Internet & Technology

A Pilot Satellite-Based Investigation of the Impact of a Deep Polar Cyclone Propagation on the Phytoplankton Chlorophyll Spatial and Temporal Dynamics in the Arctic Ocean

Description
A Pilot Satellite-Based Investigation of the Impact of a Deep Polar Cyclone Propagation on the Phytoplankton Chlorophyll Spatial and Temporal Dynamics in the Arctic Ocean
Published
of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  Chapter 11 A Pilot Satellite-Based Investigationof the Impact of a Deep Polar CyclonePropagation on the PhytoplanktonChlorophyll Spatial and TemporalDynamics in the Arctic Ocean Dmitry Pozdnyakov, DanLing Tang, Leonid Bobylev, Pavel Golubkin,Elizaveta Zabolotskikh, Dmitry Petrenko and Evgeny MorozovAbstract  A pilot satellite-based investigation of modulations exerted uponmixed-layer phytoplankton fields by deep cyclones is performed for the first timeacross the northern hemisphere polar region, viz. the Arctic Ocean. Resorting to asynergistic approach, polar cyclones were first identified from NCEP/NCAR datafor the summer time period during 2002–2005, and their propagation throughoutthe Barents Sea was further traced down. The above water wind force wasretrieved from QuikSCAT data. These data were further accompanied by oceancolour data from SeaWiFS, and MODIS to examine the spatial and temporaldistributions of surficial phytoplankton chlorophyll concentration dynamics alongthe trajectory of the cyclone’s footprint across the sea. When the wind speed,bathymetric features and cloud conditions proved conjointly favorable, apprecia-ble increases in phytoplankton chlorophyll concentration (for basically oligo-trophic waters of the Arctic Ocean) have been observed following the cyclonepassage with a time lag of about 5 days. This implies that with the ongoingamplification of climate warming at high northern latitudes, the increase inchlorophyll discussed above is potentially capable of boosting the primary pro-duction in the Arctic Ocean. However, further studies are certainly required to D. Pozdnyakov ( & )    L. Bobylev    P. Golubkin    E. Zabolotskikh    D. Petrenko    E. MorozovNansen International Environmental and Remote Sensing Centre,St. Petersburg, Russiae-mail: dmitry.pozdnyakov@niersc.spb.ruD. Pozdnyakov    L. BobylevNansen Environmental and Remote Sensing Centre, Bergen, NorwayD. L. TangState Key Laboratory of Tropical Oceanography, South ChinaInstitute of Oceanology, Guangzhou, ChinaP. Golubkin    E. Zabolotskikh    D. PetrenkoRussian State Hydrometeorological University, St. Petersburg, RussiaD. L. Tang and G. J. Sui (eds.),  Typhoon Impact and Crisis Management  ,Advances in Natural and Technological Hazards Research 40,DOI: 10.1007/978-3-642-40695-9_11,    Springer-Verlag Berlin Heidelberg 2014 241  extend the observational data up to 2012 and further on in order to statistically andphenomenologically underpin and further our understanding of the actual mech-anisms of changes in the Arctic Ocean ecosystem functioning. 11.1 Introduction Among the variety of environmental effects produced by ongoing climate change,significant variations in phytoplankton primary productivity and time- and area-integrated production are becoming increasingly evident (Greene and Pershoing2007). Undoubtedly, the shifts in primary production (PP) amply documented are areflection of serious alterations occurring to the ecosystems of the world’s oceans.Revealed in a wide variety of marine/oceanic regions, the alterations are driven bya host of climate change-related mechanisms that are yet insufficiently understood(Hanshaw et al. 2008).Deep baric formations in the atmosphere are shown to be able to strongly affectPP variations across oceanic/marine tracts (Chang et al. 1996; Lin et al. 2003; Davis and Yan 2004; Tang et al. 2004a, b, 2006; Walker et al. 2005; Zhao and Tang 2006; Rao et al. 2006; Zheng and Tang 2007; Zhao et al. 2008, 2009; Toratani2008; Byju and Kumar 2011; Sarangi 2011; Lin 2012; Chung et al. 2012; Chen and Tang 2012 and more). This is evidenced by quite a number of satellite-based studies employing synergistic approaches. These studies seemingly indicatethat the phytoplankton biomass increase provoked by cyclones might arise from avariety in-water processes (Le Fouest et al. 2011). Nearly invariably they wereconducted over low-latitude waters in the northern hemisphere: the Western NorthPacific (reportedly, the area of the highest incidence of tropical cyclones), andmore specifically, South China Sea, as well as in the Indian Ocean (Bay of Bengaland Eastern Arabian Sea), and northern Atlantic Ocean (Gulf of Mexico and the24  N–38  N latitudinal belt).There are reasons to expect that the effect of deep cyclones on the primaryproduction in the Arctic Ocean can also be appreciable/consequential (Le Fouestet al. 2011) regardless of the fact that it is a low production region of the world’soceans (Arrigo and van Dijken 2011). We are unaware, however, of any satellite-based investigations of this phenomenon at these latitudes.Although the Arctic Ocean ecosystems are subject to atmospheric and hydro-dynamic forcing of different nature (Bobylev et al. 2003), identification of the roleof deep cyclones in PP variations and quantification of the ensuing consequences isan attainable task due to the specific spatial and temporal scales inherent in thisdriving mechanism. Indeed, when studying a single Nordic Sea, the relevant scalesare of a few hundred kilometers and a few days, which are not characteristic of many, but not all, other options of external forcing (Benzeman 2009).Although attainable, the above task must be challenging because of the highincidence of heavy cloud conditions over ice-free tracts of the Arctic 242 D. Pozdnyakov et al.  (Chernokulsky and Mokhov 2012). It necessitates the application of spatio-tem-poral averaging of data in the visible, which, inevitably, will affect the spatio-temporal resolution of spaceborne images, and, thus, complicate the quantitativeassessment of the effect sought for. Obviously, a reasonable trade-off should befound to overcome the above impediment.Since, in comparison with low-latitude productive waters, the PP levels andvariations in the Arctic are rather low, the retrieval error can be significant Pe-trenko et al. 2013), and this imposes strict requirements to the inference of thedesired information. Statistical substantiation of the data is a prerequisite in thiscase.To increase the analytical capacity of the research, a synergistic approach isrequired in order to consider not solely the ocean color data (from which phyto-plankton chlorophyll concentration and PP are retrievable), but also sea surfacetemperature (SST) and near-surface winds, as well as currents, frontal zones/zonesof convergence and divergence, the bathymetry of the target region, and meteo-rological data on the baric fields.Below we present the results of our pilot study aimed at revealing and quan-titatively assessing the impact of deep cyclones on phytoplankton chlorophyll ( chl )spatial and temporal variations in the Arctic. At this stage of research, the targetregion was confined to the Barents Sea (BS), and the time period was limited to2002–2004. In the results analysis presented here not all the abovementionedauxiliary data were yet employed in depth. This will be done at later stages of theresearch implementation. 11.2 Barents Sea General Description Bordered by the shelf edge towards the Norwegian Sea in the west, the Islands of Svalbard in the northwest, and the Islands of Franz Josef Land and Novaya Zemlyain the northeast and east, the BS (Fig. 11.1) is characterized by a relatively shallowshelf and fairly complex hydrography. A seasonably variable ice-cover with itsedge retreating far to the north in summer, an inflow of warm and saline Atlanticwaters (t [ 3   C, salinity [ 35 psu) with the Norway Current and their blendingwith cold and less saline Arctic waters (t \ 0   C, salinity \ 35 psu) makes this seavery sensitive to atmospheric, hydrodynamic and, ultimately climate changeforcing (Reigstad et al. 2002). The average depth is  * 300 m plunging to amaximum of 600 m in the major Bear Island Trench.The interaction between warm Atlantic and cold Arctic waters occurs mainly inthe Polar Front Zone (PFZ). In summer the PFZ is located at 78–80  N extendingamid the BS between Svalbard and the southern coast of Novaya Zemlya.In spring the PFZ structure is frequently affected by mesoscale eddies withcharacteristic sizes of 25–40 km and the water vertical velocity reaching up to20 m day-1. The tidal/ebbing water motions are significant in the Barents Sea with 11 A Pilot Satellite-Based Investigation 243  the tidal amplitude and current direction varying greatly. The atmospheric cyclonicactivity over the Barents Sea is very pronounced throughout the year.As the PFZ enhances the vertical and horizontal mixing in the region, and hencechannels the nutrients up to the euphotic zone, it is a site of high biologicalactivity. It largely explains that, compared to other marine waters of similar lat-itude, the BS is a relatively productive high-latitude marine ecosystem. However,it should not be overlooked that in addition to the PFZ, all the aforementionedatmospheric and hydrodynamic factors can also efficiently modulate and spur upthe BS productivity. 11.3 Sources of Data The occurrence of cyclones moving across the BS has been traced down using thereanalysis data on the geopotential of the 1,000 mbar isobaric surface (http:// www.esrl.noaa.gov/psd) from the National Center for Environmental Prediction/ National Center for Atmospheric Research (NCEP/NCAR) (Kalnay et al. 1996).Satellite data on wind speed and direction over the ocean employed in the studyare from the National Aeronautics and Space Administration (NASA) QUIcK SCATterometer (QuikSCAT) (http://winds.jpl.nasa.gov/missions/quikscat/ index.cfm). The spatial resolution of the QuikSCAT and NCEP/NCAR data are,respectively, 25 km and 2.5   ( * 275 km) with the revisiting frequency of abouttwice and four times per day, respectively.Ocean colour data were downloaded from three satellite sensors: Sea-viewingWide Field-of-view Sensor (SeaWiFS), and Moderate Resolution ImagingSpectroradiometer (MODIS) [on Aqua and Terra] (http://oceancolor.gsfc.nasa.gov) Fig. 11.1  The bathymetryand limits ( blue line ) of theBS.  Black  ,  red   and  greenlines  and  arrows  indicatetracts and direction of propagation of the cyclonespassed over the sea on 15May 2003, 24–25 2003,23–25, 2004, and respectively244 D. Pozdnyakov et al.  at a spatial resolution of 1.0 and 1.1 km, respectively, and the revisiting frequencyof 2–3 and 4–6 times per day for SeaWiFS and MODIS, respectively. SeaWiFS andMODIS (conjointly on Aqua and Terra) ocean colour data were blended daily toobtain a composite image for each specific day. In addition to ocean colour data,MODISsensors also provide SST data atthe space and time resolutionequalingthatfor  chl  fields. 11.4 Methodology 11.4.1 Algorithms The concentration of   chl  was retrieved with the NASA standard algorithms OC4for SeaWiFS (O’Reilly et al. 1998), and OC3 for MODIS Aqua and Terra(O’Reilly et al. 2000). These are modified cubic polynomial functions based on theband-ratio paradigm and employing remote sensing spectral reflectance in thevisible and near infrared channels as an input parameter. The NASA data repro-cessed only recently were employed in the present study: MODIS-Aqua wasreprocessed by reprocessing R2012.0; MODIS-Terra and SeaWiFS were repro-cessed by reprocessing R2010 (http://oceancolor.gsfc.nasa.gov/WIKI/ OCReproc.html).TodetermineSST,‘‘window-split’’retrievalalgorithmswereemployed.Theyarebased on the difference between the satellite-observed water surface  apparent  ( brightness ) temperature,  T  i  determined in two several spectral channels centeredat11  l m( T  11 )and12  l m( T  12 ).TheNASAalgorithmisafour-termexpressionwithproportionality coefficients  c 1 - c 4 : SST  =  c 1  ?  c 2 T  11  ?  c 3 ( T  11  -  T  12 )  ?  c 4 [(sec E  -  1)( T  11  -  T  12 ), where  E  is the satellite zenith angle (http://yyy.rsmas.miami.edu/groups/rrsl/pathfinder/Algorithm/algo_index.html#algo-pathsst andhttp://modis.gsfc.nasa.gov/data/atbd/atbd_mod25.pdf ) (Robinson 1994). The principal of wind speed measurements rests in the backscatter of QuikSCATtransmitted microwave pulses by the water surface. The backscattered microwavesignal is modified by the wind-roughened surface in a certain (logarithmic) pro-portion to the wind speed and direction. Application of inverse modeling allowsretrievingthesoughtforwindparameters.Thesystemmeasureswindsbetween3and30 ms - 1 with an accuracy better than 2 ms - 1 or 10 % in speed and 20   in directionwith a spatial resolution of 25 km (Callahan and Lungu 2006). 11.4.2 Detection of Cyclone Impact Approach Analyzing the  NCEP/NCAR information,  a metadata base was compiled to reflectall cases of cyclones passing through the Barents during the vegetation period, i.e.the time of arrival, passage through and leaving off the BS. 11 A Pilot Satellite-Based Investigation 245
Search
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks