In order to collect samples and information from dairy farms, basic information was obtained in person regarding breeding heads, bedding materials, excretion structure and the amount of manure production after the 41 farms in Korea was investigated and the contents of N and P in those samples were measured. Methods of manure management were also investigated with detailed information recorded. Samples were collected from fresh manure and compost prior to application. For sampling on site, 20 L bucket was used for collection of fresh manure. However, compost samples were collected in 20 L bucket from vertical trisection of compost pile with 1 to 1.8 m of height.

Collected samples were stored at −4°C until analysed. For total nitrogen (TN) and phosphorus (TP), samples were mixed with strong acid and heated in the block digester (BD40, LaChat, Milwaukee, WI, USA) at 320°C. Then, the analyses were conducted using the auto water analyser (QuikChem 8500, LaChat, Loveland, CO, USA). Total solids were determined by comparison of the sample weight before and after drying oven at 105°C overnight. The determination of volatile solids (VS) as organic matter was carried out after a high temperature muffle furnace at 550°C for over 4 hours [9]. The bulk density of the samples was weight per unit volume, which was measured with 18-L of bucket on site following the methods of Cooperband [10].

The nutrient loading refers to the nutrients from dairy cattle manure that released into the environment, specially soil and streams. However, livestock manure passes through a composting or storing period prior to land application. Through the composting or storing period, the oxidation of organic matter and N stripping or transformation occurs, which leads to a loss of nutrients. Hence, the amount of nutrient loading can be calculated by multiplying the total nutrients in manure with NLCs, which are considered to account for nutrient loss during the composting or storing period. Hence, NLCs as a dimensionless number represent the opposite of nutrient loss ratio.

Where above, NLCs represents nutrient loading coefficients.

The nutrient loss rate during the composting or storing period can be calculated by dividing the nutrient quantities in compost by those of manure.

In order to obtain nutrient quantities in compost and manure, the concentrations of VS, TN, and TP were multiplied by the amount of manure and compost, respectively, but the whole quantities of manure could not be measured directly on site. Thus, the value from Ministry of environment displays of 37.7 L/head/d for Holstein was used for NLC calculation [11].

Regardless, the measurement of compost production was not feasible directly on site. Hence, the amount of compost production was assessed after consideration of three different cases to understand nutrient loading after manure production. The first experimental operations were carried out to obtain the weight reduction during the composting or storing period (case I) and were conducted at the research and development ranch of Kangwon National University, South Korea (39°56′N, 127°46′E) for 4 months. First, the floors of 3 barns were covered by 15 cm of sawdust to prevent the cattle from being affected by the cold and also to remove moisture, simulating conditions widely implemented in over 90% of dairy farms for the management of manure and cattle in Korea. A total of 12 Holstein in growing, milking, or dry stages (429±112 kg of initial body weight) were selected and allocated to barns. Each barn had 4 Holstein, allowing for 6.4, 13.5, and 16.5 m2/head of stocking density for growing, milking and dry stages, respectively, in order to meet national standard for sustainable livestock management [12]. The mixture of manure and sawdust was stepped on and turned over by cows for 3 months, after which the mixture was retrieved from the barns and stored in a compost lot for 1 month. According to the Korean Rural Development Administration, the daily manure production of Holstein was determined to account for 5.6%, 6.5%, and 10.3% of body weight in the growing, milking, and dry stages, respectively, values that were adopted for the calculation of case I. Case II was the reference experiment to determine volume reduction during the composting period. Finally, case III was calculated based on the amount of P. One of the important components in nutrients, P, should theoretically not change during the composting period. Practically, however, P loss during the composting period occurs via run-off and leachate [13], which is additional nutrient loading to the environment, whereby constant amount of P amount before and after composting can be counted. The concentrations of each nutrient in the manure and compost samples were collected and directly analysed and the amount of P in manure calculated using the total amount of manure produced based on the documents from Ministry of Environment. The total compost produced was based on the amount and concentration of P found after analysis of compost samples. Thus, manure production from Ministry of Environment was combined with the results from case II and III in order to calculate NLCs on land.

The main livestock categories in Korea are beef and dairy cattle, pig, broiler, and layers which produced manure of 44.8 million m³/yr in 2013. Ministry of Environment in Korea announced the excretion coefficients in each livestock categories in 2008 and a head of dairy cattle produces manure of 37.7 L/d. The number of dairy cattle occupied only 3% of four major livestock categories (Hanwoo, dairy cattle, pig, and chicken) in Korea (the number of chicken was counted in thousand fold) but its manure production contributed 12.9%. In addition, the daily nutrient production of N and P in dairy cattle manure are the highest of 161.8 and 56.7 g/head over the other livestock categories and the yearly nutrient production based on the dairy cattle numbers were 8.0% and 7.0% of the N and P production from the entire livestock in Korea.

In total 41 dairy farms were selected to investigate manure management and nutrient loading on land as well as farm animal databases. The head of cattle ranged from 12 to 140 with an average of 64, which was less than the average of 75 for all of Korea [14]. All dairy farms used sawdust as the bedding material with the number of time which was replaced ranging from 1 to 8/yr (average 3.2/yr), and the average consumption of bedding materials per head was 1.03 kg/d (Table 1). The thickness of the bedding material was 15 cm and it was replaced when it reached 60% to 70% moisture as determined by the farmers. In relation to composting methods, only 12% of the entire dairy farms conducted turning using a skid-loader or tractor while the remaining ones heaped simple static piles without any air-supply.

The samples of manure and compost collected from the 41 dairy farms were analysed to determine the total VS, TN, and TP and to calculate the amount of each nutrient in each (Table 2). The TN concentration (5,086 mg/kg) in manure was close to that from Ministry of Environment (5,203 mg/kg), while the TP concentration was much lower (1,289 mg/kg) than the Ministry of Environment value (1,823 mg/kg). On the other hand, a much higher concentration of TN (8,897 mg/kg) and TP (2,186 mg/kg) was found in samples collected from the composting lot. The VS content of 10.8% in manure increased to 21.4% when manure was housed in the barn and composting. This was probably because the bedding materials such as sawdust and rice husk mostly counted as VS were mixed with manure even though the VS was degradable during the storage and composting periods.

As previously described, the cases for weight reduction of dairy manure were calculated. In order to obtain practical values for Case I, the measurements were conducted in two stages; i.e., the excretion from cattle was housed in the barn and rolled with bedding materials for 3 months, after which point the mixture was stored in a composting lot for 1 month. The total reduction in weight was measured for 4 months, for which 3 months the manure was in the barn and 1 in the composting lot (Table 3). A final value of mixture weight loss of 58.8% was observed while the mixture of dairy cattle manure and bedding materials left in cattle barns for 3 months. For this experiment, the entire mixtures of manure and bedding materials were weighed from cattle barns, after which point the mixture was stored in a composting lot for 1 month, resulting in weight loss of 22.7%. The overall weight loss of the mixture was 68.1% for 4 months.

The amount of P before and after composting should not change theoretically without the leachate occurrence or run-off since all of the P remains in the compost pile in the form of ortho- or poly-phosphate in addition to other organic compounds. Thus, the weight reduction can be calculated as is described below and the weight loss of 60% on average was derived based on the constant P content as was revealed by case III (Table 4). It was found that weight loss was higher in the field study (68%) compared to the database from the Ministry of Environment (55%) or where P loss was zero (60%).

Where [P] indicates the concentration of P.

According to the data from cases related to weight reduction, the NLCs for VS, TN, and TP were calculated based on the nutrient concentrations from sample analyses (Table 5). The NLCs refer to the nutrient amount from the manure to be loaded on land and nutrient loss is the opposite means of NLCs as previously described.

According to case II, higher NLCs were found for VS (1.92), TN (0.78) and TP (0.84) whereas, the values of NLCs in case I were 1.48, 0.60, and 0.66 for VS, TN, and TP, respectively. The values of NLCs based on a reference based weight reduction might be overestimated compared to the practical investigation of this experiment. Considering agricultural land excluding N emission to atmosphere, case III could be a reasonable value to determine the impacts of nutrients from dairy cattle manure in Korea. However, with respect to the nutrient loading only on agricultural land as soil budget for nutrients, the values of NLCs in case I would be reasonable.

Since we found that the average moisture contents of cattle manure and bedding materials were approximately 81.6% and 20.0%, respectively, 11.5 kg bedding material/head/d is required to obtain an optimal moisture content for composting (approximately 65%) assuming an average daily manure production of 37.7 kg/d as obtained from the Ministry of Environment document [15,16]. However, most dairy farmers used only 1.03 kg/head/d of bedding materials on average across the existing management practices. Assuming that there were no other factors affecting the mixture, such as ambient drying and the addition of water, the moisture content in the mixture was predicted to be 79.7% with the investigated amount of bedding materials used (1.03 kg/head/d). This highly moistened condition inhibited oxygen dispersal and thereby dead zones (anaerobic condition) were formed in the static piles [17]. Composting piles under anaerobic conditions take longer to decompose and a temperature of 55°C cannot be reached for pathogen destruction, which results in lower compost qualities [18]. The moisture content decreased naturally in the cattle barn; however, it was not possible to reach the optimal moisture level of 65% only with ambient drying and the amount of bedding materials investigated. In order to perform composting normally on dairy farms, sufficient bedding materials are required with the sufficient budget. Thus, frequent turning of the composting pile may help decrease the moisture content. Higher concentrations of TN, TP, and VS in dairy cattle manure after storage and composting might be due to the reduction in moisture content and the components that were condensed with weight loss [19,20]. Sommer [7] also found a higher TP concentration (34%) in cattle manure after 132 days of composting. This might have occurred due to moisture reduction and organic matter degradation during composting.

The previous case II from the study by Cooperband [10] revealed that volume reductions of 55%, 73%, and 55% of cattle manure occurred when utilizing the composting methods of simple static pile, windrow or turning, and aeration, respectively. In the present study, 36 farms used the simple static and 5 utilized the turning pile method and the weight reduction of each farm was calculated from the volume reduction using the bulk density of the samples. Weight reduction occurred due to moisture loss and microbial degradation during the composting period. Although the majority of the existing management practices followed a simple static pile method (88%), aerobic turning was also used (12%), and in the instances where there was a higher weight reduction (68%) than expected, it could have been due to low humidity, temperature, or other environmental causes. Average organic matter degradation was measured to be 53.3% of the total in each experiment during the 4 months of storage and composting and might be due to processes, including CO₂, CH₄, and N₂O emission, as well as NH₃ volatilization, nitrification, and denitrification due to the action of microbes. Further, complex compounds in manure were transformed into their most simple and stable forms during the composting process.

The larger values of NLCs for VS (1.48 to 2.29) were produced by the addition of bedding materials. The lower TN (0.60 to 0.87) values indicated that 13% to 40% of this element was reduced, presumably by ammonia volatilization, nitrous oxide, N gas emission and leachate [6,21–23]. Except for in case III, where no change in P concentration was observed, the reduction of TP was 34% and 16% in cases I and II, respectively. Theoretically, changes in P concentration occur if there is leachate or run-off. In this regard, case III may be the best for the calculation of total nutrient loading only excluding nutrient emission to atmosphere. However, the TN reduction in case III was only 13%, which indicates that most of the TN is retained in compost and ultimately imparted to the land. Inactive composting might cause such a low reduction of TN, which meant that no turning had taken place or there was insufficient addition of bedding material. The nutrient reduction via composting dairy cattle manure was previously reported as VS values from 29% to 81% and TN from 5% to 43%, although it was dependent upon the regional climate and composting methods [24–27]. Therefore, the values of NLCs obtained (VS, 1.48; TN, 0.60; TP, 0.66) from Case I experiment may accurately reflect the nutrient loading on agricultural land. Combining these derived NLCs with the data from the statistical report in Korea, the amounts of TN and TP from dairy cattle to agricultural land would be 15,262 and 5,882 tons per year, respectively, based on a dairy cattle population of 430,678 heads [14]. However, without considering nutrient loss during composting or storage, the nutrient values in manure would be directly applied to the yearly production of nutrients resulting in 25,437 tons of TN and 8,912 tons of TP, which are values 1.6 and 1.5 fold higher, respectively.

Although in the present study, we took into account the characteristics of manure and compost, utilization of bedding materials, and weight reduction during the storage and composting periods, and found that cattle manure is often mixed with insufficient amounts of bedding materials, then stacked in the composting lot without proper handling or even sufficient time before being applied to land as fertilizer, which causes an increased nutrient loading on soil and unpleasant odour in the surrounding areas. It has been proven that cattle manure needs a longer time for decomposition compared to pig or poultry due to the presence of large amount of fibrous materials [28]. Less usage of bedding material lowers the porosity in the composting pile, leading to an insufficient air supply and microbial degradation of the mixture during composting, and ultimately in high moisture in the dairy cattle manure compost in Korea. Addition of sufficient bedding materials in the dairy barn and longer composting periods along with aeration and turning are necessary to increase the compost quality of dairy cattle manure, which may ensure lower NLCs. Thus, it is mandatory and urgent to improve the management of dairy cattle manure in Korea.

Consequently, to calculate the amount of nutrients with respect to soil budget which are being loaded on agricultural land, it is desirable to use the NLCs values, [VS, 1.48; TN, 0.60; TP, 0.66] obtained from this experiment regarding soil N budget. The NLCs developed here may be helpful for manure handling practices and beneficial for their application in agricultural land management as well as for the accurate calculation of nutrient in soil budget.

Further, when compared with animal heads in livestock categories, the number of dairy cattle occupies only 3% but 12.9% in the contribution of manure production as previously mentioned, which indicates dairy cattle produces the highest amount of manure per head as previously mentioned. The management of dairy cattle manure is more difficult than for the manure of other livestock categories due to its high moisture content. This is paradoxical since swine manure is characterized by higher moisture content than dairy cattle manure. However, most swine farms in Korea have solid and liquid separators while dairy farms do not, which results in relatively higher cost for purchasing bedding materials on dairy farms. Thus, the development of methods or protocols to decrease the moisture content in dairy cattle manure for proper composting is necessary and utilization is required in the near future.