The Egyptian Journal of Hospital Medicine Vol., 4 : 84 – 96 September 2001 I.S.S.N: 12084

Transport Processes In A Heavy Eutrophic Marine Bay, Alexandria (Egypt), Applying Environmental Impact Assessment (EIA) Model

S.K.Mikhail, W.L.Gergis and E.E.Siam*

National Institute of Oceanography and Fisheries *Arab Academy for Sscience and Technology and Maritime Transport

Abstract

Autogenous and allogenous control of the phytoplankton standing crop and biomass was proved in Mex Bay, west of Alexandria (Egypt). Among the clear signs of eutrophication are the heavy visible blooms, which became regular events, at intermittent periods during the warm seasons. The causative organisms progressed differently, the dinoflagellate, Gymnodinium catenatum, is well known toxic species. The Environmental Impact Assessment model (EIA) was applied to Mex Bay. It is a three-dimensional water quality and flow model, based on the three interacted dimensional: longitude, latitude and depth, calculations of time and space variations. The model deals with processes that affecting the water quality. It was initiated with measured field data and load input. The model was used to simulate phytoplankton biomass transport and ambient nutrient concentrations (nitrate, phosphate & ammonia). The model results and the field measurements are compared in order to fit the model. Generally, the calculated flow and concentrations corresponded well with observations. Deviations are attributed to several reasons. A notable influence of the main outfall on the concentrations, particularly in the near shore area, as well as, on the seawater quality was detected. The spatial distribution of the phytoplankton standing crop coincided clearly with the accompanied chlorophyll a concentrations calculated by the model. The results stressed the need to reduce load input into Mex Bay, in order to counteract eutrophication in Alexandria waters. Nitrogen reduction is more important for the whole area, while phosphorous reduction is more effective in the bay.

Introduction

Although coastal marine transiti - ication. This phenomenon is of wide onal ecosystems located between land occurrence in various parts of the world and open sea represent only small area (Vollenweider, 1981). Over 183x106 of the world’s ocean, they are of great m3 of partially treated or untreated ecological and economic importance. sewage and waste waters are discharged The land inflow of waste waters annually into the coast of Alexandria, and subsequently the excessive input of but mostly into it’s the western part. nutrients, mainly nitrogen and phosp - About 40% of the industry surrou - horus is underlying cause of eutroph - nds Alexandri city and as a conseque -

Refree : Prof ; Dr. Abla M . Helmy Afify 84 S.K.Mikhail, et al

nce of the rapid increase of the human Bay caused drastic environmental population and other activities, the task changes. The bay has distinct structural of controlling pollution have become a dynamic properties that easily differe - serious problem. The discharge waters ntiate it from the adjacent Mediter - rendered the neritic waters of ranean open oceanographic system. It is Alexandria highly eutrophic, developing a site of heavy phytoplankton blooms. the occurrence of regular toxic and/or The aim of the present study is to non toxic red tide blooms (Labib, 1992; set up the Environmental Impact Labib, 1994, a, b; Labib, 1996; Labib Assessment (EIA) model to Mex Bay & Halim, 1995). Attempts have been to: - provide an overview of the proce - made to evaluate the quantity of the sses governing the transport of the different sources of pollutants, west of dissolved inorganic nutrients (ammonia, Alexandria (e.g. Said et al., 1991; Abdel nitrate and phosphate); and the seasonal Moneim et al., 1994; Nessim, 1994), and spatial distribution of phytoplan - and their ecological impacts on the kton biomass (Chlorophyll a); - to standing crop (Dorgham et al., 1987; El- determine the extent of eutrophication Sherif, 1989; Samaan et al., 1992). and to discuss the hydrodynamic characters of the bay. A case study: Mex Bay The results gained can give a Mex Bay has a surface area of better insight into practical processes about 19.4 Km2, mean depth of about have to be carried out in order to ameliorate the situation, if possible. 10 m and a volume of about 190.3x106 The final goal of the work is not just to 3 m (Labib,1997). The bay receives study of nutrient cycles and dynamics as directly daily agricultural discharge well as biological processes, but to water, mean of 6.5x106 m3,(Emar ,et.al predict future trends of eutrophication 1992) from neighbouring Lake in changed physical and chemical Maryout, through Umum Drain. This conditions. water is mixed with municipal and The bay is ideally suited for industrial wastes. Five main industries, achieving the purposes of the study Chemical, Chloro-alkali, Tanneries, because its circulation patterns are a Cement and Petroleum, dispose their complicated function of winds, huge waste waters directly into the western input of waste waters of different part of the bay. Due to water sources and pollutants and because of circulation, it is also affected by the ecological and economic problems additional amount of municipal associated with the progressive wastewater from the main sewer of eutrophication. Alexandria located at its eastern part. The study is based on surface data Man-made eutrophication in Mex collected during 1992-1995.

Water Dis charge (m3) 250 200 150 100 50 0 y y t b l 5 2 3 4 a c e u 9 9 9 9 F M O . J

Figure 1: water discharge from Umum Drain (Nessim,1994)

85 Transport Processes In A Heavy Eutrophic… … .

1 Nutrient concentrations and water maximum of 33.8µg. l- . The accomp - discharge input anied minor species were Rhizosolenia

The monthly amount of the fragilissima, Scrippsiella trochoidea and discharged water (m3) into Mex Bay Prorocentrum triestinum. Again, 5-11 October 1994 was a bloom period throughout Umum Drain was shown in 5 -1 Figure. 1. Data obtained from Ministry (maximum 44.2.10 cell.l ), Gymno - of Irrigation, Directorate of Mex dinium catenatum (53.13%), and Pumping Station, Alexandria. Euglena granulata (26.56%), were the responsible species. Chlorophyll a The daily nutrient injection from - land-based sources led to continuous content attained 27.3 µg. l 1. These regeneration of nutrient. The high previous blooms covered the entire ammonia, nitrate and phosphate bay, but with different degrees. The concentrations during May 1992 and western area was always dense. The phytoplankton standing crop was very February 1995 (ammonia reached 60.5 3 -1 µ mol. l-1 during the latter month), poor in February (70.1.10 cell. l ), associated with peaks of discharge despite the remarkable nutrients incr - water input, and very limited ease, Phacus triquter contributed 51% to phytoplankton growth. However, severe the total. Temperature was seen signi - impoverishment occurred during the red ficantly influencing the density. tide bloom periods in July 1993 and Nutrients for their daily replenishment October 1994 (ammonia and nitrate at seen of negligible effect. 4.2 and 3.5 µ mol. l-1 during the latter month, respectively). The Environmental Impact Assess - ment (EIA) model Virtanen et al. (1994) desc Phytoplankton standing crop, species ribes composition and chlorophyll a the Environmental Impact Assessment (EIA) model in detail. The phytoplankton standing crop and species composition progressed The model calculations differently during the different months. Water masses The density was relatively low in May 1992, attaining 0.12.106 cell. l-1. The The water masses are treated as community was mainly represented by horizontal layers interfaced by levels at microflagellate species, 56.54% to the selected depths. In the horizontal plane total standing crop, followed by the model area is subdivided into Scendesemus quadricauda (12.91%), rectangles with arbitrary mesh intervals Protoperidinium steinii (4.9%) and in both directions. Explicit finite- Prorocentrum minimum (3.8%). difference schemes are used for the However, chlorophyll a was relatively numerical solution of flow velocities high (7.3 µg. l-1), influenced by the and of water level elevations. occurrence of fresh-water forms, Mathematically, the physical interact - mainly, Euglena acus and E. gracilis.. A ions and mechanisms affecting water water discoloration was observed on 12 currents are described with Reynold's July 1993, the centric diatom, equations of motion, with the equation Skeletonema costatum, culminating a of hydrostatic pressure as a proper population size at 12.106 cell. l-1, approximation of its vertical compon - 96.2%.Chlorophyll a reached its ents and with the equation of

86 S.K.Mikhail, et al

continuity(1): complex net of interactions. In the model these are gathered within the ∂u transport equation, with internal loading =−∇.p/ ρ +∇.( K.∇)u−2Ω.u−u.∇u+F ∂t o as one of its boundary conditions (Eq. η 3):

P=Pa +g.(η−z).ρo +g—( ρ+ρo)dz where:µ - ≈ ’ ≈ ’ ≈ ’ z ∂c s∂c s∂c s∂c ∂ ∆ ∂c÷ ∂ ∆ ∂c÷ ∂ ∆ ∂c÷ =−u −v −w + Dx + Dy + Dz + ∂u ∂v ∂w ∂t ∂x ∂y ∂z ∂x« ∂x◊ ∂y« ∂y◊ ∂z« ∂z◊ + + =0 ∂x ∂y ∂z +∂(qL) /∂n+R(T,c,...) qL = + + flow velocity vector (µ,v,w), η -water B Eo E1 E2 Where level elevation, P -pressure, ρo -average water density, Ω -angular velocity of c - Concentration (or strength) of any quantity computed, s , s , s -flow the earth rotation, pa -air pressure, K – u v w dispersion matrix of momentum, g - velocity components obtained from Eq. gravity acceleration, F -external forcing 2., , , -dispersion coefficient Dx Dy Dz vector = (0,0,-g). components of concentrations, qL -

The water currents are calculated loading discharges (from point source P, 3D multi layer (Simons, 1980; with the - rivers R, drainage area D, bottom B and Virtanen et al., 1986; Koponen et al., through water surface A), n - discharge 1992). direction co-ordinate, R - changes from The calculation of water currents internal (biogeochemical) reactions is detached from that of transport and within water. water quality (Virtanen and Koponen, In the solution, the simultaneous 1985; Halfon et al., 1990; Koponen et complex interactions are governed by al., 1992). considering each process after the other The flow fields UL induced by with its immediate consequences and winds W and water discharges Q are going through all of the processes step calculated in advance from several by step with short intervals until the levels of water surface η . When the L desired time period as the sum of the water level is changed, the flow fields time-steps for each process is equally for transport calculation changed expired (Virtanen et al., 1994). accordingly. The surface area covered with water is updated. For any water Model data and initial conditions level η between two levels of flow field The in-flowing Mediterranean waters at the boundaries were calculated calculatiηL andηL+1 the flow field U- i.e. the flow velocities u = (u,v,w) at with an average current velocity of 8 cm every location (m,n,k) of the 3D grid - sec-1. The wind data was obtained from will be : the Egyptian Navy Meteorological station for the different months (Fig. 2). η−ηL+1 L ηL −η L+1 U = .U + .U (2) η −η η −η L L+1 L L+1 The model application in Mex Bay The model was calibrated for 4 Calculation of water quality different seasons (May 1992, July 1993, Release, transport, mixing, October 1994 and February 1995), on a resettling and the changes of water point, where measurements were quality and bottom properties are available. Modelled results and field coupled with each other in a most data were compared.

87 Transport Processes In A Heavy Eutrophic… … .

Results and Discussion state circulation patterns at the sea For numerical simulation Mex surface are sketched in Fig. 3. Bay is subdivided with equal mesh The simulation results of the resolution of 300 m into 40 by 17 model run for the different months are horizontal grid squares and in the shown in Figures 4-8. The main vertical direction into 10 one meter indicators of water quality in the present thick layers. The amount of the study were the nutrients (ammonia, discharge water from the Umum Drain nitrate and phosphate). The phytopl - is given as monthly constant input data ankton standing crop was expressed as for the model. The open boundary to the biomass. Mediterranean has a monthly constant The results are shown as time- concentration. The model bathymetry series of their concentrations, as was obtained from the Admiralty Chart, concentration maps, and as numerical by linear interpolation of the depth data values, ranging from smallest to into the model grid. maximum on the computer screen, The simulation results of the expressing the development of nutrients and chlorophyll a are concentrations with colours along a presented as pictures of horizontal selected horizontal lines. concentration distribution on surface. The spatial distribution of the Wind frequency distributions are shown nutrients and chlorophyll a obtained in Fig. 2. from the model results indicated that: During May 1992, the • under the north-westerly wind in May, predominant wind direction was NW the maximum concentrations were and relatively less NE. In July 1993, the restricted to the area around the outlet. winds were less variable, mostly NW. A However, very high nitrate and noticeable change occurred in October ammonia concentrations were also 1994, when the wind direction was seen to spread towards the west. The dominated by NE, relatively less time series at the selected points powerful in NW. Again in winter reflected such pattern. (February 1995), the strong winds were • the predominance of NE wind mostly of NW direction, less affective direction in July affected clearly the towards south and east. distribution patterns, maximum The flow velocities, induced by concentrations pushed to a narrow several combination of wind, jet coastal zone, and a gradual decrease discharge water from Umum Drain and far from the shore and towards west. discharge water at the western part of The time series data showed the bay, as well as the partial effect of variability in ammonia and nitrate the main sewer of Alexandria at its concentrations, while phosphate eastern vicinity, are calculated with a exhibited a very limited range of time-step of 30s for periods of almost 2 variation. weeks. • in October, following the prevailing To show the differences in the wind directions, three distribution resultant flow patterns under the wind patterns were observed. Variations of action, a case study is presented. The phosphate with the time series were water mass is subjected to the free still limited surface shear induced by a uniform and -1 • the jet effect of land run-off on steady wind of 5-m sec , blowing from ammonia distribution was seen in 14 the main eight directions. The steady February under the SN wind direction.

88 S.K.Mikhail, et al

High concentrations covered most of mainly attributed to the effect of the bay. The distribution patterns changing species composition, excha - changed on 17 February, with the nge with the open sea flora and that reverse wind direction, maximum coming from Lake Maryout. A compa - concentrations were located near the rison of the model results and field data coast, extending towards the inner is shown in Fig. 10. In general term, the part of the bay. behaviour of the model results is rather • the characteristics of nutrient distrib - similar to the field observations, and the utions on the surface were in signif - calculated values correspond well with icant connection with wind speed and the real data, but their values were direction. lower, and there are some differences • the spatial distribution of chlorophyll between them. These might be for some a was clearly influenced by the trans - reasons; *the model results represent port processes and the jet effect of the the mean values of the grid squares of discharged water. High concentrations the whole bay. were detected in front of the main *local irregulations in sampling can’t be outlet, as well as towards the western avoided. part of the bay, decreasing towards the *other land-based sources of nutrients open sea. in the bay are not included. As shown in Fig. 9, the correspo - *water exchange at the model open nding spatial distribution of the standing boundaries, significantly influence the crop was somewhat similar to the concentration values in the modelled modelling results of chlorophyll a. area. However, noticeable deviations occur,

89 Transport Processes In A Heavy Eutrophic… … .

*vertical mixing in the model is more a weaker NW gradient, persisting most intense compared to nature. of the time, but modified at times by Subsequently, the concentrations are prevailing meteo-climatic conditions. distributed more rapidly in the vertical This powerful outflow creates a strong direction, causing reduction of concent - offshore surface current, and as a result rations in the surface layer. there is a subsurface compensation *The ecological factors affecting the current of higher salinity towards the concentrations are not taken into coast (Labib, in preparation). account. the species composition which creates *the high supply from Umum Drain sharp deviations between the counts and determines a prevalent SN gradient and biomass.

90 S.K.Mikhail, et al

91 Transport Processes In A Heavy Eutrophic… … .

92 S.K.Mikhail, et al

93 Transport Processes In A Heavy Eutrophic… … .

Conclusions

The hydrodynamic and water and environmental problems, not only quality model (the Environmental in Mex Bay, but also along the coast of Impact Assessment model, EIA), with Alexandria. The increased frequency of available field data, was applied to the visible algal blooms, including toxic highly eutrophic marine basin (Mex species, associated with heavier Man- Bay). The bay, subjecting to a huge made eutrophication could lead to volume of discharge water, has distinct undesirable situations. physical, chemical and biological aspects that easily differentiate it from Acknowledgement the adjacent Mediterranean waters. The bay is a site of heavy algal blooms. The This work was done in the r 4 different months model was run fo framework of LAND-3 Project during the period from 1992-1995, in “Protection of the Coastal Marine order to follow the patterns of the Environment in South Mediterranean”, spatial distribution of the dissolved ICSC World Laboratory, Erice, Italy. inorganic nutrients, ammonia, nitrate We appreciate the help and discussion and phosphate, and chlorophyll a. with our colleagues in this project. Generally, the calculated flow and concentrations corresponded well to observations and the model reacts References: logically to environmental changes. The 1. model describes the seasonal variations Abdel-Moneim, M.A; Khaled, A.M. and Iskander, M .F., 1994. A study of of nutrients and chlorophyll a rather the levels of some heavy metals in Mex well. The largest deviations in the algal Bay, west of Alexandria, Egypt. The 4th biomass results are observed in summer Conf. Environ. Prot. is a must, 10-12 and autumn, affected mainly by the May, 155-174. variability in species composition. A 2. Dorgham, M. M.; El-Samra, M.; notable influence of the main outfall on Moustafa, M . I., 1987. Phytoplankton the concentrations, reflecting a jet in an area of multi-polluting factors effect, was distinct. The concentrations west of Alexandria, Eg. Qatar Univ. decreased away from shore and towards Sci., Bull., 7: 393-419. the open sea. The western part of the 3. ElSherif, Z. M ., 1989. Distribution and ecology of phytoplankton in El-Mex bay, affected by transport processes, Bay (Egypt). Bull. Inst. Oceanogr. was denser and richer. Deviations are Fish., A.R.E. 15 (2): 83-100. attributed to several reasons. Finer 4. Emara, H.L., Shiradah, M.A., vertical resolution or stratification Moustafa, T. H. and El-Deek, M. S., would be taken into account to reduce 1992. Effect of sewage and industrial the vertical mixing. In future, the wastes on the chemical characteristics model must be tested and verified with of the Eastern Harbour and Mex Bay extensive nutrient and phytoplankton waters of Alexandria, Egypt. In: Marine data. coastal Eutrophication, Proc. Int. Conf. It is recommended that the load Bologna, Italy, 21-24 March, 1992 reduction of nutrients, particularly .Elsevier, 773-784. 5 nitrogen and phosphorous, and the . Halfon, E.; Simons, T. J.; Scherzer, W . M ., 1990. Modelling the spatial water discharge input are very important distribution of seven halocarbons on to overcome serious health, economic Lake St. Clair in June 1984 using the

94 S.K.Mikhail, et al

TOXFATE model. J. Great Lakes Res., in Alexandria, Egypt. Mar. life, 91: 11- 16: 90-112. 17. 6. Koponen, J.; Alasaarela, E; 13. Nessim, R.B., 1994. Environmental Lehtinen, K.; Sarkkula, J.; characteristics of Mex Bay. 1st Proc. Simbierowics, P.; Vepsa, H.; Arab cof. On Marine Environ. Protec. Virtanen, M ., 1992. Modelling the Alexandria, 5-7 February: 221-243. Dynamics of a Large Sea Area, 14. Said, M. A.; El-Deek, M. S.; Bothnian Bay Research Project 1988- Mahmoud, Th. H. and Shridah, M. 1990. Publ. of the W ater and M . A., 1991. Physicochemical Environment Research Institute, No 11, characteristics of different water types National Board of waters and the of El-Mex Bay, Alexandria, Egypt. Environment, Helsinki, Finland, 73p. Bull. Nat. Inst. Oceanogr. Fish., A.R.E., 7. Labib, W .,1992. Amphicrysis 17 (1): 103-116. compressa Korshikov red tide bloom 15. Samaan, A. A.; Abdella, R. R. and off Alexandria. Biology of the Gergis, W . L., 1992. Phytoplankton bloomand associated water chemistry. population in relation to hydrographic CERBOM, Centre d’etudes et de rech - conditions along the West Coast of erches de biologie et d’ Oceanographie. Alexandria (Egypt). Bull. Nat. Inst. Tomes, 107-108: 13-23. Oceanogr. Fish., A.R.E., 18: 53-71. 8. Labib, W .,1994 a. Ecological study of 16. Simons, T. J., 1980. Circulation spring-early summer phytoplankton Models of Lakes and Inland Seas. Can. blooms in a semi-enclosed estuary. Bull. Fish. Aquat. Sci., No 203. 145p. Chemistry and Ecology, 9: 75-85. 17. Virtanen, M . and Koponen, J. 1985. 9. Labib, W .,1994 b. Massive algal Simulation of transport under irregular pollution in highly eutrophic marine flow conditions. Aqua Fenn., 15 : 65- basin, Alexandria, Egypt. The 4th Conf. 75. of the Environ. Prot. is a must, 10-12 18. Virtanen, M.; Koponen, J.; Dahlbo, May 1994, 181-194. K.; Sarkkula, J., 1986. Three- 10. Labib, W .,1996. Water discoloration in dimensional water quality transport Alexandria, Egypt, April 1993. I- model compared with field Occurrence of Prorocentrum triestinum observations. Ecological Modelling, 31: Schiller (Red Tide) bloom and 185-199. associated physical and chemical 19. Vitranen, M; Koponen, J.; Hellsten, conditions. Chemistry and Ecology, 12: S.; Nenonen, O.; Kinnunen, K., 1994. 163-170. Principles for calculation of transport 11. Labib, W .,1997. Eutrophication In and water quality in strongly regulated Mex Bay (Alexandria, Egypt) reservoirs. J. Ecological Modelling, 74 : Environmental Studies And Statistical 103-123. Approach.Bull. Nat.Inst. of 20. Vollenweider, R.A.,1981. Eutrophi - Oceanogr.&Fish., A.R.E.,23: 49-68. cation- algal problem.WHO water 12. Labib, W . and Halim, Y., 1995. Diel Quel. Bull., 6. vertical migration and toxicity of Alexandrium minutum Halim red tide,

95 Transport Processes In A Heavy Eutrophic… … .

>?@ABCDEFG HCIJK LMNO PFMQ >GRSO TUF?V >MNIW XMMYZ [\M]JK XMMY^NJ _]?`aZ bcRId eKAf^DFG

qFI?r/U** -Hm`m nM]J oAmp / U.K * -iMjFfMQ iQFk >MQFD /U* eFMs XMtK`Gr FGHIJK LMNOJ PQMRJK STOQ* UV]HIJK UiaIJK UVGWX YZ[ \]HIJK ^R_JK` GabMJM_cdJK` LMNONJ Uae]OJK UaWVfGghK **

j]k[ lkcWJK makNn oGkaQ UeMkpn qfGkVr sMktMQ UkauHIJK Uk[FMJK ovkw x[Gk_y FMkTz` {akpJK` |kae]JK PNkp} ~k} UkeFGRdQ •K]kd} PkN€ •kJ‚ FMkTz` UVFS_ckƒ„K .…MNdJK †Q K]aug ‡IZy ~dJK UQGZJK •GaˆMZJK KSkWdOQ UkaiaIJK •K]aŽ•kdJK YaaRdJ ~eGZX ‰‚MWŠ ‹Œn †Q sMtMWJK Kvw U•[G_Q Yy ]Ž•ky ~kdJK •GkaNWOJK PN€ •Jvg` ‘WOJK` ’ ]OJK` ‹M“JK UŽŒuJK UaƒGƒ”K fGOe”K PN€ .]HIJK oGaQ UJGX PN€ ™kašHdJ •Jvkg LS•dkZV` ˜ UkabFG•JK ^kQKMOJK ]aŽ•ky YŽ UaiaIJK ^QKMOJK –GaRe —SIV` †kQ UkVvœWJK •ŒkQ”K UIkZŠ ™kag]y • Jvkg` GkTdg]X` UkayGI_JK •GkW›GTNJ UkVMaHJK UkNdcJK †kQ GkTaN€ G_NkpX ~kdJK m›Gkd_JK` UkaNRHJK m›Gkd_JK UkŠFGRQ YkŽ •GškƒM}` •K]kdŠ` GkaŠMQ— .^uQ”K ‰‚MW_JK PJž ^p_J ~IV]ŸdJK ‰‚MW_JK •GkƒKFSJK |kQ GkQGWy UkW›ŒQ kŠGg UkaNRHJK •GkƒGaRJK ¡— kIudJ m›Gkd_JK •¢Gkb` UkŸadŠ UkanG_WJK •K]kaœdNJ ^kO} fF ŠGc} m›Gd_JK ¤ Oe ~} •G}K]HŠhK GQ— •G£XŒWJK` . ‹MpšJK ¥Œdnh ^kka}`FMNcJK ™kag]dJ GkQrŒQ ¡Gkg UkayGI_JK •GkW›GTJK |kkVrMy ¡— Gk¦V— m›Gkd_JK •]kTz—` .~ƒGaRJK ‰‚MW_JK U“ƒKMe jMZHWJK (—)§MaHJK ~k} ‡kpy ~kdJK UkabFG•JK •GkŽMNWJK PkN€ q]“akZJK qF`]kt m›Gd_JK Ht`—` ™kkag]y qfGkkVr ^kkaNRy ‘kkV]ˆ †kk€ •kkJ‚` UeMkkp•JK qfGkkVr UkkJGX ^kkaNRdJ lkkcWJK makkNn .maN•JK oGaQ ~} FMšZšJK` †ab`]da_JK

96