This study was conducted in the sugar belt regions of the Nzoia River catchment area of Kenya. River Nzoia Catchment lies between latitudes 1˚30'N and 0˚05'S and longitudes 34˚ and 35˚45'E as shown in Figure 1. One of its major tributaries of the River Nzoia is River Isiukhu. The sampling locations within Nzoia River Catchment were selected depending on the social-economic activities and waste generation and disposal methods within the study location. The specific sampling locations were; Shirere, Rostaman, Mwibatsiro, Rostamandumpsite, Mmust dumpsite, Shirere, Shibeye, Ekero, West Kenya sugar company, Nzoia sugar company, Mumias sugar company and within Kakamega Municipality. Soil samples were collected from Mumias farm, Mumias nuclear estate. Sludge soil was collected from the Municipal dumpsite and Mmust dumpsite. Sediment and water samples were collected from Mumias sugar company stream, at Ekero, Shibeye and Rosterman all along Isiukhu river, Mwibatsiro and MMUST stream. Within the sugar company, the by products as waste from the sugar production proceses sampled for analysis were; Baggasse waste from Nzoia Sugar Company (NSC), Baggasse waste from West Kenya Sugar Company (WKSC), Filter cake WKSC, Filter cake NSC, Boiler water NSC and Boiler water WKSC. Table 1 describes the area where samples were collected.

The selection of these points was based upon human activities taking place near the river banks such as sand harvesting, farming, waste and effluents disposal methods and accessibility. The area surrounding the sampling sites had sugarcane farming as the main economic activity. The sampling sites, rivers, and the forest within the study area are shown in Figure 1.

High purity standard for PAHs was purchased from Sigma Aldrich, UK. Solid Phase Extraction cartridges and nylon micro filters were purchased from Merck

Chemicals. Anhydrous Sodium sulfate (Na₂SO₄), Dichloromethane and Iso-octane were of analytical grade reagent purchased from Kobian Chemicals Nairobi. Internal standard dodecane was purchased from Sigma Aldrich, UK and Reference standards of PAHS obtained from IoIc (Warsaw: Puder, Poland). Soxhlet extraction apparatus, SPE cartridges, 0.45 μm PTFE (polytetrafluoroethylene) syringe filter, cellulose extraction thimble, from Kobian Chemicals Nairobi and GC-MS instruments with the following specifications: GC Model of 7890A manufactured by Agilent technologies, Inc, Beijing China, with a MASS detector of 5973C by Agilent Technologies, Inc. Santa Clara, CA, USA attached to an auto sampler of 7683 by Agilent Technologies.

Reconnaissance study was done to obtain information regarding the locations where sampling was done, based on the activities that take place within the locations (see Table 1). Sample was collected from sites believed to have previous history of anthropogenic activities that include open burning or combustion such as dump sites, sugarcane fields, urban areas, rivers in the vicinity of sugar factories, and also in sites that have no history of interferences by anthropogenic activities such as Kakamega forest, for comparison. Control samples were obtained within the Kakamega forest. Sediment sample was collected using an Ekman Grabber. Water sample was collected in amber bottles and soil samples were collected using a stainless steel hoe and scooped using stainless steel shovel. The samples were then wrapped in clean piece of aluminum foil, packed in black polythene bag, and labeled. The entire sample was placed in cooler boxes and transported to the university laboratory where they were stored in the refrigerator at 4^°C.

Extraction of water samples was done by solid phase extraction procedure which was adopted from US EPA Method 3510C [26]. 500 mls of water sample was measured using a glass measuring cylinder, transferred into a 1.0 L beaker and its pH recorded. The water sample was filtered using what man filter paper No.1 to remove other impurities. The pH was adjusted by adding drops of 0.1 M hydrochloric acid or 0.1 M sodium hydroxide solutions with carefully stirring to adjust the pH to 7.0. The neutral solution was transferred to 500 ml beaker and treated with 25 g of sodium chloride to salt out the PAHs from aqueous phase to organic phase. The mixture was further shaken gently and allowed to settle for 30 minutes to enhance homogeneity. Extraction process was repeated thrice each time with 30 ml dichloromethane. The combined extracts were dried using activated anhydrous Na₂SO₄. 2 ml of isooctane was added as keeper of the PAHs then concentrated to about 2 ml. The concentrated extracts were filtered with 0.45 μm PTFE (polytetrafluoroethylene) syringe filter, and the volume adjusted to the 2 mls mark using the mobile phase solvent before the extract was stored in a fridge at −4˚C prior to GC-MS analysis [27]. Dodecane was used as the internal standard.

Sediment samples were removed from freezer and allowed to thaw for about 4 hours prior to extraction. Samples were air-dried in the dark, crushed and sieved with a 2 mm sieve. About 10 grams of the sieved samples were weight and put into the Soxhlet extraction thimble and 0.1 of 0.01 µl/ml dodecane solution added as the internal standard. This was extracted with 250 ml of dichloromethane in a round bottomed flask for at least 8 hours in a Soxhlet extractor set-up. The extracts were dried using 2 g of anhydrous sodium sulphate. 2 ml of isooctane was added as keeper, and the mixture concentrated to 2 ml. The concentrated extracts were filtered through a 0.45 μm PTFE (polytetrafluoroethylene) syringe filter, stored in a fridge at −4˚C prior to GC-MS analysis.

The samples were run using GC-MS: GC Model of 7890A hyphenated with a MASS detector of 5973C by Agilent Technologies, Inc. Santa Clara, CA, and USA attached to an auto sampler of 7683 by Agilent Technologies. The GC was equipped with a DB-5 silica fused high performance capillary column, (30 m length, 0.25 mm internal diameter and 0.25 μm film thickness). The carrier gas was helium (99.99%) with a constant flow rate of 2 ml/min. Make up gas was (99.99%) with a constant flow rate of 2 ml/min. The injection volume was 1 μl with a pulsed split less injection mode. The oven temperature was started initially at 80˚C where it was held for 3 mins. Programmed for two temperature ramps was from 80˚C to 280˚C with increase at 10˚C∙min⁻¹ and held for 20 mins. Injector temperatures were 250 and detector temparature was 300˚C. Solvent delay was programmed at 5 mins. The mass selective detector was operated in the electron ionization (70 eV) and the selected ion monitoring modes. Mass range was between m/z 30 - 450. Samples for GC-MS analysis were evaporated to dryness under reduced pressure and residue re-dissolved in 50 μL of iso-octane. Analysis by GC-MS was performed in split/splitles mode. Mass spectrometry was operated on electron impact ionization mode. Analyte identification was also based on the comparison of mass spectra of unknowns with those in the NIST/EPA/NIH MASS SPECTRAL LIBRARY library [28]. Quantitative analysis of PAHs was achieved through the integration of selected GC chromatograms.

Extraction efficiency was tested by obtaining recovery values. Five blank samples which were prepared in the same way that the real analytical samples were prepared were used for each set of ten real analytical samples analysed. Calibration curves were drawn for all the elements to be analyzed and regression equations determined for every calibration curve, to check the instrumental response to various analyte concentrations. In all the cases regression was found have r² = 0.99 and above. Accuracy and precision of the method were also tested.

The distribution of PAHs varied among the sampling locations and the sample matrices. There was no single sample where all the priority PAHs were detected. Variation in concentrations indicated different sources of PAHs [32]. The observed concentrations distribution in the Nzoia catchment region, indicates varied sources of the studied PAHs. Figure 2 shows the distribution in percentage

(%) of PAHs in water and sediments samples. Mumias sugar company stream sample had only two PAHs present with Nap distribution in the sample at over 75%. Ekero, Shibeye, Mwibatsiro and Rosterman which are located along Isiukhu river had at least three PAHs in the water samples.

However, Acy, Nap and Ace make up to more than 50% of total PAHs in water of the three locations respectively. Benz(a)anthracene (BaA) is dominat in water sample from MMUST stream (over 90%) and Mwibatsiro (Over 50%) (see Figure 2). In a study by [33] that BaA was among the PAHs that showed a good correlation indicating that they originate from the burning processes. Classification of PAHs is based on molecular weights (MW). The classes of PAHs strongly correlate with their environmental fate. Lower molecular weight is PAHs of 152 to 178 g/mol. The medium (middle) molecular weight PAHs, have about 202 g/mol and High Molecular weights have MW of 228 to 278 g/mol. Predominance of low and medium molecular weight PAHs in water and sediment samples reflected a similar source. The high MW class of PAHs has low solubility in water. These could explain the observed low or/and below method detection limit concentrations of BaA, Char, BbF, BaP, InP and DBC in water samples. The possible contributors of PAHs determined in sediment samples might be through surface runoffs from wastes dumping sites along the river banks. Discharge sewage effluents from the sewerage plant may be contributing significantly to Nap and Flu present in the Mwibatsiro sediments. High hydrophobicity of PAHs is associated with their adsorption on sediments, organic matter and suspended particles [34]. Naphthalene and Phenanthrene are degraded rapidly in sediments but higher MW PAHs such as Benzo(a)pyrene are more recalcitrant [35].

Soils are likely to retain high MW PAHs. The EPA has classified Benz(a)anthracene (BaA), Benz(a)pyrene (BaP), Benz(b)flourancene (BbF), chrysene (Chr), Dibenz(ah)anthracene, and Indeno(1,2,3-cd)pyrene (InP), as probable human carcinogens [7]. Of these carcinogenic PAHs, Benz(a)pyrene (BaP), Benz(b)flourancene (BbF) and Indeno(123,cd)pyrene (InP), were major contributors to the river water and sludge soil. This observation is of major concern since water from the rivers within the study region is abstracted for domestic use. Also, the sludge soil is used as fertilizer in agricultural farms (Figure 3). Show the percentage of distributions of individual PAHs in soil samples obtained from different localities within the Nzoia River catchment area. Naphthalene (Nap), was distributed in all soil samples except the sludge soil sample. It was higher in the MMUST sample (60% distributed). The MMUST soil sample also contained Indeno(123,cd)pyrene (InP), which tends to adsorption strong onto the soil organic matter. Anthracene recorded the highest concentration of 20 µg/kg and fluorine with the least concentration level of 0.8 µg/kg in Mumias nuclear farm. Observed concentrations of PAHs at this sampling location can be attributed to agricultural activities such as cane farming within the study area is associated with cane burning, during land preparation for another crop. At the

Kakamega municipal dumpsite, the high concentration of anthracene of 1200 µg/kg and Benzo(b)flourancene may therefore be as a result of open burning, which was the observed method used to reduce the waste volumes at dumpsites within the study region. The distribution of anthracene in soil samples was over 85%, 65% and 25% in municipal dumpsite, mumias nuclear estate and municipal sludge soil samples respectively (see Figure 3). Anthracene is to be one of the most prevalent PAHs in soils in the study region. PAHs are generated by incomplete combustion, thermal means such as fires and burning or by other anthropogenic sources [36] [37]. Higher concentrations of PAHs in urban soils than in rural soils is a clear indicator of urban centers associated anthropogenic activities contribution of PAHs in the environment.

Anthracene was quantified at the municipal dumpsite but not in water and sediment within the dumpsite area. This observation indicated a diffuse source of anthracene. Naphthalene with a concentration of 0.06 µg/L and 160 µg/kg within the vicinity of munipal dumpsite at Rosterman could indicate that the source of naphthalene in water was the municipal dumpsite because of the ratio of 160:0.06 for the dumpsite and water respectively. However concentration of naphthalene at MMUST dumpsite was 6.0 µg/kg and was lower than the concentration obtained in water within the same vicinity which was 38.4 µg/L (ratio of 0.06:38.4) an indication of other various sources of naphthalene in water.

Decomposition of waste at dumpsites, other pyrogenic processes and open burning that take place at the municipal dumpsite may produce and/or release PAHs to the soil. Phenanthrene was distributed in about 50% of the mumias nuclear estate farm and about 20% of the soils from municipal dumpsite respectively, even though phenanthrene is espected to be removed from soils due to its relatively high water solubility.

Polycyclic aromatic hydrocarbons are ubiquitous in the modern environment [38]. Results from around the globe from various researchers generally show a variation in concentrations of PAHs among countries, and sampling locations. [21] researched on PAHs in water and sediments of car wash and kisat areas of winam Golf of Lake Victoria in Kenya. The PAH concentrations in sediment and water ranged from 0.04 to 31.95 µg/kg dry weight and 3.32 to 55.8 µg/L, respectively, depending upon the sampling location. [39] studied the distributions of PAHs in surface waters, sediments and soils of Hangzhou city, China and reported that the total concentrations of 10 PAHs in water ranging from 0.989 µg/Lto 9.663 µg/L, with a mean concentration of 3.717 µg/L. The composition pattern of PAHs in the water showed that PAHs (naphthalene, fluorene, phenanthrene, and pyrene) were the most abundant in all water samples. Five-ring PAHs Benzo(a)anthracene, Benzo(b)fluoranthene, chrysene, and Benzo(a)pyrene concentrations gave the lowest concentrations. Additionally, another study reported in Niger Delta, Nigeria where PAHs concentrations were averaged at 0.1761 mg/L was determined in selected water bodies in Niger Delta and 1.95 μg/L to 10.9 μg/L determined in Ekpan Creek [40]. In Ghana, total PAHs reported their concentration range up to 14.587 μg/L [41]. Another study by [42] from China reported that areas with intense navigation and large industrial facilities including oil refineries had higher concentration of PAHs in sediments. In the study, the range of PAHs in concentration obtained ranged from 464 to 2.62 μg/kg as shown in Table 5.

Study done in China reported PAHs levels from 0.016 μg/kg in rural to 3.884 μg/kg in the suburban soils of Beijing, China [43]. These levels are higher than those obtained from this study. According to [44] observed that industrial sites had higher PAHs levels than other more remote regions studied. Range in concentrations of PAHs determined by Oketola and [45], in the leachate samples in Lagos Nigeria was 0.85 to 1.47 mg/L and for top soil around the dumpsite, the values 0.94 - 2.79 mg/kg. Poor management of dumpsites can cause the leaching of leachate to ground water and surface water through run-off, and can have adverse effects on human health and the ecosystem. [46] obtained high concentrations of total PAHs, inside the landfill (1475 μg∙kg⁻¹ dw), whereas concentrations in the surrounding soils were by far lower ranging between 11.2 - 28.1 μg∙kg⁻¹ dw for PAHs.