This study was conducted in a small agricultural catchment (132 ha) which is located in Yoeju, Gyeonggi province of Korea (37˚16'32.22"N, 127˚35'21.32"E) (Figure 1). The catchment consists of 17% paddy rice field and 20.5% upland. The experimental plots in paddy rice field and cultivated upland were located in the catchment. Soils in a paddy rice field and upland were all sandy loam, and soil pH were 5.8 and 6.4, respectively (Table 1). In addition, the organic matter contents were 19.6 and 26.4 g∙kg⁻¹ for a paddy rice field and upland, respectively. The available phosphate content from upland was 967 mg∙kg⁻¹ which was much higher than onesfrom paddy rice field soil.

Two different pig manure-based liquid fertilizers were applied in this study, which were bio-filtration process

(SCB) and storage process (SP) (Table 2). Nitrogen contents in SCB and SP liquid fertilizers were 0.26 and 0.40%, respectively. The content of potassium in SCB liquid fertilizers was especially greater than the other. It was caused by the content of the soluble potassium in the liquid type fertilizers. The content of the soluble potassium in urine was greater than ones in feces. Therefore, it was easily drained to the outside of filtering system without being filtered out during the process of SCB.

They were applied to the total of 16 plots (4 for paddy rice field and 12 for upland) in the studied catchment, and the size of each plot was 20 × 36 m for a paddy rice field and 2.7 × 10 m for upland. The amount of N requirement was pre-calculated for the crop cultivation. For comparison, the experimental plots also included no fertilization plots (control A) and chemical fertilization plots (control B). Paddy rice (Oryza sativa L.) was transplanted into each experimental plot at the space 15 × 30 cm and grown for five months. Corn seeds (Zea mays L.) were sowed in each experimental plot at the space of 20 × 60 cm and grown for three months.

In order to assess the effect of livestock manure appli-

Table 1. Chemical properties from paddy rice field and upland before the experiment.

Table 2. Chemical properties and heavy metals from liquid fertilizers that were applied to paddy rice field and upland.

^*SCB: Liquid fertilizer with slurry composting and bio-filtration process; SP: Liquid fertilizer with storage process.

cation on land at catchment-scale, water quantity and quality were monitored at the catchment outlet. A set of equipments was installed to measure water level and velocity (Starflow, Ultrasonic Dopper Instrument, Unidata Co., Australia) at the monitoring site. Water quality was monitored in stream nearby the investigated paddy rice field and cultivated upland during the cultivation period from May to October in 2009.

Water samples were collected from paddy rice field and upland as well as stream water to monitor the changes of water quality properties during the cultivation period. All samples were refrigerated at 0 to 4˚C soon after collected and stored in the refrigerator until the analyses. Water samples were transported to the laboratory in RDA (Rural Development Administration, Suwon, Korea) and were shaken to obtain homogeneous aliquots for the analyses.

Electrical conductivity (EC) was measured in the field with portable meters (Model 145A, Orion Co., USA). The samples from the drainage water/runoff and stream water were analyzed for suspended solids (SS) and chemical oxygen demands (COD), nitrogen and phosphorous concentrations. SS and COD were analyzed by the Standard Methods [15] and by the dichromate reflux method [16]. The NO₃-N and NH₄-N were analyzed using ion chromatography (DX-100, Ion Chromatograph, Dionex Co., USA) and using the indophenols-blue method [17] through 0.45 mm nylon membrane filter, respectively. Total N (T-N) was analyzed by the standard methods according to the Federal Water Pollutant Clean Act (FWPCA) manual [18] and total P (T-P) was measured by the isobutanol extraction method described by Golterman and Glymo [19].

The combined use of livestock manure and mineral fertilizer results in considerable nutrient enrichment in soil surface [2] and overland flow from such areas during or after a rainfall event. The flow contains relatively high levels of nutrients (e.g. N, P, K), organic matter and suspended particles [20,21].

Table 3 shows the observed concentrations of N and P in surface drainage waters from paddy rice field during the cultivation period. The SS concentration in water samples were ranged from 0.01 to 0.07 mg∙L⁻¹ for all treatments. The concentrations of T-N and T-P in drainage waters were 1.85 - 3.45 and 0.18 - 0.44 mg∙L⁻¹, respectively. The concentrations of N and P and the EC values from SP and SCB treated plots were much higher than the ones from control A (no fertilization) and control B (chemical fertilization) (Table 3). Therefore, there is a need to manage drainage water from SP and SCB treated areas.

Manure is generally composed of NH₄-N and organicN. Thses components are often mixed with equal ratios. NH₄-N is more flavorable to plants than organic-N, because NH₄-N is readily available to plant. In the meantime, organic-N needs to be mineralized to become available for plants. NH₄-N originated from manure was found to be biologically oxidised into NO₃-N before it was converted into organic-N. This causes N losses in Manure [22]. Therefore, the effect of nitrate loss by manure application needs to be largely determined by plant uptake and the factors that control the processess of N mineralisation and nitrification. Phosphorus transport in surface runoff occurs both in soluble and particulate forms. Phosphorus is generally transported in particulate form [8,23]. The transport is influenced by several factors, including land use, management practices, soil type, time and rate of fertilizer and manure application.

Table 4 shows the observed concentrations of N and P in runoff waters from upland during the cultivation period. The SS concentrations in water samples were ranged from 1.0 to 1.7 mg∙L⁻¹ for all treatments. The T-N and T-P concentrations in runoff water samples were 6.07 - 8.45 and 1.39 - 3.02 mg∙L⁻¹ during the rainfall events. Losses of N and P by the runoff waters from SP and SCB treated plots were higher than the ones from the control plots (A and B). Particularly, the loss of nitrate-nitrogen was greater than the loss of ammonium-nitrogen, and this can be explained by the nitrification process in manure as

Table 3. Water quality properties in the drainage waters from paddy rice fields.

^*CF: Chemical fertilization; SP: Liquid fertilizer with storage process; SCB: Liquid fertilizer with slurry composting and bio-filtration process; ^**Data are expressed as mean ± SD (n = 8).

Table 4. Water quality properties in run-off water with application of liquid pig manure at the study upland.

^*CF: Chemical fertilization; SP: Liquid fertilizer with storage process; SCB: Liquid fertilizer with slurry composting and bio-filtration process; ^**Data are expressed as mean ± SD (n = 7).

described above.

Similar to N loss in Manure, several factors affect phosphorous loss by surface runoff; land use and management practices, soil type, time and rate of fertilizer and manure application. Consequently, nutrient leach from agricultural land flows into surface water and this causes acceleration of eutrophication in the surface water. In the end, the influx of excess nutrients contributes to the degradation of surface waters [20,23]. This nutrient leach from agricultural land can be an important factor to not only soil or water, but to the whole environments.

Figure 2 shows the changes of nutrient concentrations in stream water near the experimental sites of a paddy rice field and upland during the cultivation period. A lot of nutrient inputs for crop cultivation were applied to agricultural lands, particularly from May to June (the cultivation period). These nutrient inputs caused higher concentration of nutrients to the drainage waters during the cultivation period. Therefore, the concentrations of COD, T-N and T-P during the early cultivation period, May and June, were higher than the ones in July through August when it was the rainy season.

Rainfall events characteristically govern the nutrient of streams draining from livestock areas. Generally, when there is no rainfall, the level of nutrients including BOD remains within normal range in surface waters unless farm effluents are discharged directly into streams. This effluent may have even greater nutrient concentration if there has been no rainfall event [24]. In addition, the impact intensity of land use on water quality was as great as the amount of surplus nutrient in cropland from the respective drainage watershed [3]. Therefore, a proper management of cropland is very important in protecting the quality of stream/river water, e.g. managing paddy rice field and upland proportions in drainage watersheds and of surplus nutrient.

Table 5 shows the monthly distribution of discharge water and nutrient loads into stream water during the cultivation period. The total amount of precipitation in

Table 5. Monthly discharge water and nutrient loads in streams located nearby paddy fields and uplands during the crop cultivation period.

2009 was 1163 mm and the monthly precipitation was constantly high in July and August. The highest monthly precipitation was occurred in July, 2009 and the amount of precipitation was almost 55.5% of the annual precipitation in Korea. Discharge water, chemical and nutrient loads were recorded the highest in July, the highest precipitation month. It shows that precipitation is the most significant factor on water, chemical and nutrient loads which influence the water quality in streams.

Generally, small scale sites have a clear positive relationship between the loads of leached nutrient and rainfall [25]. Previous study reported that the changes in the nutrient concentrations of the drainage water over rainfall events indicated the transport of the particulate materials and of the nutrients such as nitrogen and phosphorus were mainly influenced by direct runoff and base flow, respectively [6].

Therefore, it is important that the loss of nutrients in surface runoff should be minimized for the sustainable agriculture and water quality. This can be done by adapting the best management practices considering the climate and crop cultivation condition. Because of this, riparian croplands have been major targets to manage nutrient losses in surface runoff. Since the dissolved nitrogen ratio was found to be high in base flow, managing spring water can be a solution to effectively reduce dissolved nitrogen [6]. Therefore, in case of upland cultivation, buffer strips and wetlands are effective managements to reduce both particulate and dissolved materials in runoff. In case of rice cultivation in paddy fields, it is important to manage irrigation and drainage waters at the early cultivation period in paddy rice fields to increase nutrient use efficiency and prevent nutrient losses.