Next Article in Journal
Logistics Village Location with Capacity Planning Problem, an MILP Model Approach
Previous Article in Journal
Life Cycle Assessment of Greenhouse Gas Emissions from Recycled Asphalt Pavement Production
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 15
Issue 5
10.3390/su15054632
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Evaluation of the Potential of Sentinel-5P TROPOMI and AIS Marine Traffic Data for the Monitoring of Anthropogenic Activity and Maritime Transport NOx-Emissions in Canary Islands Waters

by
Manuel Rodriguez Valido
*,
Javier Perez Marrero
,
Argelio Mauro González
,
Peña Fabiani Bendicho
and
Carlos Efrem Mora
Department of Industrial Engineering, University of La Laguna, 38200 San Cristóbal de La Laguna, Spain
*
Author to whom correspondence should be addressed.
Sustainability 2023, 15(5), 4632; https://doi.org/10.3390/su15054632
Submission received: 24 January 2023 / Revised: 22 February 2023 / Accepted: 27 February 2023 / Published: 5 March 2023
(This article belongs to the Section Environmental Sustainability and Applications)
Browse Figures
Versions Notes

Abstract

:
This work studies air quality by analysing NOx emissions in the inland waters of the Canary Islands, with particular emphasis on determining how maritime transport activity contributes to the emission of NO2 in the environment of the two main islands, Tenerife and Gran Canaria. We explored the capabilities of tropospheric NO2 density derived from the TROPOMI sensor onboard ESA’s Sentinel 5P Satellite to be used as an air quality monitoring tool at the regional scale of the Canary Islands. The studied mesoscale emission scenarios allowed us to identify the main sources, associated with urban areas, heavy roads, power plants, ports, and to a lesser extent, shipping routes. Mean values for the metropolitan area of Santa Cruz de Tenerife were 1.38 × 1015 molec cm−2. Similarly, in port areas, mean values of 1.22 × 1015 molec cm−2 were found. These levels can confidently be associated with anthropogenic activities. These were clearly distinguishable from background (noise) values of 7.08 × 1014 molec cm−2 obtained in maritime areas away from the influence of the islands. To investigate the maritime contribution to the NO2 emissions, ship tracks were obtained from an Automated Identification System (AIS) receiving station that covered the channel between the Tenerife and Gran Canaria islands. Multitemporal, and hence accumulative, NO2 scenarios were compared with the ship traffic density within a given temporal window before satellite overpass. We found good spatial agreement between NO2 signal and frequent ship routes between the major islands at several time scales, particularly in weekly averaged scenarios. Enhancements up to 2.0 × 1015 molecules cm−2 relative to surrounding waters were identified in the middle of the main shipping routes between the main islands. Thus, multitemporal NO2 scenarios derived from TROPOMI can lead to an estimate of the ship traffic contribution to NOx emissions in complex environments, such as this one, influenced by land emissions.

1. Introduction

Nitrogen dioxide, NO2, is a fraction of all nitrogen oxides, jointly called NOx. These chemicals are produced during the combustion of fuels, resulting in atmospheric pollutants that affect human health [1]. Additionally, NO2 participates in the formation of free radicals, which may lead to the formation of tropospheric ozone, O3, which is another pollutant harmful to health; NO2 also contributes to acid rain [2,3]. Except for volcanic eruptions, wildfires, and specific edaphological processes and lightning (none of which occurred during the studied period), the emission sources of NO2 are mainly anthropogenic in origin [4]. This variety of chemical processes results in a relatively short atmospheric NO2 persistence. The atmospheric residence time of NO2 is assumed to be a few hours, particularly under sunlight conditions [5]. Consequently, observed variability in NO2 concentration is detected more clearly in the proximity of its primary sources [6]. Therefore, satellite imagery is an excellent tool for studying mesoscale and synoptic NO2 behaviour, such as frequent ship routes as emission sources. High spatial resolution satellite-derived NO2 measurements provided by the Sentinel-5P TROPOMI sensor can thus be used to estimate the aggregated effects of myriad point emission sources that evolve in the low troposphere, producing a synoptic mesoscale picture of the study area. Previous studies have used TROPOMI data to characterise emissions in large cities or countries [7,8,9]. Spotlighted precedents linked to this work have characterised the effects of the COVID-19 lockdown on industrial emissions worldwide: [9,10,11,12,13,14,15]. Regarding the specific question of maritime operations, there are fewer examples in the literature that use data from TROPOMI precursor sensors, such as SCIAMACHY and OMI [16,17], that possessed lower spatial and radiometric resolutions. Another study compared the satellite signal with emissions model results [18] and they found NO2 enhancement of the order of 8 × 1014 molecules·cm−2 that could be detected by satellite sensors over studied shipping routes. More recently, other studies used data from the TROPOMI sensor and marine traffic data obtained from an automated identification system (AIS) to address the issue of emissions due to maritime traffic [19,20].
Currently, the health of citizens and the environment is a major concern for local and global authorities. In this sense, any contribution to the Sustainable Development Goals (SDG) and the de-carbonization of the economy [21,22] are important. In this context, the studies and developments aimed to reduce dependence on fossil fuels, i.e., alternative fuels, are growing in relevance [23,24].At the same time, in order to support environmental regulations, knowledge-based governance requires an understanding of the sources, processes, and flows of pollutants.
This work is supported by the Smart Ports and Electrification of Inland Traffic in Canary Harbors project funded by the local Canarian government. The project is aimed to improve the sustainability of Canarian ports by making them more energy-efficient, less-polluting, and more competitive, and to improve the quality of life in the densely populated areas close to harbour locations. This study was conducted in the Canary Islands region (Figure 1) [25,26]. This archipelago is formed of eight main islands, representing an area of 7493 km2 that is home to a population of approximately two million stable inhabitants. In addition, ten to twelve million tourists visit the islands annually, using the territory and resources of the region. The main industry is tourism, which implies significant dominance of the tertiary sector and air and maritime transportation in the archipelago economy [27].
The Canary Islands maritime waters are declared a particularly sensitive sea area [28,29]. These waters support a high maritime traffic density because of the archipelago’s position close to important commercial maritime routes that cross it.
In this study, we proposed the use of TROPOMI data together with ship activity records to estimate (monitor) the contribution of marine traffic to NO2 pollution, thus addressing the distribution of NO2 emissions and their variability at the scale of the Canary Islands. This study tried to identify primary sources and evaluate their relative importance. In particular, this work was focused on the estimation of the contribution of maritime-marine operations to the NO2 balance in the area.
The possibility of using a multitemporal approach to contribute to unravelling the maritime component of NO2 emissions was explored. This is of great importance in situations where direct detection of plumes from ships is not possible due to the influence of terrestrial emissions. Using TROPOMI as a monitoring tool, characterised by its large coverage and high repetitivity, would lead to improved governance [21,22] and can also be used to complement ship-based emission models [23,24].

2. Material and Methods

Quantifying the contribution of NO2 of the maritime industry requires identifying all possible other contributors, including those from static sources (e.g., power plants and large terrestrial industries, and urban agglomerations) and those from mobile sources linked to transportation. Figure 2 depicts the positions of the main static sources of pollution that are coincident with main cities all over the archipelago. These data were obtained by extracting the standard deviation of the NO2 density during a thirty day period.
There are cases in the literature showing the detection of NO2 plumes produced by ships (especially large vessels) using TROPOMI satellite data [19,20]. However, the conditions of these detections were very restrictive regarding the number and type of satellite passes used (zenith passes and sunglint situations).
The usability of satellite measurements for pollution monitoring on a regional scale requires determining the noise level and anthropogenic contributions to NO2 distributions, and so the maritime component must be calculated afterward. This study takes into account TROPOMI’s dependence on the visualisation angle. NO2 content calculation in the atmospheric column uses a multi-stage algorithm [30] that initially calculates the concentration in the slant column (between the emission and the sensor). This algorithm converts angular densities into normal (vertical) columnar densities by applying an atmospheric air mass correction factor (AMF). The application of this factor is the most controversial aspect in determining NO2 from satellite data [31].
We approached the problem of ship emissions in a multi-temporal manner by integrating satellite data over short-period observations (1–15 days) and comparing them with the maritime traffic within the channel that connects the central islands, which is the area covered by the AIS station (Figure 3). This area has density, high-frequency traffic between the islands of Tenerife and Gran Canaria, mainly occupied by large ferries and, to a lesser extent, inter-island cabotage vessels. This type of traffic was predominant during the time window before the satellite passed.
Wind conditions must be considered when dealing with atmospheric constituents. In this work, this task was accomplished using hourly averaged data from surface (10 m) winds obtained from the Copernicus reanalysis ERA5 database. The average wind vectors during the hour before the satellite overpass were extracted.
In order to assess the operability of satellite-derived data for monitoring NO2 emissions at the local scale, we searched for “relatively long” periods of low cloud cover in the Canary Region based on a quick-look revision available at Mundi Web Service [32]. A total of 110 satellite scenes were selected from January to April during the years 2020 and 2021. This includes slightly more than 50% of the total potential observations for that period. Satellite measurements that had low quality index values (qa < 0.75) were rejected from the analysis, following the recommendations of the Algorithm Theoretical Basis Definition (ATBD) and the User’s Manual of Products Level 2 [30,33].
The TROPOMI sensor [34], installed on ESA’s Sentinel 5 Precursor platform of the ESA, began its operational phase in April 2018. This instrument can measure the exoatmospheric, irradiated, and reflected radiation at different wavelengths in the UV-VIS (270–495 nm), NIR (675–775 nm), and SWIR (2305–2385 nm) regions. With these characteristics, TROPOMI measures the concentrations of O3, NO2, SO2, CH4, CO, HCHO, aerosols, and clouds with high spatial and temporal resolution [35] (Loyola et al., 2018). The TROPOMI sensor is a push-broom sensor with an angular field-of-view of 108°, equivalent to a footprint on the ground of approximately 2600 km in a direction perpendicular to the flight of the satellite. It has 450 multispectral detectors. This configuration produces a spatial resolution of 3.5 × 5.5 km at the nadir, decreasing progressively with the viewing angle [33]. The estimation of total and tropospheric NO2 is based on differential spectral absorption techniques (DOAS) [16]. It makes use of the approximation DOMINO [36], for which it is necessary to use a pre-calculated factor of atmospheric air mass (AMF), together with assimilated data from chemical transport models, as well as auxiliary data established in the European Quality Assurance for Essential Climate Variables (QA4ECV) project. This study used level 2 geophysical data files obtained from Copernicus. The characteristics of these data files are defined in the TROPOMI ATBD of the total and tropospheric NO2 data products [33] available on public data servers of the European Copernicus program [32].
To assess the contribution of maritime operations and ship traffic emissions at the local scale, nearly continuous records of ship positions registered from a locally operated AIS receiving station were used. This station monitors mainly the channel between the Gran Canaria and Tenerife islands, where most of the marine traffic of the inland waters of the Canary Archipelago occurs.
Positional data corresponding to the identification of vessels in the study area were obtained from an AIS station installed at the Faculty of Physics of the University of La Laguna (28°28′40.18″ N, 16°14′37.06″ W). The AIS station samples every 3 s for vessels equipped with an AIS transmitter within the coverage of the receiving station (Figure 3). From the AIS records, we extracted latitude, longitude, vessel identifier, time, and date. When comparing ship data to TROPOMI data, a temporal filter was established due to the short lifetime of NO2 in the atmosphere. Eventually, only the data corresponding to 1.5 h prior to the satellite overpass were considered in this study. The maritime traffic data were georeferenced and overlaid (geographically and temporally) on the grid defined for the satellite data.
Comprehension of tropospheric gas behaviour requires knowledge of wind conditions at the time of satellite observations. In this study, we drowned the Copernicus reanalysis RA-5 data [37]. From which, hourly wind vectors coincident with the satellite observations were extracted and added to the integrated data framework.
The behaviour of the emitted gases (dilution, diffusion, advection, and extinction) depends mainly on the wind factor. Hourly wind data that coincided with the study area and period were selected for this study. Wind data were retrieved from the climatological collection of ERA5 hourly wind data of wind at 10 m georeferenced and overlapped on the same reference frame defined for satellite data.
In Figure 4, wind vectors are over plotted on a TROPOMI scene using wind barbs, showing wind intensities below 5 ms−1 in the channel between the central islands. Furthermore, ship traffic data were superimposed over the corresponding NO2 distribution. An integrated reference framework allows accounting for the effects of the sensor visualisation angles. The example shown (Figure 4) supports the idea of reverting AMF producing more sharp scenes, as pointed out by [19].
Data processing routines used were developed in Python; the scripts include proven methods for treating geospatial information such as bilinear interpolations, accumulations, histograms, standard deviation, average, AIS and wind data alignment, etc. The radiometric values: NO2 content, AMF and qa values, and associated metadata X were extracted from the Sentinel Level 2 files for the studied area. The extracted data matched the entire Canary Islands in the window 27°N to 30°N and 10°W to 20°W.
The application of the TROPOMI observations to the operational monitoring of NO2 emissions must also consider the considerable variation in ground resolution with the angle of view and from one day to the next. Furthermore, the algorithm that determines the NO2 amount is strongly dependent on the viewing angle. Depending on the viewing angle, the weekly cycle of Sentinel 5P overpasses in this region provides three zenith view scenes (Figure 5, panels b, c and g); two passes with intermediate angle view (Figure 5, panels a and d); and two passes with oblique view (Figure 5, panels e and f). The analysis system developed considers these circumstances in the data processing.
The methodological contribution of this work is the development of computer tools and libraries in Python that allow the coherent integration of data from various sources including NO2, AIS-derived ship traffic, and wind data on a common dynamic cartography; a characterisation of the spatio-temporal variations in the NO2 signal; an assessment of the effects of the viewing geometry on the integration of the signal; statistical operations (central, dispersion, and correlation measurements); and an analysis of the time series.
Satellite observations were resampled over an equiangular cylindrical mesh grid with a spatial resolution equivalent to that of the TROPOMI sensor (3.5 × 7 km). Quality flags and high-relocation RMSE errors were excluded from this process. This resampling process is the usual method when observations whose spatial resolution varies over the days are available [11,13]. The result or data structure is a multiparametric three-dimensional array A = [x, y, t, P1 ..., Pn] where x is longitude, y is latitude, t is the observation time, and P1,...Pn are the observed properties, such as: NO2, ship presence, wind vector, and sensor viewing angle.
Once aligned on a spatial and temporal frame of reference, these data were analysed by applying statistical measures of centrality and dispersion, and linear regression analysis in the spatial and temporal dimensions, as well as image analysis.

3. Results

In the Canary Islands, cloud cover is generally abundant, although they tend to have periods of relatively low cloudiness between January and March due to the weakening of the trade winds.
Regarding the characterization of the noise level and anthropogenic variability of the NO2 content in the region, Figure 6 shows the averaged time series of NO2 for the period spanning January 1st to 30 April 2020. It was performed at selected, representative locations (Table 1). The averaged value for the Santa Cruz metropolitan area (blue line), harbour area (orange line), and oceanic control area (green line) are presented in Figure 6. Measures over land and harbour areas were significantly higher, by an order of magnitude, than those over the oceanic control areas. The labour cycle is evidenced in the urban areas (green line) with four peaks found from Julian day 14 to 45, approximately. From 45 to 50 Julian days, a drop in NO2 was observed, coinciding with the “Carnival holiday”. A remarkable decrease in both metropolitan and harbour areas was observed after the COVID-19 lockdown, which occurred on Julian day 79 (19 March) of 2020. A drop in peak values from day 75 onwards can be observed. Although a longer time series would be required, monthly distributions before and after the lockdown, presented in Figure 7, also support the idea that this reduction is actually due to the effects of the lockdown. Control areas far from the islands’ influence (red line) were used to depict the background or noise levels. Consequently, this variability is not associated with anthropogenic activities [38]. Table 2 summarises the statistics associated with these measurements.
The variance found in metropolitan areas corresponds to approximately 60% of the total variance in the sample. The port area variance contributed 28% to the total variance, whereas that of the oceanic control area accounted for approximately 12% of the total variance of the series. The first two correspond mainly to anthropic activities, and the third corresponds mainly to other causes.
The effects of the reduction of activities caused by the COVID-19 lockdown (see Table 3) show a sharp decrease in the mean values of NO2 from one month before the lockdown (19 March 2020–Julian Day 79) to one month later.
The monthly average drop was 41% in the metropolitan area and 7% in the maritime zone. The control area grew by 5%. This reduction of more than 40% in the metropolitan area overlaps, in general terms, with one reported in various articles in the scientific literature [11,13,31,39]. Figure 7 shows these changes in the spatial context.
The NO2 spatial distributions were analysed by averaging at various time scales. This analysis allowed us to study the influence of adding oblique passes to determine the NO2 concentration and how time scales affect the distribution of average values. We finally discarded the most oblique passes, accepting types A and B (Figure 5).
As expected, the spatial distribution found in NO2 concentration measurements in the Canary Islands showed a clear association with the population centres and areas of higher human activity on the islands. The average image (Figure 3) shows the effects of socioeconomic development (anthropization) on NO2 emissions from the main population centres, roads, industrial areas, harbours, and power plants. The maritime channel separating the two central islands also has high NO2 values. Figure 8 shows the effect of time scales on the averaging process. Increasing the timespan reduced the noisiness of the images because of the cancellation of random errors. Nevertheless, the relative importance of the maritime channel between Tenerife and Gran Canaria can be observed, regardless of the time scale considered. Enlarging the time span reduces the averaged values, given that point sources change their positions randomly.
In many cases, the concentration found in the central part of the channel exceeds the values found in the areas adjacent to the coast, which implies that not all of the NO2 detected in the channel comes from advection from land. Therefore, maritime activities play a non-negligible role in the NO2 balance in the area.
The maritime traffic distribution was superimposed on the NO2 distribution obtained from the satellite data to understand the contribution of maritime traffic to NO2 emissions in the study area. The data were extracted immediately 1.5–2.0 h prior to the passing of the satellite, given the short residence time of NO2 in the atmosphere [5].
The analysis was carried out in two ways. First, it was conducted based on individual passes, for specific dates, as most publications do when detecting emissions from ships. Second, the present study analysed the situations that occur when integrating various time series between 2 and 7 days. For individual passes, this study included the influence of applying the AMF atmospheric correction factor. In the case of multitemporal images, the variable used was always the normalised component of the satellite signal.
There was a high degree of spatial coherence between the average distribution of NO2 over the ocean and the density of maritime traffic just before the satellite overpass. The extent to which the contribution of maritime traffic to an area can be identified is what we want to explore using cumulative and averaged NO2 records.
Several examples of multi-temporal images, with different time spans, are presented in Figure 8. A stable pattern is observed in the spatial distribution of tropospheric NO2, regardless of the time period considered. The main characteristic of this distribution is the preponderance of emissions on the two main islands, which extends beyond the borders of each island. However, there is an enrichment effect in the channel that connects Tenerife and Gran Canaria, including a fork to the port of Agaete in Gran Canaria, which significantly contributes to maritime traffic between these islands.
When comparing these distributions with the maritime traffic observed in a maximum period of 2 h prior to the satellite observation, with the average distribution of NO2 (Figure 9), we observed that this increase in the concentration in the channel is consistent with the traffic density in the area, both on a weekly scale (Figure 9a) and on a 3-day scale (Figure 9b). This indicates that the multi-temporal images behave as valid indicators of the emission behaviour and the effect of emissions produced by high-frequency maritime traffic between the main islands of the archipelago.

4. Discussion

The obtained data structure was used to analyse the spatial and temporal variability of the atmospheric gas content and its relationships with other observations, such as wind conditions or the presence of ships at a given location and time. Moreover, this data structure can easily incorporate other data, such as PM2.5, PM10, O3, or SO2, from in-situ monitoring networks or other satellite products.
In the metropolitan area of Santa Cruz de Tenerife, a long-term mean value of 1.37 × 1015 molec NO2 cm−2 was calculated, while peak values frequently reached 5 × 1015 molec cm−2. These results are comparable to medium-sized urban areas of Texas [40], and are lower than the levels observed for large cities, such as Helsinki [41]. Nevertheless, TROPOMI data are accurate enough to catch the variability on a weekly scale related to work cycles and vacation periods (Figure 4), in concordance with what has been found in other studies [9,41]. These levels are consistent with an economy based on the tertiary sector, with a significant influence from travel and transportation [27]. The contribution of the industrial sector to emissions is low compared to other industrialised areas [9,17]. However, remarkable industrial facilities, such as port areas and power plants, were detected in the standard deviation TROPOMI images (Figure 2). The main source areas were directly related to the economic activities of the islands. They match the metropolitan areas geographically with the main roads or highways located on the eastern facade of each island.
The central islands (Tenerife and Gran Canaria) account for about 80% of the population and economic activities of the archipelago. In the rest of the islands, the sources are clearly identified with the capital areas of each one of them, which in all cases include a port area of relative importance. In the central part of the Tenerife and Gran Canaria islands, with mountains higher than 1500 m above sea level, very low values of NO2 are usually detected. This clearly shows that these gases are produced, mostly, in the lowest layer of the troposphere.
The effects of the pandemic lockdown on NO2 concentrations produced a reduction of approximately 40% in land areas on a monthly scale (Figure 7). This reduction was moderate in the maritime zone during the same period, and comparable to what has been found in similar studies [11].
The levels of NO2 found in the port area, with mean values of 1.22 × 1015 molec cm−2 and maxima of 2.97 × 1015 molec cm−2, were comparable to those found over land and one order of magnitude higher than the oceanic, “non-anthropogenic” background values of 7.1 × 1014 molec cm−2.
Given the proximity of land emissions in the study area, the most relevant difficulty in estimating the component due to ships was the separation of the contribution generated in situ and advected from land at the local level. These difficulties have not been faced in other studies that detect emissions from ships, such as [19], given that in all cases, these authors used oceanic locations very far from the coast and in sunglint conditions, and only for very large cargo vessels. These conditions occur at very few moments throughout the year.
Another limiting factor of the advective–diffusive component is the short half-life of gas. Therefore, it is possible that under low wind speed conditions (less than 5 m/s or 18 km/h equivalent to approximately 2 pixels/hour), a threshold of 4 or 5 pixels of distance from the coast can be established in which the NO2 values found would be essentially generated locally, that is, due to marine emissions.
In this scenario (areas far from the coast), it can be ensured that the NO2 values found in the central part of the channel, which coincide with the positions occupied by the vessels that cover the route between the main islands, are essentially due to emissions from ships. The coincidence of the NO2 “cloud” position with the highest traffic density that we find in the multi-temporal images (Figure 9) would indicate the importance of maritime emissions in this environment. An enrichment in NO2 values above 2 × 1015 molec cm−2 was observed in the central zone of the canal, coinciding with the high-frequency maritime routes between the Tenerife and Gran Canaria islands in weekly aggregated scenes (Figure 8), which is not compatible with mere transport, advection, and diffusion phenomena.
This enhancement in NO2 by ship traffic is comparable to other estimations about ships emissions such as [19,20] or [42].

5. Conclusions

The multitemporal analysis of the satellite-derived spatial distribution of NO2 showed persistent patterns geographically distributed according to the main sources of emissions. In the maritime channel between the Tenerife and Gran Canaria islands, accumulations in the NO2 content were found, indicative of source-type structures, which generally coincide with the occupation of the marine space by ships in the period immediately prior to the satellite pass.
We have verified that the cumulative signal of the sensor over different periods of time reflects a spatial pattern quite like the distribution of the presence of vessels in the Tenerife–Gran Canaria channel. This study proposes an improvement in the potential use of the TROPOMI sensor to monitor, analyse, and develop applications to estimate the emissions inventory associated with shipping. In other words, emission inventory monitoring is a cost-effective tool for promoting improved environmental performance, facilitating public access to information on emissions, tracking developments, demonstrating progress in reducing pollution, checking compliance with certain international agreements, setting priorities, and assessing progress achieved through community, national, and international environmental policies and programs.
This outcome will enhance the usefulness of the TROPOMI data as sustainability indicators for the management of industrial and maritime operations in an area with moderate, mixed-source NO2 emissions.
Although the study was carried out in the Canaries, the outcomes can be applicable in any other location where the same conditions are present, enhancing the possibilities of satellite measurements as monitoring tools that allow implementation of environmental and health regulations.

Author Contributions

Conceptualization, M.R.V. and J.P.M.; methodology, A.M.G., P.F.B.; software, J.P.M., A.M.G.; validation, M.R.V., J.P.M. and C.E.M.; formal analysis, M.R.V. and P.F.B.; investigation, A.M.G., J.P.M. and M.R.V.; resources, A.M.G.; data curation, M.R.V.; writing—original draft preparation, M.R.V., A.M.G. and J.P.M.; writing—review and editing, P.F.B.; visualization, C.E.M.; supervision, M.R.V.; project administration, M.R.V.; funding acquisition, M.R.V. All authors have read and agreed to the published version of the manuscript.

Funding

This work is supported by the projects entitled “Smart Ports and Electrification of Inland Traffic in Canarian Ports” (ProID2020010080) and “Mobile Electric Platform for Training in Marine Technologies and Sustainable R+D Support (PLEAMAR)” (EIS 2021 09), both funded by the Canarian Agency for Research, Innovation, and Information Society.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable

Data Availability Statement

Research data supporting this publication are available from the Family of ERA5 datasets—Copernicus Knowledge Base—ECMWF Confluence Wiki‘s repository located at https://confluence.ecmwf.int/display/CKB/The+family+of+ERA5+datasets [43] (accessed on 1 October 2021).

Conflicts of Interest

The authors declare that they have no affiliations with or involvement in any organisations or entities with any financial interest in the subject matter or materials discussed in this manuscript.

References

  1. Gauderman, W.J.; Avol, E.; Lurmann, F.; Kuenzli, N.; Gilliland, F.; Peters, J.; McConnell, R. Childhood asthma and exposure to traffic and nitrogen dioxide. Epidemiology 2005, 16, 737–743. [Google Scholar] [CrossRef]
  2. Notholt, J.; Hjorth, J.; Raes, F. Formation of HNO2 on aerosol surfaces during foggy periods in the presence of NO and NO2. Atmos. Environ. Part A Gen. Top. 1992, 26, 211–217. [Google Scholar] [CrossRef]
  3. Stemmler, K.; Ndour, M.; Elshorbany, Y.; Kleffmann, J.; D’anna, B.; George, C.; Ammann, M. Light induced conversion of nitrogen dioxide into nitrous acid on submicron humic acid aerosol. Atmos. Chem. Phys. 2007, 7, 4237–4248. [Google Scholar] [CrossRef] [Green Version]
  4. Minallah, N.; Khan, M.N. Remote sensing based analysis of disparity in tropospheric NO2 during COVID-19. Remote Sens. 2021, 10, 10. [Google Scholar]
  5. Liu, F.; Beirle, S.; Zhang, Q.; Dörner, S.; He, K.; Wagner, T. NOx lifetime and emissions of cities and power plants in polluted backgrounds estimated using satellite observations. Atmos. Chem. Phys. 2016, 16, 5283–5298. [Google Scholar] [CrossRef] [Green Version]
  6. Cooper, M.J.; Martin, R.V.; McLinden, C.A.; Brook, J.R. Inferring ground-level nitrogen dioxide concentrations at fine spatial resolution applied to the TROPOMI satellite instrument. Environ. Res. Lett. 2020, 15, 104013. [Google Scholar] [CrossRef]
  7. Barten, J.G.M.; Ganzeveld, L.N.; Visser, A.J.; Jiménez, R.; Krol, M.C. Evaluation of nitrogen oxides (NOx) sources and sinks and ozone production in Colombia and surrounding areas. Atmos. Meas. Tech. 2020, 20, 9441–9458. [Google Scholar] [CrossRef]
  8. Gerrit, K.; Henne, S.; Meijer, Y.; Brunner, D. Quantifying CO2 emissions of power plants using CO2 and NO2 imaging satellites. Front. Remote Sens. 2021, 2, 689838. [Google Scholar] [CrossRef]
  9. Goldberg, D.L.; Anenberg, S.C.; Kerr, G.H.; Mohegh, A.; Lu, Z.; Streets, D.G. TROPOMI NO2 in the United States: A detailed look at the annual averages, weekly cycles, effects of temperature, and correlation with surface NO2 concentrations. Earth's Futur. 2021, 9, e2020EF001665. [Google Scholar] [CrossRef]
  10. Griffin, D.; Zhao, X.; McLinden, C.A.; Boersma, F.; Bourassa, A.; Dammers, E.; Degenstein, D.; Eskes, H.; Fehr, L.; Fioletov, V.; et al. High-resolution mapping of nitrogen dioxide with TROPOMI: First results and validation over the canadian oil sands. Geophys. Res. Lett. 2019, 46, 1049–1060. [Google Scholar] [CrossRef] [Green Version]
  11. Vîrghileanu, M.; Săvulescu, I.; Mihai, B.A.; Nistor, C.; Dobre, R. Nitrogen dioxide (NO2) pollution monitoring with Sentinel-5P satellite imagery over Europe during the Coronavirus pandemic outbreak. Remote Sens. 2020, 12, 3575. [Google Scholar] [CrossRef]
  12. Vitali, F.; McLinden, C.A.; Griffin, D.; Krotkov, N.; Liu, F.; Eskes, H. Quantifying urban, industrial, and background changes in NO2 during the COVID-19 lockdown period based on TROPOMI satellite observations. Atmos. Chem. Phys. Discuss. 2021, 22, 4201–4236. [Google Scholar] [CrossRef]
  13. Jeong, U.; Hong, H. Assessment of tropospheric concentrations of NO2 from the TROPOMI/Sentinel-5 precursor for the estimation of long-term exposure to surface NO2 over South Korea. Remote Sens. 2021, 13, 1877. [Google Scholar] [CrossRef]
  14. Schneider, P.; Hamer, P.; Kylling, A.; Shetty, S.; Stebel, K. Spatiotemporal patterns in data availability of the Sentinel-5P NO2 product over urban areas in Norway. Remote Sens. 2021, 13, 2095. [Google Scholar] [CrossRef]
  15. Zara, M.; Boersma, K.F.; Eskes, H.; van der Gon, H.D.; de Arellano, J.V.-G.; Krol, M.; van der Swaluw, E.; Schuch, W.; Velders, G.J. Reductions in nitrogen oxides over the Netherlands between 2005 and 2018 observed from space and on the ground: Decreasing emissions and increasing O3 indicate changing NOx chemistry. Atmos. Environ. X 2021, 9, 100104. [Google Scholar] [CrossRef]
  16. Richter, A.; Eyring, V.; Burrows, J.P.; Bovensmann, H.; Lauer, A.; Sierk, B.; Crutzen, P.J. Satellite measurements of NO2 from international shipping emissions. Geophys. Res. Lett. 2004, 31, L23110. [Google Scholar] [CrossRef]
  17. Ding, J.; van der A, R.J.; Mijling, B.; Jalkanen, J.-P.; Johansson, L.; Levelt, P.F. Maritime NOx emissions over Chinese seas derived from satellite observations. Geophys. Res. Lett. 2018, 45, 2031–2037. [Google Scholar] [CrossRef] [Green Version]
  18. Franke, K.; Richter, A.; Bovensmann, H.; Eyring, V.; Jöckel, P.; Hoor, P.; Burrows, J.P. Ship emitted NO2 in the Indian Ocean: Comparison of model results with satellite data. Atmos. Meas. Tech. 2009, 9, 7289–7301. [Google Scholar] [CrossRef] [Green Version]
  19. Georgoulias, A.K.; Boersma, K.F.; van Vliet, J.; Zhang, X.; van der A, R.; Zanis, P.; de Laat, J. Detection of NO2 pollution plumes from individual ships with the TROPOMI/S5P satellite sensor. Environ. Res. Lett. 2020, 15, 124037. [Google Scholar] [CrossRef]
  20. Solomiia, K.; Veenman, C.J.; van Vliet, J.; Verbeek, F.J. Estimating individual sea-vessel NO2 emissions using spatial autocorrelation on S5P-TROPOMI satellite data. In EGU General Assembly Conference Abstracts; Copernicus Meetings: Vienna, Austria, 2021. [Google Scholar] [CrossRef]
  21. Roseland, M.; Spiliotopoulou, M. Converging urban agendas: Toward healthy and sustainable communities. Soc. Sci. 2016, 5, 28. [Google Scholar] [CrossRef] [Green Version]
  22. Serra, P.; Fancello, G. Towards the IMO’s GHG goals: A critical overview of the perspectives and challenges of the main options for decarbonizing international shipping. Sustainability 2020, 12, 3220. [Google Scholar] [CrossRef] [Green Version]
  23. Perčić, M.; Vladimir, N.; Fan, A.; Jovanović, I. Holistic energy efficiency and environmental friendliness model for short-sea vessels with alternative power systems considering realistic fuel pathways and workloads. J. Mar. Sci. Eng. 2022, 10, 613. [Google Scholar] [CrossRef]
  24. Perčić, M.; Vladimir, N.; Fan, A. Life-cycle cost assessment of alternative marine fuels to reduce the carbon footprint in short-sea shipping: A case study of Croatia. Appl. Energy 2020, 279, 115848. [Google Scholar] [CrossRef]
  25. Marine Traffic Density. Available online: https://www.marinevesseltraffic.com/NORTH-ATLANTIC-OCEAN/ship-traffic-tracker (accessed on 24 April 2021).
  26. Infraestructura de Datos Espaciales de Canarias. Available online: https://visor.grafcan.es/visorweb/ (accessed on 11 April 2021).
  27. Luis, J.H. The role of inter-island air transport in the Canary Islands. J. Transp. Geogr. 2004, 12, 235–244. [Google Scholar] [CrossRef]
  28. Ortega, A.; Diaz, E. Proposal for a new traffic separation scheme in the Canary Islands. J. Marit. Res. 2015, 12, 19–26. [Google Scholar]
  29. Kraska, J. Particularly sensitive sea. In Freedom of Seas, Passage Rights and the 1982 Law of the Sea Convention; Brill Nijhoff: Leiden, The Netherlands, 2009; pp. 511–572. [Google Scholar]
  30. Van Geffen, J.H.G.M.; Eskes, H.J.; Boersma, K.F.; Maasakkers, J.D.; Veefkind, J.P. TROPOMI ATBD of the Total and Tropospheric NO2 Data Products; Ministry of Infrastructure and Water Management: Amsterdam, The Netherlands, 2019.
  31. Levelt, P.F.; Zweers, D.C.S.; Aben, I.; Bauwens, M.; Borsdorff, T.; De Smedt, I.; Eskes, H.J.; Lerot, C.; Loyola, D.G.; Romahn, F.; et al. Air quality impacts of COVID-19 lockdown measures detected from space using high spatial resolution observations of multiple trace gases from Sentinel-5P/TROPOMI. Atmos. Meas. Tech. 2022, 22, 10319–10351. [Google Scholar] [CrossRef]
  32. Mundi Web Services. Platform to Access Free Copernicus Data, Download and Process Satellite Images, Sentinel-1, Sentinel-2, Sentinel-5P, Sentinel-3. Available online: https://mundiwebservices.com/data/sentinel-5P (accessed on 1 October 2021).
  33. Eskes, H.; van Geffen, J.; Boersma, F.; Eichmann, K.U.; Apituley, A.; Pedergnana, M.; Sneep, M.; Veefkind, J.P.; Loyola, D. Sentinel-5 precursor/TROPOMI Level 2 Product User Manual Nitrogendioxide; Document number: S5P-KNMI-L2-0021-MA.; Royal Netherlands Meteorological Institute: Amsterdam, The Netherlands, 2021. [Google Scholar]
  34. Veefkind, J.P.; Aben, I.; McMullan, K.; Förster, H.; de Vries, J.; Otter, G.; Claas, J.; Eskes, H.J.; de Haan, J.F.; Kleipool, Q.; et al. TROPOMI on the ESA Sentinel-5 Precursor: A GMES mission for global observations of the atmospheric composition for climate, air quality and ozone layer applications. Remote Sens. Environ. 2012, 120, 70–83. [Google Scholar] [CrossRef]
  35. Loyola, D.G.; García, S.G.; Lutz, R.; Argyrouli, A.; Romahn, F.; Spurr, R.J.D.; Pedergnana, M.; Doicu, A.; García, V.M.; Schüssler, O. The operational cloud retrieval algorithms from TROPOMI on board Sentinel-5 Precursor. Atmospheric Meas. Tech. 2018, 11, 409–427. [Google Scholar] [CrossRef] [Green Version]
  36. Hains, J.C.; Boersma, K.F.; Kroon, M.; Dirksen, R.J.; Cohen, R.C.; Perring, A.E.; Bucsela, E.; Volten, H.; Swart, D.P.J.; Richter, A.; et al. Testing and improving OMI DOMINO tropospheric NO2using observations from the DANDELIONS and INTEX-B validation campaigns. J. Geophys. Res. Atmos. 2010, 115. [Google Scholar] [CrossRef]
  37. Hersbach, H.; Bell, B.; Berrisford, P.; Hirahara, S.; Horányi, A.; Muñoz-Sabater, J.; Thépaut, J.N. The ERA5 global reanalysis. Q. J. R. Meteorol. Soc. 2020, 146, 1999–2049. [Google Scholar] [CrossRef]
  38. Povey, A.C.; Grainger, R.G. Known and unknown unknowns: Uncertainty estimation in satellite remote sensing. Atmos. Meas. Tech. 2015, 8, 4699–4718. [Google Scholar] [CrossRef] [Green Version]
  39. Barré, J.; Petetin, H.; Colette, A.; Guevara, M.; Peuch, V.-H.; Rouil, L.; Engelen, R.; Inness, A.; Flemming, J.; Pérez García-Pando, C.; et al. Estimating lockdown induced European NO2 changes. Atmos. Chem. Phys. Discuss. 2020, preprint. [Google Scholar] [CrossRef]
  40. Goldberg, D.L.; Harkey, M.; de Foy, B.; Judd, L.; Johnson, J.; Yarwood, G.; Holloway, T. Evaluating NOx emissions and their effect on O3 production in Texas using TROPOMI NO2 and HCHO. Atmos. Meas. Tech. 2022, 22, 10875–10900. [Google Scholar] [CrossRef]
  41. Ialongo, I.; Virta, H.; Eskes, H.; Hovila, J.; Douros, J. Comparison of TROPOMI/Sentinel-5 Precursor NO2 observations with ground-based measurements in Helsinki. Atmos. Meas. Tech. 2020, 13, 205–218. [Google Scholar] [CrossRef] [Green Version]
  42. Sundström, A.-M.; Majamäki, E.; Jalkanen, J.P.; Ialongo, I.; Tamminen, J. Detecting single ship plumes from TROPOMI NO2. In Proceedings of the 23rd EGU General Assembly, Online, 19–30 April 2021. [Google Scholar]
  43. ERA5. The Family of ERA5 Datasets—Copernicus Knowledge Base—ECMWF Confluence Wiki'sf. Available online: https://confluence.ecmwf.int/display/CKB/The+family+of+ERA5+datasets (accessed on 1 March 2022).
Sustainability 15 04632 g001 550
Figure 1. (a) The Canary Islands are located near the Northwestern African coast; they are part of very important maritime routes in the Atlantic Ocean. (b) The archipelago as a whole, the main routes marked; (c) the central islands that concentrate more than 80% of the population and economic activities of the region, main ports are marked with red push pin.
Figure 1. (a) The Canary Islands are located near the Northwestern African coast; they are part of very important maritime routes in the Atlantic Ocean. (b) The archipelago as a whole, the main routes marked; (c) the central islands that concentrate more than 80% of the population and economic activities of the region, main ports are marked with red push pin.
Sustainability 15 04632 g001
Sustainability 15 04632 g002 550
Figure 2. Standard deviation of [NO2] 1015 molecules cm−2 from 12 to 31 January 2021.
Figure 2. Standard deviation of [NO2] 1015 molecules cm−2 from 12 to 31 January 2021.
Sustainability 15 04632 g002
Sustainability 15 04632 g003 550
Figure 3. The effective area covered by the AIS station is obtained by representing all marine traffic detected during a week (between 12:00 and 16:00 UTC from 2 to 7 April 2021). Each vessel is geo-located every 3 s and is represented by a coloured dot in the plot (different colours indicate different vessels).
Figure 3. The effective area covered by the AIS station is obtained by representing all marine traffic detected during a week (between 12:00 and 16:00 UTC from 2 to 7 April 2021). Each vessel is geo-located every 3 s and is represented by a coloured dot in the plot (different colours indicate different vessels).
Sustainability 15 04632 g003
Sustainability 15 04632 g004 550
Figure 4. Ship traffic data superimposed over the corresponding NO2 distribution; only ship contributions within the 1.5 h period previous to the satellite pass are shown. (a) Using normalised values (normal view) and (b) the right oblique values (slant-view) for the indicated date.
Figure 4. Ship traffic data superimposed over the corresponding NO2 distribution; only ship contributions within the 1.5 h period previous to the satellite pass are shown. (a) Using normalised values (normal view) and (b) the right oblique values (slant-view) for the indicated date.
Sustainability 15 04632 g004
Sustainability 15 04632 g005 550
Figure 5. S5-P TROPOMI weekly coverage over the Canary Islands. Differences in visualisation geometries implies variations in ground resolution cells for each scene.
Figure 5. S5-P TROPOMI weekly coverage over the Canary Islands. Differences in visualisation geometries implies variations in ground resolution cells for each scene.
Sustainability 15 04632 g005
Sustainability 15 04632 g006 550
Figure 6. Time series of NO2 for the period from 1 January to 30 April 2020, at selected locations of the Santa Cruz metropolitan urban area (blue), the port area (orange), and oceanic control area (green).
Figure 6. Time series of NO2 for the period from 1 January to 30 April 2020, at selected locations of the Santa Cruz metropolitan urban area (blue), the port area (orange), and oceanic control area (green).
Sustainability 15 04632 g006
Sustainability 15 04632 g007a 550Sustainability 15 04632 g007b 550
Figure 7. Effects of the COVID−19 lockdown over the monthly averaged distribution of NO2. Monthly averaged distribution just before (a) and just after (b) the COVID−19 lockdown, which occurred on 19 March 2021, corresponding to Julian day 79. Reduction of emissions is evident.
Figure 7. Effects of the COVID−19 lockdown over the monthly averaged distribution of NO2. Monthly averaged distribution just before (a) and just after (b) the COVID−19 lockdown, which occurred on 19 March 2021, corresponding to Julian day 79. Reduction of emissions is evident.
Sustainability 15 04632 g007aSustainability 15 04632 g007b
Sustainability 15 04632 g008 550
Figure 8. Average NO2 distribution at different timescales: 3 months (a), one month (b), one week (c), and 3 days average (d). Increasing the timespan reduces the noisiness of the images due to a smoothing effect of the averaging process.
Figure 8. Average NO2 distribution at different timescales: 3 months (a), one month (b), one week (c), and 3 days average (d). Increasing the timespan reduces the noisiness of the images due to a smoothing effect of the averaging process.
Sustainability 15 04632 g008
Sustainability 15 04632 g009a 550Sustainability 15 04632 g009b 550
Figure 9. Relationship between marine traffic and NO2 distribution at various time scales. Marine traffic within 1.5 h before the satellite overpass is overplotted on the corresponding mean NO2 distributions for a 3-day period (a) and for a three-week period (b).
Figure 9. Relationship between marine traffic and NO2 distribution at various time scales. Marine traffic within 1.5 h before the satellite overpass is overplotted on the corresponding mean NO2 distributions for a 3-day period (a) and for a three-week period (b).
Sustainability 15 04632 g009aSustainability 15 04632 g009b
Table
Table 1. Selected positions for the time series analysis.
Table 1. Selected positions for the time series analysis.
LocationsCoordinates
Urban areas−16.288, 28.4355
−16.288, 28.4674
−16.3518, 28.4355
−16.3518, 28.4674
−16.288, 28.4993
Port Area−16.226, 28.4355
−16.226, 28.4036
Oceanic Background (control point)−16,000, 29,000
−16.07, 28.96
Table
Table 2. Statistical summary of the [NO2] values in the sampled areas in the period from 1 January to 30 April 2020 (in molecules·cm−2).
Table 2. Statistical summary of the [NO2] values in the sampled areas in the period from 1 January to 30 April 2020 (in molecules·cm−2).
MetropolitanControlHarbour
1N300120120
Max5.19 × 10152.31 × 10152.97 × 1015
Mean1.38 × 10157.08 × 10141.22 × 1015
SD4.23 × 10141.95 × 10142.89 × 1014
Table
Table 3. Variation of NO2 content (normalised values) associated with the COVID-19 pandemic lockdown in the metropolitan, harbour, and control areas.
Table 3. Variation of NO2 content (normalised values) associated with the COVID-19 pandemic lockdown in the metropolitan, harbour, and control areas.
BeforeAfterVar (%)
Metropolitan0.44550.2637−40.7906
Control0.18840.19805.0825
Harbour0.35590.3295−7.4121
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Rodriguez Valido, M.; Perez Marrero, J.; Mauro González, A.; Fabiani Bendicho, P.; Efrem Mora, C. Evaluation of the Potential of Sentinel-5P TROPOMI and AIS Marine Traffic Data for the Monitoring of Anthropogenic Activity and Maritime Transport NOx-Emissions in Canary Islands Waters. Sustainability 2023, 15, 4632. https://doi.org/10.3390/su15054632

AMA Style

Rodriguez Valido M, Perez Marrero J, Mauro González A, Fabiani Bendicho P, Efrem Mora C. Evaluation of the Potential of Sentinel-5P TROPOMI and AIS Marine Traffic Data for the Monitoring of Anthropogenic Activity and Maritime Transport NOx-Emissions in Canary Islands Waters. Sustainability. 2023; 15(5):4632. https://doi.org/10.3390/su15054632

Chicago/Turabian Style

Rodriguez Valido, Manuel, Javier Perez Marrero, Argelio Mauro González, Peña Fabiani Bendicho, and Carlos Efrem Mora. 2023. "Evaluation of the Potential of Sentinel-5P TROPOMI and AIS Marine Traffic Data for the Monitoring of Anthropogenic Activity and Maritime Transport NOx-Emissions in Canary Islands Waters" Sustainability 15, no. 5: 4632. https://doi.org/10.3390/su15054632

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop