Characterization of the Particle Emission from Ships Operating at 1 Sea Using Unmanned Aerial Vehicles 2

This research demonstrates the use of an unmanned aerial vehicle (UAV) to characterize the gaseous (CO2) and 10 particle (10 500 nm) emissions of a ship at sea. The field study was part of the research voyage “The Great Barrier Reef as 11 a significant source of climatically relevant aerosol particles” on-board the RV Investigator around the Australian Great 12 Barrier Reef. Measurements of the RV Investigator exhaust plume were carried out while the ship was operating at sea, at a 13 steady engine load of 30%. 14 The UAV system was flown autonomously using several different programmed paths. These incorporated different altitudes 15 and distances behind the ship in order to investigate the optimal position to capture the ship plume. Five flights were 16 performed, providing a total of 27 horizontal transects perpendicular to the ship exhaust plume. Results show that the most 17 appropriate altitude and distance to effectively capture the plume was 25 m above sea level and 20 m downwind. 18 Particle number (PN) emission factors (EF) were calculated in terms of number of particles emitted (#) per weight of fuel 19 consumed (Kg fuel). Fuel consumption was calculated using the simultaneous measurements of plume CO2 concentration. 20 Calculated EFPN were between 9.19 x 10 14 and 5.15 x 10 15 #∙(Kg fuel) -1 . These values are in line with those reported in the 21 literature for ship emissions ranging from 0.2 6.2 x 10 16 #∙(Kg fuel) -1 to 6.2 x 10 16 #∙(Kg fuel) -1 . 22 This UAV system successfully assessed ship emissions to derive emission factors (EFs) under real world conditions. This is 23 significant as, for the first time, it provides a reliable, inexpensive and accessible way to assess and potentially regulate ship 24 emissions. 25


Introduction
Shipping is the most significant contributor to international freight, with almost 80% of the worldwide merchandise trade by volume transported by ships in 2015 (UNCTAD, 2015).Emissions from this transportation mode are a significant contributor to air pollution, both locally and globally.Ships are a major pollutant source in areas surrounding harbours (Viana et al., 2014), with over 70% of emissions reaching 400 km inland (Fuglestvedt et al., 2009).In 2012 exhaust from diesel engines, the predominant source of ship power, was classified as a group 1 carcinogen by the International Agency for Research on Cancer (IARC).In 2007, pollution from ship exhaust was found to be responsible for approximately 60,000 cardiopulmonary and lung cancer deaths worldwide annually (Corbett et al., 2007a).Such emissions are also a strong climate forcing agent, contributing to global warming through the absorbance of solar and terrestrial radiation (Winnes et al., 2016;Hallquist et al., 2013a;Lack et al., 2011).Despite these findings, emissions from shipping have consistently been subject to less regulation than those of land-based transport with ship emissions in international waters remaining one of the least regulated parts of the global transportation system (Cooper, 2001(Cooper, , 2005;;Corbett and Koehler, 2003;Corbett and Farrell, 2002;Eyring et al., 2005;Streets et al., 1997;USEPA-OTAC, 2012).Currently, no specific restrictions for ship-emitted particulate matter (PM) exist, with the only regulated pollutants being NOx and SO2.The International Maritime Organization (IMO) recently revised the regulation of these gaseous pollutants through the Annex VI of the International Convention for the Prevention of Pollution from Shipsthe Marine Pollution Convention (MARPOL).The IMO expected that these regulations would lead to an indirect decrease in particle number (PN) concentration due to the reduction of NOx emissions and the use of fuel with lower sulphur content [14].
However, it has been found that the use of some low sulphur fuels lead to increased PN concentrations at lower engine loads (Anderson et al., 2015), which stresses the importance for regulation specifically addressing particulate matter (PM).
The majority of emitted PM is in the ultrafine size range, < 0.1 µm, which have been demonstrated to have a particularly significant impact on health and the environment (WHO, 2013).However, due to the lack in regulation, ultrafine particles, in terms of PN concentration, emitted from ships have remained unassessed in real world conditions.Quantifying PN concentration is critical to improve our understanding of shipping's impact on health and climate (Chen et al., 2005;Cooper, 2001;Corbett and Farrell, 2002;Isakson et al., 2001;Williams et al., 2009;Reda et al., 2015;Mueller et al., 2015;Anderson et al., 2015;Blasco et al., 2014;Ristovski et al., 2012;Corbett et al., 2007b).To achieve this, wide-scale evaluation of ship emission factors (EFs) is necessary.EFs are commonly expressed as the amount of pollutant (x) emitted per unit mass of fuel consumed g(x).(Kg fuel) -1 .Different methods have been used to investigate ship EFs, including laboratory test-bench studies, on-board measurements, and measurement of ship emission plumes.
Test-bench studies (Reda et al., 2015;Mueller et al., 2015;Anderson et al., 2015;Petzold et al., 2010;Petzold et al., 2008;Kasper et al., 2007) have been used to characterize emissions from different engines at various loads in laboratory conditions.However, engine performance and emissions have been shown to be different in real world operations when compared to laboratory studies.This calls for measurements of ship emissions in-situ to collect reliable data for EF calculations (Blasco et al., 2014;Murphy et al., 2009;Agrawal et al., 2008).To date, only a few studies have been undertaken on-board ships to calculate real emission factors (Juwono et al., 2013;Hallquist et al., 2013b).This is attributed to the prohibitive costs and time commitments of setting up and maintaining on-board measurement equipment on commercial ships.Airborne ship plume measurements (Westerlund et al., 2015;Schreier et al., 2015;Pirjola et al., 2014;Cappa et al., 2014;Beecken et al., 2014;Balzani Lööv et al., 2014;Berg et al., 2012;Lack et al., 2009;Lack et al., 2008;Sinha et al., 2003) offer an alternative method of in-situ measurements without requiring on-board monitoring stations.In the past the deployment cost of these systems, and the risks associated with manned aircrafts have limited their feasibility.However, this has recently changed with the rapid advances being made in commercially available Unmanned Aerial Vehicle (UAV) technology.
Hexacopter UAVs have seen a wide scale increase in industry and research applications due to their ease of use and comparatively low cost (Malaver Rojas et al., 2015;Gonzalez et al., 2011;Brady et al., 2016).Used in conjunction with air monitoring equipment, these systems provide, for the first time, the ability to perform relatively simplistic and cost-effective airborne measurements of ship emissions.However, to date no studies have reported the use of a UAV system capable of collecting data to calculate the EF of PN concentration for ships at sea.This research utilized a customized hexacopter UAV carrying instruments for PN concentration and CO2 measurements to derive EFPN.The UAV system was deployed from the RV Investigator research vessel while at sea.Autonomous measurements of the RV investigators exhaust plume were taken over several flights at various altitudes and distances from the ship.Data collected was used to optimize the sampling flight path and successfully quantify the RV investigators EF for PN concentration.

Methodology and Measurement system
Measurements were conducted as part of the research voyage "The Great Barrier Reef as a significant source of climatically relevant aerosol particles" aboard the RV Investigator research vessel over a two day period of the 13 and 14 October 2016 (day 1 and day 2).Measurements of PN and CO2 concentration emitted by the RV Investigator were taken using a PN and CO2 monitor mounted on a customized DJI EVO S800 hexacopter UAV (DJI, 2014).

The RV Investigator and the voyage
The RV Investigator is an ocean research vessel configured to enable a wide range of atmospheric, biological, goescience and oceanographic research.The vessel is 94 m long, has a gross weight of 6,082 tons, a fuel capacity of 700 tons of ultra-low sulphur diesel fuel.It is powered by three 9 cylinder 3000 kW MaK diesel engines, each coupled to a 690V AC Generator.
Ship propulsion is achieved using two 2600 kW L3 AC reversible propulsion motors powered by these generators.The RV Investigator can host up to 30 crew members and 35 researchers for a maximum voyage period of 60 days with a maximum cruising speed of 12 knots.
A suite of instrumentation for atmospheric research is available on the RV Investigator.This includes a radar system capable of collecting weather information within a 150 km radius of the vessel, and instruments measuring: sunlight parameters; aerosol composition, particle concentration and size distributions; cloud condensation nuclei; gas concentrations; and various other components of the atmosphere.These instruments are housed inside two dedicated on-board laboratories for aerosol and for atmospheric chemistry research.An atmospheric aerosol sample is continuously drawn into the laboratories for analysis through a specialized inlet fitted to the foremast of the ship.Of particular interest to this study, the ship contains a PICARRO (PICARRO Inc., Santa Clara, California, USA) G2401 analyser (Inc., 2017) that continuously measures CO2, CO, H2O and CH4.It has an operation range between 0-1000 ppm and a parts-per-billion sensitivity (ppb) for CO2.
The two day UAV measurement study was possible as part of the RV Investigator voyage "The Great Barrier Reef as a significant source of climatically relevant aerosol particles", which started in Brisbane on the 28 th of September 2016.The ship was used as both: a floating platform to allow launch and recovery of the UAV system; and as the source of an exhaust plume measured by the UAV system for EF calculation.During a several day stationary period on the Great Barrier Reef off the coast of Australia, it was possible to measure the ship plume under stable real world conditions over two consecutive days.One of the three ship engines was maintained at a steady engine load of 25 -30 % of the maximum engine power during all measurements.

UAV system
Measurements of PN and CO2 concentrations in the ship plume were performed using two commercial sensors mounted onboard a hexacopter UAV.The UAV used (Figure 1) is a composite material S800 EVO manufactured by DJI (DJI, 2014).The UAV is 800 mm wide and 320 mm in height, with an unloaded weight of 3.7 kg.Minimum and maximum take-off weights are 6.7 kg and 8 kg, respectively.The UAV contains a 16000 mAh LiPo 6 cell battery, which provides a hover time of approximately 20 min when operating at minimum take-off weight.The telemetry range of the UAV is 2 km, which was adequate to cover the desired sampling area (See Figure 2).
The payload consisted of a PN concentration and a CO2 monitor mounted on-board underneath the UAV.Careful placement of the payload was required to prevent flight issues caused by an altered centre of gravity.Also included was a carbon fibre rod, which extended outward horizontally from the UAV.The sampling lines for the monitors were attached to the end of this rod to ensure that measurements were not affected by the downwash of the UAV rotors.The total weight of the payload was (1.2 kg), which allowed the UAV system to fly for 12-15 min before landing at the home point (A) (See Figure 2).
The S800 was used in conjunction with the DJI Wookong autopilot.The software provides an intuitive and easy to use interface where autonomous flight paths can be planned, saved, and uploaded into the UAV.In addition to this, the ground station allows for continuous, real-time monitoring of the status of the UAV during operation; which includes its longitude, latitude, altitude, waypoint tolerance and airspeed.
The DJI S800 was chosen for this study because it is designed to operate under the 20 kg all up weight (AUW) class of UAV.
This reduces operational costs and avoid subjection to the tighter regulations of larger platforms.Small UAV cannot be operated above any person, or closer than 30 m of populated areas, houses and people.Furthermore, current Civil Aviation Safety Australia (CASA) regulations restrict the use of small UAV (2 and 20 kg) to visual line-of-sight daylight operation, with a maximum altitude of approximately 120 m and within a radius of 3 nmi of an airport.UAVs in this category are not permitted for research unless the research institution has been granted a permit exception.These exceptions can be granted if the institution in question has or collaborates with an UAV operation team who must have: an experienced UAV pilot who is also radio controller specialist; a license for commercial UAV operation; and appropriate liability insurance (NPRM 1309OS -Remotely Piloted Aircraft Systems).Queensland University of Technology (QUT) has an unmanned operator certificate and four pilots who have UAV controller licenses.

Instrumentation for PN concentration
This study measured PN concentration using a Mini Diffusion Size Classifier (DISCmini), developed by the University of Applied Sciences, Windisch, Switzerland (Fierz et al., 2008).The DISCmini is a portable monitor used to measure concentration of particles in the 10-500 nm diameter size range, with a time resolution of up to 1s (1 Hz).It can measure PN concentrations between 10 3 and 10 6 N/cm # .Measurement accuracy is dependent upon the particle shape, size distribution, and number concentration.The advantages of using the DISCmini are its relatively small dimensions (180 x 90 x 40 mm), low weight (640 g, 780 g with the sampling probe, Figure 1) and long battery life of up to 8 hrs.These characteristics allow it to be easily integrated on the UAV.

Instrumentation for CO2 concentration measurements
A TSI (TSI, Shoreview, Minnesota, United States) IAQ-calc 7545 model was chosen to measure CO2 concentrations.Its sensor is based on a dual-wavelength NDIR (non-dispersive infrared) with a sensitivity range between 0 to 5,000 ppm and an accuracy of ± 3.0% of reading or ± 50 ppm (whichever is greater).The measurement resolution is 1 ppm with a maximum time resolution of 1s.Similar to the DISCmini, the advantages of using the IAQ-calc are: its small dimensions (178 x 84 x 44 mm); low weight (270 g, with batteries, significantly lower than the DISCmini), and a battery life of 10 hours.
The readings of the IAQ-clac for CO2 were compared with those measured by the on-board PICARRO G2401 analyser.
Both the DISCmini and the IAQ-calc were tested and calibrated in the on-board laboratory using ambient aerosol measurements at sea prior to the commencement of the measurements (Figure S2 in the Supplementary Material).All data were logged with a 1 s time interval.

Meteorological data
Meteorological data (including air temperature, relative humidity, atmospheric pressure, wind speed and direction) were recorded by the RV Investigators on-board instrumentation during the entire voyage with a 60 s time interval, 24/h a day.

Study design
During the two measurement days of this study, the vessel was heading into the wind whilst idling the UAV missions at sea.
This positioning caused the exhaust plume to extend downwind, directly behind the ship.The UAV system was launched off the back deck, autonomously sampling at varying altitudes and distances into the downwind plume.Flight speed of the UAV was 1.5 m/s, the minimum for the S800.
Day 1 was used to optimise the study design, focusing on finding the flight path most suitable to capture the ship plume.were fitted on the underside of the UAV at the beginning of each measuring day.Five flights were performed across the two measurement days, providing a total of 27 horizontal transects perpendicular to the ship's exhaust plume.

Emission factors
The calculation of an emission factor for particle number concentration (EFPN) from the collected ship plume measurements was performed using Eq. ( 1).This method has previously been used for ship (Westerlund et al., 2015), road vehicle (Hak et al., 2009) and aircraft (Mazaheri et al., 2009) emissions.The measured values of PN concentration were related to the amount of fuel consumed by the engine in question through the use of the simultaneous measurements of CO2 concentration taken by the UAV.This was achieved by using a published value for a ship emission factor of CO2 (EF &'( ) of 3.2 Kg CO2 (Kg fuel) -1 (Hallquist et al., 2013b;Hobbs et al., 2000) .
x  /01 (1) The ∆ and ∆ in Eq. (1) represent the maximum particle concentration change above background in the measured particle number and CO2 concentrations, respectively.The DISCmini measurements were corrected against a reference CPC.
For each transect data series of PNC and CO2, the averaged background concentration were subtracted from the peak data corresponding to measurements inside the plume.The corrected peak data series were then fit with a Gaussian curve using the inbuilt Matlab curve fitting application.The least absolute residuals (LAR) condition was used as this most closely fits the curve to the highest magnitude data points in the series.The maximum peak height of the fitted Gaussian curves were used as ΔPNC and ΔCO2 in the calculation of emission factors for each transect.

Meteorological and Investigator data
Wind conditions were very stable during both day 1 and day 2, following one main pattern for the entire flight time.The wind speed ranged from 3 -13 m s-1.The wind direction was predominantly from the NE during day 1 and ESE during day 2. The wind rose graphs in Figure 3a and 3b illustrate the wind data recorded with the on-board weather instrumentation during all horizontal transects flown during day 1 and 2 respectively.The prevalent wind direction was ESE, which corresponded to the heading of the RV Investigator (indicated by the rose triangle).The wind direction changed occasionally to E during the flight, causing the UAV to fail to capture the RV Investigator plume during some transects.As a result, 2 of the 8 horizontal transects collected on day 2 were excluded from the analysis.

UAV system horizontal transects inside and outside the plume
The UAV system acquired data for a total of 27 horizontal transects for day 1 and day 2. Data were collected at altitudes between 25 m and 65 m above the water surface.During day 1 the plume was captured once when the UAV was at 25 m 8 altitude and 20 m downwind of the ship; and again at both 25 and 35 m altitude 100 m downwind of the ship.These observations lead to the optimized flight used on day 2, which started downwind at 25 m above the surface and 20 m behind the ship.On day 2 the UAV system successfully captured the plume during 6 of the 8 transects performed.Across the two days this lead to a total of 9 transects that captured the plume and which have been considered for discussion, shown in Table 1.
Table 1 -  The PN concentration profiles for the (a) and (b) transects in Figure 4 show that the concentration varied by five orders of magnitude between the outside and inside the plume, while the CO2 profiles show an increase up to 140 ppm above the background.
The profiles in (c) show that the PN concentration was four orders of magnitude greater inside the plume at 100 m behind the ship and that the CO2 concentration was up to 70 ppm higher inside the plume.measurements show clear, single peaks as the UAV crosses the plume.As a consequence, these transects show a good fit with the corresponding Gaussian distribution curves with R 2 values of above 0.9 for both PNC and CO2.In contrast Figure 4 (c) shows substantially less defined, wider peaks with lower pollutant concentrations.This is attributed to a difference in flight paths, with Figure 4 (c) representing data from a transect 100 m behind the ship.The additional time between emission and sampling has allowed the plume to broaden, become less homogenous, and take on a skewed cross-section.This results in a significantly lower R 2 value for the fitted Gaussian curves, with a value of 0.4998 for the CO2 data in this transect.Therefore, whilst the 100 m transect does provide more data points inside the plume, the randomized variations inside the plume lead to less accurate calculations of emission factors.
Of further note in Figure 4, the maximum PN concentrations measured in (a) (7.5x10 5 #.cm -3 ) is approximately three times greater than those in (b) (2.4x10 5 #.cm -3 ) and the CO2 concentrations in (a) are 43 ppm greater than (b).The transect flight plan and ship engine load remained constant throughout these measurements.The variations between (a) and (b) are attributed to several factors which reduce the effectiveness of the UAV transect for capturing the plume.Slight changes in ambient conditions such as temperature, wind direction and intensity will alter the path of the plume as it moves away from the ship.
The UAVs automated flight path cannot account for these variations.Therefore, the degree to which the UAV enters the plume, and thus the concentrations it measures, will be different on each transect.Both CO2 and PN concentration measurements will be similarly affected by this variance.However, differences in instrument response rates in conjunction with these variances will be one of the major contributors to variations in calculated emission factors.

PN Emission Factors
Table 2 shows the distance and altitude of each transect, the R 2 values of the fitted Gaussian curves for PNC and CO2 data, the calculated values of ∆PNC and ∆CO2, and the calculated EFPN.The calculated EFPN values for the RV Investigator ranged from 1.15 x 10 15 to 1.73 x 10 16 #.Kgfuel -1 .The two 100 m transects provided the worst Gaussian fits as well as the highest and lowest calculated emission factors.This indicates that it is important to filter out transects with data which does not fit the expected Gaussian distribution suitably as they can generate significant error.To this end, the 100 m transects were excluded from further analysis.∆PNC and ∆CO2 values for remaining transects were plotted against each other as shown in Figure 5. Relying on a single sample point leads to the potential for random instrumentation effects heavily biasing results in a way which does not strongly impact the R 2 values of Gaussian fits used to identify successful transects.Therefore, it is unclear whether this is a variation in the ship emissions or an instrumentation error.

Instrumentation Limitations
Lightweight UAVs present an opportunity to achieve aerial measurements at significantly less upfront and operational costs than fixed wing and manned aerial vehicles.Lightweight UAVs can be deployed faster with limited or no required launch and landing area compared to their manned and fixed wing counterparts.Yet, their primary disadvantage, particularly in this application, is a severely limited payload weight.To overcome this limitation, this project used the lightweight and portable DICSmini and IAQ-calc sensors.However, these instruments have lower sensitivities and greater uncertainties when compared to a high accuracy CPC and CO2 monitor for measurements, which can influence results.
The DISCmini has a manufacturer listed measurement cut-off size of 10 nm.A previous study listed in Table 3 (Lack, Corbett et al. 2009) shows that the cut-off size of instruments used to measure PNC is directly linked to the value of EFPN, with the measured EFPN doubling when the cut-off size is changed from 13 nm to 5 nm due to the large number of particles in this size range.This may have been another contributing factor to the EFPN measured in this study being in the lower end of measured values in literature.
The two 100m transects were not accounted for in the final calculation of EFPN due to their poor Gaussian curve fits.Whilst this has been attributed to the skewing of the plume at this distance, the limitations of the instrumentation could also have contributed.The lower concentrations of CO2 at this distance result in the difference above background inside the plume being the same order of magnitude as the manufacturer specified error margin.Hence, the variability in the plume either side of the central peak as shown in figure 4 (c) could be due in part to instrumentation error.
Calibrations of sensors in this study were performed by comparison with reference instruments for ambient measurements at sea.Ideally, calibration should be performed with in-plume measurements, however it was not possible to access the plume with reference instrumentation on board the ship.Whilst this study provides a successful proof of concept with consistent results over multiple days and flights, a validation study is needed.This should include independent measurements of EFPN using other established methodologies to ascertain more precise correction factors and uncertainties.

Summary and conclusion
The UAV system used in this study successfully measured PN and CO2 concentrations from the exhaust plume of the RV Investigator whilst operating at sea.Several different flight paths were tested and an optimal transect flying perpendicular to the plume at a distance of 20 meters from the ship was adopted.The EFPN calculated for the RV investigator 7.6 ± 1.4 x 10 15 #.kgfuel -1 at a constant 30% engine load.This EFPN was in agreement with values reported in literature, indicating this novel UAV system has potential for EFPN quantification pending further evaluation.
In comparison with other methods, the UAV system presented provides a cost effective and accessible solution for the rapid measurement and quantification of ship EFPNs.Its ability for deployment both in harbour and at sea, coupled with the possibility of altering its flight path to account for variances in wind conditions; gives this UAV system a distinct advantage over ground based and manned aerial vehicles.Furthermore, the UAV can sample considerably closer to the plume emission source than other methodologies, providing higher concentration measurements for the calculation of EFPN.
Whilst further validation is necessary, results present here indicate that this UAV system has the potential to be used a low cost tool for quantification of ultrafine particle emission factors from commercial shipping.This is critical to improve our understanding of shipping's impact on climate and health.

Recommendations
The potential of this UAV system extend far beyond what is described here.This study is intended as both: a proof of concept; and to provide useful information both for the future of this project, as well as any other UAV sampling systems being developed.The most significant improvement to the method described would be the use a UAV with a lower minimum airspeed.This would allow for more data points per transect and would minimize the impact potential outliers in instrumentation data.Other related improvements to this include: the use of different sensors with higher response rates; and additional flightpath investigations to find an optimal transect distance which provides the broadest plume cross-section, without the plume becoming distorted and impacting accuracy.
Further optimization of the transect approach is also possible.After location of the plume the system could be set to make several repeat passes across the plume in rapid succession to increase the sample size.Another alternative would involve the UAV hovering inside the plume over a period of time collecting a continuous series of measurements from the centre of the plume.These methods would both require real time sensor feedback to the UAV pilot and potentially adaptive autonomous controls to achieve a suitable result.This methodology could also be expanded to measure other important ship emission factors, including NOX and volatile organic compounds (VOCs).
Investigator as well as the on board MNF support staff as without their support and effort this research would not have been possible.

Figure 1 .
Figure 1.The UAV system with the on-board instrumentation: the DISCmini and the IAQ-calc.

Figure 2
shows the programmed flight path, which consisted of a continuous flight beginning at a distance (D) and from an altitude (H) above the surface.Point A, located on the back deck of the RV Investigator, represents the 'home point'.In UAV terminology this refers to the position where the UAV system takes off and lands.The UAV system was programmed to move horizontally by a distance (2d), perpendicular to the ship, then climb vertically for 10 m (h) before flying in the opposite horizontal direction for the same distance (2d).The UAV was then programmed to climb another 10 m (h) before repeating this pattern until the UAV reached an altitude of 65 m above the ocean.During day 1, the UAV system followed three different flight paths, each one with both a different distance D behind the ship (20, 50 and 100 m), and a different horizontal distance 2d (50, 100 and 150 m).The optimised flight path for day 2 started 20 m behind the ship and 25 m above the surface, with no altitude variation.The UAV path was limited to a continuous horizontal flight of 50 m (2d) at steady speed of 2 m s -1 .This path and flying speed allowed up to 4 horizontal transects to capture the ship plume.

Figure 2 .
Figure 2. Flight path used to capture the plume: H -height from the ocean, D -distance behind the ship to the flight beginning point, h -rising altitude after the horizontal transect, 2d -full length of the horizontal transect 2.5.Experimental procedureThe UAV can fly either manually or autonomously.As a safety precaution, every take-off and landing was performed using the manual flight mode.Once in the air, the UAV was switched to autonomous flight mode, allowing the platform to follow the pre-programmed flight path discussed in the previous section.The flight path consisted of waypoints, which are three-

Figure 3a -
Figure 3a -Wind rose showing wind speed and direction during day 1. Rose triangle shows RV Investigator direction during the measurements.

Figure 3b -
Figure 3b -Wind rose showing wind speed and direction during day 2 optimized flight.Rose triangle shows RV Investigator direction during the measurements.
Specifications of the transects considered for the data analysis.The (*) indicates the transect of Day 1 of which PN concentration and CO2 profiles are presented in Figure 4.

Figure 4 -
Figure 4 -(a) and (b) show the measured PN and CO2 concentration profiles and fitted Gaussian curves for two different transects 20 m behind the ship 25 m above the surface during day 2. (c) shows the PN and CO2 concentration profiles and fitted Gaussian curves collected during flight 3 of day 1 at 100 m behind the ship, 25 m above the surface.

Figure 4
Figure 4 (a) and (b) both show transects at 25 m altitude and 20 m behind the ship.Both the PN concentration and CO2

Figure 5 -
Figure 5 -(a) ∆PNC against ∆CO2 with 95% confidence interval for the six transects considered for the data analysis. (b) ∆PNC against ∆CO2 with 95% confidence interval with the removal of the outlier transect from the first flight of day 2