring will be used to characterize the surface energy balance, in particular the effects of shortwave cloud forcing on the surface insolation, on timescales much longer than the IFP. Comparison of the measurements from the PROTEUS and IMET (Improved METeorological observation) moorings with ship-based ISS (during TOGA-COARE) suggest that data from a single mooring is sufficient to characterize the long-term statistics of cloud forcing and surface winds for a region several hundred km in radius.

6.4. Satellite Component

(J. Coakley, W. Collins)

Scientific Objectives

* Estimate the optical depth and the direct forcing due to aerosols in the Arabian Sea and Indian Ocean, and quantify the contribution to the forcing due to the anthropogenic sulfate component.

* Quantify the indirect forcing due to the aerosols on cloud droplet size distributions.

* Measure the top-of-atmosphere radiation budget, and apportion the radiation budget into the contributions by various cloud systems and the cloud-free background.

* Estimate the sea surface solar radiation budget from observed spectral radiances, broad band fluxes and correlative water vapor and aerosol data.

Specific Tasks

(before INDOEX Field Phase)

* Develop retrievals of aerosol optical depth from AVHHR and/or Sea-viewing Wide-Field Sensor (SeaWiFS) data validated against in situ observations of column optical depth from surface stations and pre-INDOEX ship cruises.

* Calibrate narrowband radiometers on operational weather satellites against broad-band satellite instruments (ScaRaB and CERES).

* Analyze archival satellite measurements for direct radiative effects of aerosols.

Specific Tasks

(during INDOEX Field Phase)

* Capture full-resolution AVHRR and geostationary satellite imagery, TIROS Operational Vertical Sounder (TOVS) infrared sounding data, SeaWiFS radiances, Ocean and Color Temperature Scanner (OCTS) data, and measurements from DMSP satellites.

* Provide real-time information concerning large-scale variations in column-integrated aerosol optical depth, the spatial and temporal evolution of convective systems, and the positions of low-altitude stratus clouds to guide mission planning.

* Coordinate satellite overpasses with aircraft missions for clear-sky radiation closure (Section 6.1.2).

* Test existing algorithms for converting top-of-atmosphere shortwave radiances and fluxes to sea surface solar radiation fluxes.

Description of Specific Tasks

Satellite Instrumentation

The suite of satellite observations include AVHRR and geostationary imagery data, TOVS sounding data, and Earth radiation budget data from CERES/TRMM and ScaRaB. The historical Earth radiation measurements from ERBE and ScaRaB will also be analyzed by the US investigators. As part of its contribution to INDOEX, C4 will undertake the analysis of the geostationary and AVHRR imagery data (Table 3a and Table 9). CERES/TRMM data is expected to be analyzed as part of NASA's EOS, and C4 investigators on the CERES science team will obtain the analyzed results. LMD and C4 will analyze ScaRaB data, and LMD will perform retrievals of temperature and humidity from TOVS data. It is likely that CERES and a new ScaRaB instrument will be launched in April 1997. These instruments will not provide sufficient measurements to resolve the diurnal variations in shortwave radiation. Radiation budgets will also be derived, as they were during CEPEX, from the imaging instruments on various weather satellites (AVHRR, INSAT, and other geostationary platforms).

Estimation of Column-Integrated Aerosol Optical Depth

For INDOEX, the investigators will use numerical radiative transfer to determine the relation of aerosol optical depth and the visible radiances measured by satellite imaging instruments. This method is similar to the current methods for determining optical depths [e.g., Stowe et al, 1992; Kaufmann, 1995]. However, the investigators will incorporate several improvements over most operational methods:

* An empirical model of the variability in clear-sky ocean reflectivity.

* Treatment of the departures of ocean directional reflectivity from the Lambertian model.

* Treatment of the effects of multiple scattering by aerosols.

* Parameterization of aerosol optical properties based upon in situ ship and aircraft measurements from the INDOEX region.

Figure 23. Mean 0.63 µm reflectivities for AVHRR data from March 1989. Solid lines show calculated reflectivities for zero ocean surface albedo and a maritime aerosol with optical depths of 0.0, 0.05, and 0.1. Dashed line is for an ocean albedo of 0.015. (Rao et al., 1989)
In most of the existing algorithms, the ocean surface is assumed to have a constant reflectivity ranging from 0 [Durkee et al., 1991] to 0.015 [Rao, et al., 1989]. The retrieved aerosol optical depths are quite sensitive to the assumed value. For example, Figure 23 shows monthly mean reflectivities of the cloud-free ocean obtained through application of the spatial coherence method [Coakley and Bretherton, 1982] to daytime observations of the NOAA-11 AVHRR for the month of March 1989. The figure also shows the reflectivities expected assuming zero ocean surface albedo and a maritime aerosol having an optical depth of 0.0, 0.05 and 0.1 at 0.63-µm based upon the single scattering approximation [Ignatov et al., 1995]. The dashed curve in the figure is obtained using a surface albedo of 0.015, the value used previously in NOAA's operational processing [Rao et al., 1989]. Based on the comparison of the observations with the dashed curve, the optical depths retrieved with the previous NOAA scheme would have been near zero. This would lead to large errors in calculations of the direct forcing, since an anthropogenic sulfate haze layer with optical depth of 0.05 in this latitude band is expected to produce a radiative forcing of about -1.5 Wm-2 [Kiehl and Rodhe, 1995]. The investigators will also incorporate the variation of reflectivity with the sun-satellite viewing geometry (i.e., departures from the Lambertian model). The dependence of the reflectivity with the viewing geometry will be derived from AVHRR data for pristine ocean regions with historically low aerosol concentrations.

Validation against Radiative Closure Missions During INDOEX

Satellite observations of optical depth can be taken only as an index of the actual optical depth. Reflected sunlight is affected by the aerosol particle size and by the absorption cross sections, which are expected to be nonneglible for the dust component. Consequently, the retrieval of optical depth as measured by the optical depth in the 0.63 µm AVHRR channel will be compared with surface-based observations of optical depth obtained from the Maldives and Reunion stations and from the Indian Ocean ship expeditions prior to INDOEX. Much more detailed analysis of the sources of error in the satellite algorithms will be facilitated by the data from the clear-sky radiative closure missions during the field phase (Section 6.1.2). In particular, it will be possible to remove or mitigate the ambiguity in the satellite algorithms related to multiple aerosol layers [Porter et al, 1996], and to replace the optical and physical properties of the aerosols adopted for the initial version of the retrieval with direct observations of these quantities.

Analysis of the Direct Forcing from Satellite Measurements of Planetary Albedo

Radiation budget data from previous and current satellite sensors will be analyzed in conjunction with imagery data to determine the direct radiative effects of aerosols. Analysis of coincident high-resolution imagery data will help distinguish the aerosol signal from the radiative effects of thin or scattered clouds. For example, under the relatively high coarse particle concentrations witnessed in the flow of desert dust off the African continent, the ERBE scene identification incorrectly identifies dust outbreaks as partly cloudy conditions. Collocated ERBE and AVHRR observations from NOAA-9 and NOAA-10 for the Arabian Sea will be used to measure the effect of the aerosol on the radiation budget under cloud-free conditions. The spatial coherence method combined with cloud screening techniques like those described by Saunders and Kriebel (1988) will be used to identify cloud-free conditions. ERBE fields of view found to be cloud-free will be used to measure the broadband radiances reflected by the cloud-free scenes. An analogous procedure for scene identification for ScaRaB will be implemented by LMD. Scene identification for CERES will be based upon the operational processing of the visible infrared scanner (VIRS) on the TRMM satellite. The satellite estimates of the direct radiative effect will be validated against the measurements of the surface-troposphere albedo from the ER-2 for each of the radiative closure aircraft missions (Figure 21).

Estimation of the Indirect Effect of Aerosols on Cloud Droplet Size

The objective is to measure the effect of the aerosols on cloud microphysics that could produce a change in cloud reflectivity at visible wavelengths. The effect of the aerosol on cloud microphysics first manifests itself as a shift in droplet size, although not all clouds are expected to be susceptible to aerosol influences [Platnick and Twomey, 1994]. Clouds affected by aerosols have droplets that are smaller than those in unaffected clouds. Han et al. (1994) suggests that a dramatic shift occurs in the sizes of droplets inferred for low-level clouds in the Indian Ocean. The droplet sizes range from 6-10 µm in the Arabian Sea just south of the continent, where aerosol concentration is presumably high, to 12-16 µm in the supposedly pristine air just south of the ITCZ.

The Han et al. (1994) retrieval scheme for cloud droplet radius will be applied to pixels determined to be overcast by low-level cloud, as deduced from the results of the spatial coherence analysis. By choosing to use only the overcast pixels for the retrieval of droplet radii, enhancements of the reflectivities at 3.7 µm due to cloud edge effects will be avoided [Coakley and Davies, 1986; Coakley, 1991]. Droplet sizes for cloud systems in the Arabian Sea will be compared with those for similar systems found south of the ITCZ. The similarity of the cloud systems will be judged on the basis of cloud height determined from the difference between the satellite inferred SST and the cloud top emission temperature and from textural analysis of the cloud fields [e.g., Chen et al., 1989; Gollmer et al., 1995]. Shifts in droplet sizes will be correlated with the optical depth deduced from the nearby cloud-free pixels. Changes in the reflectivities at visible wavelengths will also be correlated with the optical depth. Measurements of the shortwave radiative forcing will be deduced from collocated ScaRaB and CERES observations.

The strategy for measuring the indirect effect is admittedly exploratory. It does not constrain other factors which influence cloud reflectivity, most notably the cloud liquid water. It is unlikely that cloud liquid water can be retrieved from the Special Sensor Microwave Imager (SSM/I) observations [Greenwald et al., 1993] with sufficient accuracy to provide the needed constraint. At visible wavelengths, reflectivities are relatively sensitive to the column amount of cloud liquid water and insensitive to droplet size. Cloud liquid water varies dramatically from pixel to pixel at the 1-4 km scales of typical imagery data. Within this variability the 10% shift caused by changes in droplet number and size is all but lost. Examples of the difficulties encountered in detecting such shifts were shown in the observations of ship tracks off the coast of California [Coakley et al., 1987].

If evidence is found for the indirect effect, the investigators will develop new schemes for further constraining the analysis and refining the observations. The properties of upper level, layered clouds will be treated with the same techniques. While upper level clouds contain ice crystals which make the inference of microphysical properties dubious, reflected sunlight at 3.7 µm and emission at 11 and 12 µm indicate the common occurrence of small particles in these clouds [Inoue, 1987; Prabakhara et al., 1988; Lin and Coakley, 1993; Baum et al., 1994].

Top of the Atmosphere Radiation Budget

Calibrated radiation budget observations at the top of the atmosphere will be obtained from ScaRaB and CERES if these satellites are operational during INDOEX. Prior to INDOEX, measures of the direct and indirect radiative forcing will be derived from ScaRaB and ERBE data combined with high-resolution imagery from AVHRR and other satellites. These measures will be used to determine if the observations obtained during the INDOEX field phase are representative of climatological conditions.

Figure 24. 1-minute averages of coincident shortwave albedos from the NASA ER-2 and GMS satellite during CEPEX.
While the CERES and ScaRaB satellites will provide useful estimates of the earth-radiation budget on weekly and monthly time scales, the temporal sampling is too infrequent to study diurnal variations. In order to follow the diurnal variation of cloud radiative effects, geostationary sensors calibrated against aircraft radiometers will be used to estimate radiation budget components at the top of the atmosphere. This approach has been developed and tested using CEPEX data. Figure 24 shows 1-minute averages of the shortwave albedos obtained from the ER-2 during CEPEX, and corresponding instantaneous estimates of the albedos constructed from geostationary meteorological satellite (GMS) visible data. To construct these estimates, the radiances obtained with the GMS satellite were integrated over the upwelling hemisphere in the reference frame of the aircraft. The root-mean-square error in the satellite estimates is near (approximately 2x) the theoretical limit imposed by the course digital resolution of the satellite visible imagery. This calibration procedure will be applied to radiation measurement systems RAMS data from the ER-2 and WB-57 during INDOEX (Table 6a) using the MAS imager on the ER-2 and MCR on the WB-57 to correct navigational errors in the satellite images. A similar method has been developed for relating the satellite mid-infrared measurements to broadband longwave fluxes measured on the ER-2. These procedures will be used to examine the diurnal variability of the shortwave albedos of clouds and aerosols over the Indian Ocean and Arabian Sea.

The identification of individual cloud systems and the derivation of the radiative properties of these clouds will be based upon two methods: the spatial coherence technique for identifying layered cloud systems [Coakley and Bretherton, 1982], and a cloud cluster identification method that traces the growth and evolution of individual clouds [Boer and Ramanathan, 1995].

Calibration of Satellite Sensors

Through the AVHRR Pathfinder project, the AVHRR on the afternoon polar satellites have been calibrated using terrestrial targets [Rao and Chen, 1994]. NOAA plans to continue to update the calibration of the shortwave channels (Rao, private communication). The AVHRR calibration can be transferred to INSAT and other geosynchronous satellites observations by comparing the observations with data from the afternoon AVHRR in similar sun-target-satellite geometries. During INDOEX, the ScaRaB and CERES satellites should be available as calibration standards for the geosynchronous satellites. Using collocated observations with similar sun-target-satellite geometries, the visible sensor on the geostationary satellite will be calibrated against CERES and ScaRaB.

The satellite imaging instruments will also be calibrated against the MAS multispectral imager on the ER-2 (Table 6a) on approximately half of the flights envisioned for INDOEX. Individual legs of ~100 km length will be aligned to facilitate the intercomparison by matching sun-target-sensor geometry of the satellite and aircraft instruments [e.g., Smith et al., 1988]. The legs will be initiated within +/- 10 min of the satellite overpass.

Field Operations

During INDOEX, geostationary imagery and polar orbiting data from AVHRR, TOVS, SeaWiFS, DMSP, and possibly OCTS will be recorded and processed using portable ground stations. One station will process the data from polar orbiting platforms, and the second will be dedicated to the geosynchronous imagery. The AVHRR and geostationary imagery data will be used to plan missions and to monitor the relationship of the aircraft to clear and convective regions in support of the closure experiments (Section 6.1.2) and radiation profiling flights (Figures 19b,c). Through a cooperative agreement, the SeaSpace Corporation will conduct investigations using AVHRR, TOVS, SeaWiFS, OCTS and DMSP data in collaboration with C4. These investigations are expected to yield value-added products such as aerosol optical depth retrievals and ocean color from SeaWiFS and OCTS, and the initialization of mesoscale models with TOVS soundings. SeaSpace will interact with other INDOEX investigators to make these products (based upon standard community algorithms) available for INDOEX analyses.

The sources and availability of geostationary observations have not been determined, but the investigators are actively pursuing several options. Plans are underway to acquire and use INSAT data in collaboration with the Indian Space Agency. In early 1996, China is expected to launch a geostationary satellite (FY-2) that provides data similar to that of the GMS satellite. Likewise, Russia plans to launch a second geostationary satellite in the ELEKTRO series within the next two years (1996-1997). Observations from either satellite would allow adequate coverage of the INDOEX region. The possibility of moving a redundant meteorology satellite (METEOSAT) over the Indian Ocean is also being actively pursued by European PIs (R. Sadourny, LMD). Based upon the published specifications of these sensors, the SeaSpace ground stations will be capable of processing data from any of the geostationary satellites that view the INDOEX region. While the specific sources of geostationary imagery data are still unclear, it is very likely that the required observations will be available before the start of the INDOEX field phase.

NOAA will be flying the K,L,M series of satellites with a modified AVHRR sensor by the time INDOEX is underway. The modification replaces the 3.7 µm channel with a dual channel that measures reflected sunlight at 1.6 µm for daytime observations and 3.7 µm emission for nighttime observations. The channel at 1.6 µm is expected to be more sensitive to sunlight reflected by aerosols than the channels at 0.63 or 0.89 µm currently available on the AVHRR, but it will be less sensitive to changes in the radii of cloud droplets than the channel at 3.7 µm. The new observations should help quantify the direct effect of the aerosols but may prove troublesome for measuring the indirect effect, although some indication of droplet size should remain [Wetzel and Vonder Haar, 1991].