Microsoft Word - 26jonassen.docx CHEMICAL ENGINEERING TRANSACTIONS VOL. 40, 2014 A publication of The Italian Association of Chemical Engineering www.aidic.it/cet Guest Editor: Renato Del Rosso Copyright © 2014, AIDIC Servizi S.r.l., ISBN 978-88-95608-31-0; ISSN 2283-9216 Modelling of Odour Emission Rates for Fattening Pigs as Time-resolved Input for Dispersion Models Günther Schaubergera*, Martin Piringer b, Albert J. Heber c aWG Environmental Health, Unit for Physiology and Biophysics, University of Veterinary Medicine, Vienna, Austria bCentral Institute of Meteorology and Geodynamics, Vienna, Austria cAgricultural and Biological Engineering Department, Purdue University, West Lafayette, Indiana * gunther.schauberger@vetmeduni.ac.at For the calculation of the separation distance to avoid odour annoyance, the odour emission rate of the livestock building has to be known as input for the dispersion model. In general, an annual mean value is used although the influence of the live mass growth of the animals as well as of indoor air temperature and ventilation rate on the emission rate is known. This simplified approach was compared with various emission scenarios for a continuous fattening system with a constant live mass of 75 kg and an all-in/all- out system. For the last one a Monte-Carlo based model (inverse transfer sampling technique) was used to avoid an interaction between the growth of the animals and the annual variation of the outdoor temperature. The variation of the emission factor was taken into account twofold, first by a schematized diurnal emission pattern for the various seasons and second by a steady state simulation of the indoor climate to calculate indoor temperature and ventilation rate which both influence the odour release. The results indicate an underestimation of the odour emission rate of a livestock building during summer compared with winter when using an annual mean value. For the all-in/all-out system, this effect is superposed by an overestimation at the beginning of the fattening period and an underestimation at the end. Using emission models which take into account the growth function and/or the indoor climate, a more realistic description of the odour emission characteristics can be achieved. This will help to increase the goodness of the assessment of environmental odour by dispersion models. 1. Introduction For dispersion modelling of environmental odour, the emission rate is typically estimated as an annual constant value (e.g. Guingand, 2003; Hayes et al., 2006; Nicolas et al., 2008; VDI 3894 Part 1, 2011). In general this annual mean is calculated as a product of the mean live mass of the animals and the odour emission factor. In reality, however, the two parameters show distinct variations over time. This was discussed in detail in Schauberger et al. (2013a) for fattening pigs. Therefore the use of an annual mean to represent livestock odour emission rates is inappropriate for dispersion model inputs. For fattening pigs, an emission model is used with the indoor temperature, the ventilation rate which are calculated by a simulation model (Schauberger et al., 2000b) as well as the animal activity as predictors (Schauberger et al., 2013a). As a reference, a constant odour emission rate is used, which is conventionally calculated by an annual mean live mass and a constant odour emission factor. Then, step by step, this constant value is substituted by a live mass which increases during the fattening period using an animal growth model and/or the emission model which takes into account the influence of the indoor temperature, ventilation rate, and animal activity. Six emission scenarios with increasing complexity are thus obtained and presented in the paper. DOI: 10.3303/CET1440009 Please cite this article as: Schauberger G., Piringer M., Heber A.J., 2014, Modelling of odour emission rates for fattening pigs as time- resolved input for dispersion models, Chemical Engineering Transactions, 40, 49-54 DOI: 10.3303/CET1440009 49 2. Materials and Methods 2.1 Growth model For an all-in/all-out (AIAO) production system, an animal growth model describes the increase of emissions by the growing of the animal live mass of the herd. If the AIAO production system is applied only to individual pens within the building, a constant mean live mass value for the entire livestock building is used over time. This is called the continuous flow production system (CONT). We used a growth model with a constant average daily gain (linear increase) of the live mass m (kg) as a function of time t (days), applied here to fattening pigs reared in an AIAO system. The live mass values at the beginning and end of the fattening period were selected to be mstart = 30 kg and mend = 120 kg, respectively. The average daily gain is assumed to be ADG = 0.780 kg d-1. This results in a duration of the fattening period of tf = 116 d. The duration between two consecutive fattening periods, when the livestock building is empty, is assumed as te = 10 d. In this period, the odour emission is set to zero. Hence the overall duration of a fattening period is tp = tf + te with 126 d or 18 weeks. The growth function is given by startm m AD G t= + . The time course of the live mass of fattening pigs behaves like a saw tooth wave with a period duration of 18 weeks (about 1/3 year). These growth periods are superimposed and interact with the time course of the outdoor temperature. To avoid this interaction we calculate the live mass on the basis of a Monte Carlo method, called inverse transform sampling, a useful tool for environmental sciences (e.g. Schauberger et al., 2013b; Wilks, 2011). This method was applied to the growth function (Schauberger et al., 2014). For each half hour mean value, a live mass m for AIAO pig production systems is calculated using this Monte- Carlo approach. 2.2 Simulation of the indoor climate We adapted a simulation model for the indoor climate of a livestock building which calculates inside air temperature and the ventilation rate. The model is reduced to the sensible heat balance of a livestock building in a moderate climate (Schauberger et al., 2001). To reduce the complexity of the simulation, the influence of the latent heat balance to the sensible heat balance is neglected. The two parameters indoor air temperature Θi (equal to the temperature of the exhaust air) and the volumetric ventilation rate V are calculated as a function of the outdoor temperature under the assumption of steady-state conditions. The simulations were conducted for an AIAO system, taking into account the growth of the animals, and for a continuous flow production system CONT with a constant live mass of m = 75 kg. 2.3 Meteorological data The air temperature for the simulation of the indoor climate was measured at Wels, a site representative of the Austrian flatlands north of the Alps, with a temporal resolution of 30 minutes over a 2 year period between January 30, 1992 (JD=30) and January 31, 1994 (JD=761). 2.4 Odour release model In general, the odour emission rate of a livestock building is calculated by a live mass specific emission factor (ou s-1 LU-1) and the total live mass (LU, 1 LU is equivalent to 500 kg) of the animals inside the building. The odour release is also modified by the indoor climate (e.g. temperature, ventilation rate and time of the day) of the livestock building described in the Sections above. Therefore we developed an odour release model to describe the modifications based on these parameters. The live mass specific odour emission factor e (ou s-1 LU-1) is calculated by multiplying the standardized live mass specific odour emission factor e0, by the release modification factor R, which modifies the odour emission due to the indoor climate 0e e R= ⋅ (1) In this paper two different release modification factors R were selected. The first one R1 is derived from Nicholas et al. (2002) in a schematic way using rectangular functions with different durations of the day/night cycles and different emission levels to describe the diurnal variations for winter, summer, and spring/autumn (Figure 1). 50 Figure 1. Diurnal course of the release modification factor R1 for winter, spring/autumn, and summer time derived from Nicholas et al. (2002). Second we use the emission model of Schauberger et al. (2013a) to calculate the release modification factor R2 on the basis of the indoor air temperature Θi, ventilation rate V, and physical activity of animals A as a function of daytime t. The release modification factor R2 is calculated by 2 V AR F F FΘ= (2) with FΘ for indoor temperature, FV for ventilation rate and FA for relative animal activity. The release modification factor R2 is thus given by ( )( ) ( )1 0.318 2 0 2 sin (2 1) exp 0.0314 1 0.25 2 1i R n i i t R V i π ϕ τ =   ⋅ + ⋅ +     = Θ − Θ ⋅ ⋅ + +       (3) with the reference temperature ΘR = 20°C, the the normalized ventilation rate Vn per animal place. The ventilation rate per animal place V is normalized to unity by Vd, according to Vn = V / Vd. This ventilation rate is the maximum ventilation rate for growing pigs of Vd = 200 m 3 h-1 (55.6 10-3 m³ s-1) per animal place (MWPS-32, 1990). The relative animal activity depends on the period τ = 24 h, time of day t (h), and the time lag ϕ = -6 h. 2.5 Odour emission scenarios The odour emission rate E (ou s-1) is calculated for one animal place with a time resolution of one half hour according to 0E e R m= ⋅ ⋅ (4) with the standard live mass specific odor emission factor e0, the live mass of one animal m (LU; 1 LU = 500 kg), and the release modification factor R. The entire emission rate of the livestock building can be calculated by multiplying this emission rate E by the number of animals N. Table 1. Input parameters for odour emission scenarios with growing complexity to calculate the odour emission rate E (ou s-1) per animal place by the standard live mass specific odour emission factor e0 (ou s-1 LU-1), the release modification factor R, and CONT (m = 75 kg) and AIAO pig production systems. Scenario Release modification factor R Pig production system SC0 R = 1 CONT SCm R = 1 AIAO SCR1 R1 = f (season, t) CONT SCR1+m R1 = f (season, t) AIAO SCR2 R2 = f (Θi, V, A) CONT SCR2+m R2 = f (Θi, V, A) AIAO The input parameters of the scenarios are summarized in Table 1. The standard live mass specific odour emission factor e0 is selected from Schauberger et al. (2013a) with e0 = 48 OU s -1 LU-1. For the first two emission scenarios, the release modification factor is set to R = 1. For scenario SC0 a continuous pig 0.0 0.5 1.0 1.5 R el ea se m od ifi ca ti o n fa ct or R 1 0 2 4 6 8 10 12 14 16 18 20 22 24 Time (h) Summer Spring / Fall Winter 51 production system is assumed, which means that the live mass of the pigs is constant over the year with m = 75 kg or m = 0.15 LU. For scenario SCm, the odour emission rate per animal place is calculated for an AIAO pig production system where the live mass is calculated by the Monte Carlo based growth model (Section 2.1). The next four emission scenarios differ by the way the release modification factor is calculated, either R1 (Nicholas et al., 2002) or R2 (Schauberger et al., 2013a). Both are applied either to a continuous flow (m = 75 kg) or to an AIAO production system. a b c Figure 2 Time course of the odour emission rate E (ou s-1) per animal place for the six odour emission scenarios, calculated for the two year period (JD Julian day; 1.1.1992 = 0) (zero values during the service period are not displayed). (a) SC0: Continuous fattening: m = 75 kg and release modification factor R = 1, SCm: all-in/all-out and release modification factor R = 1, (b) SCR1: Continuous fattening: m = 75 kg and release modification factor R1 (Nicholas et al., 2002), SCR1+m: all-in/all-out and release modification factor R1, and (c) SCR2: Continuous fattening: m = 75 kg and release modification factor R2 (Schauberger et al., 2013a), SCR2+m: all-in/all-out and release modification factor R2. (Source: Schauberger et al. (2014)). The CONT emission scenarios are shown as black lines, the AIAO data as grey dots. 3. Results and Discussion The time course of the six odour emission scenarios (Table 1) with increasing complexity is shown in Figure 2. The grey dots represent the emission E, calculated by the Monte Carlo simulation. Therefore the values are randomly distributed according to the live mass. The corresponding cumulative distribution functions (CDFs) of the six scenarios are depicted in Figure 3. Whereas scenario SC0 with a constant live mass shows also a constant emission, the odour emission of scenario SCm changes with the changing live mass of the AIAO production system (Figure 2a). Neither the influence of the indoor climate nor the influence of the animal activity is considered. An improvement of the odour release was reached for scenario SCR1 (CONT) and SCR1+m (AIAO) by a schematised diurnal time pattern of the release modification factor for the winter, summer, and spring/autumn (Figure 1). The last two scenarios (Figure 2c) CSR2 and SCR2+m are calculated with the release modification factor R2, which relates the odour emission to the indoor climate of the livestock building. Therefore we combined the simulation of the indoor climate of a livestock building of fattening pigs (Schauberger et al., 2000b) with an odour emission model (Schauberger et al., 2013a) using indoor air temperature, ventilation rate as well as the animal activity as predictors. In this way, the most realistic yearly odour emission cycle is obtained, 0 365 730 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 O do ur e m is si on r at e pe r an im al p la ce E ( ou /s ) Time (JD) SC 0 CONT SC m AIAO 0 365 730 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 O do ur e m is si on r at e pe r an im al p la ce E ( ou /s ) Time (JD) SC R1 CONT SC R1+m AIAO 0 365 730 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 O do ur e m is si on r at e pe r an im al p la ce E ( ou /s ) Time (JD) SC R2 CONT SC R2+m AIAO 52 with low emissions in winter and higher emissions in summer. The variability of the AIAO production system is always larger than for the CONT system. a b Figure 3 Cumulative frequency of the odour emission rate E (ou s-1) per animal place for the continuous flow production system CONT (a) and the all-in all-out (AIAO) production system (b), calculated for the two year period. (a) CONT: SC0: m = 75 kg and release modification factor R = 1, SCR1: m = 75 kg and release modification factor R1 (Nicholas et al., 2002), SCR2: m = 75 kg and release modification factor R2 (Schauberger et al., 2013a). (b) AIAO: SC0: m = 75 kg and release modification factor R = 1 (as reference), SCm: animal growth model and release modification factor R = 1, SCR1+m: animal growth model and release modification factor R1, SCR2+m: animal growth model and release modification factor R2. (Source: Schauberger et al. (2014)) Even if the scenarios calculated with R1 and R2 do not show much difference in the shape of the CDFs, only the release modification factor R2, which is based on the indoor climate simulation, shows a high correlation with the meteorological situation, which is to be expected. The scenarios with the AIAO production system also deliver, in general, larger odour emissions and greater variability. The odour emission rate of a livestock building is a major input to the calculation of separation distances by dispersion models. These model calculations are generally based on hourly weather data (Capelli et al., 2013; Schauberger et al., 2000a; Yu et al., 2009). Nevertheless, most of the calculations of separation distances are performed with constant emission rates (e.g. Hayes et al. (2006); VDI 3894 Part 1 (2011)). Due to the lack of data, none of the existing setbacks or odour dispersion models consider diurnal and seasonal variations in odour emission rates, which may result in great uncertainty in the predicted results (Guo et al., 2007). Capelli et al. (2013) pointed out the importance of the influence of meteorological parameters on the odour emission rate. They suggested the recalculation of the emission rate of an area source on an hourly basis taking into account the influence of wind velocity on the area-specific emission factor. Sun et al. (2010) suggested the use of different emission rates on a monthly basis. For other farm animals as fattening pigs, such emission models are unavailable in detail. If empirical data on odour emission are missing then such odour emission models can also be based on other substances or gases which are closely related to odour (e.g. PM10 for turkeys (Li et al., 2008)), ammonia for poultry (Cheng et al., 2011). Even if such a model is only an educated guess, we are convinced that this will result in a better prediction quality compared with a constant emission rate. 4. Conclusions The odour emission rate of a livestock building is a major input to the calculation of separation distances by dispersion models. In general, only an annual mean value is used even if it is evident that the growth of the animals and the influence of the indoor climate on the odour release are neglected thereby. In this paper we use a constant emission rate with an annual mean only as a reference. In addition, emission scenarios with increasing complexity have been developed. Calculations were carried out for a continuous flow production system with a constant live mass during the year and an all-in/all-out system with a growing live mass during the fattening period. The emission scenarios which take into account the modification of the odour release due to the indoor climate show an increasing variability of the emissions year-round. During summer the odour release will be underestimated when an annual mean value is used. This effect will be amplified when high outside temperatures and the ends of the fattening periods coincide. In winter, using an annual mean value, the odour emission will be overestimated compared with variable 0 2 4 6 8 10 12 14 16 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 C um ul at iv e fr eq ue nc y f Odour emission factor per animal place E (ou E /s) SC 0 CONT SC R1 CONT SC R2 CONT 0 2 4 6 8 10 12 14 16 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 C um ul at iv e fr eq ue nc y f Odour emission factor per animal place E (ou E /s) SC 0 CONT SC m AIAO SC R1+m AIAO SC R2+m AIAO 53 emission scenarios. The step from an annual mean value to a more realistic emission scenario will thus increase the reliability of the calculation of separation distances to avoid odour annoyance. References Capelli, L., Sironi, S., Del Rosso, R., Guillot, J.-M., 2013. Measuring odours in the environment vs. dispersion modelling: a review. Atmos. Environ. 79, 731-743. Cheng, W.-H., Chou, M.-S., Tung, S.-C., 2011. Gaseous Ammonia Emission from Poultry Facilities in Taiwan Environmental Engineering Science 28, 283-289. Guingand, N., 2003. Air quality and reduction of slatted floor in growing-finishing pig units, Gaseous and Odour Emissions from Animal Production Facilities. CIGR, Horsens, DK, pp. 80-87. Guo, H., Dehod, W.D., Agnew, J., Feddes, J.R., Lague, C., Pang, S., 2007. Daytime odor emission variations from various swine barns. Transactions of the ASABE 50, 1365-1372. Hayes, E.T., Curran, T.P., Dodd, V.A., 2006. Odour and ammonia emissions from intensive pig units in Ireland. Bioresource Technology 97, 940-948. Li, H., Xin, H., Burns, R.T., Hoff, S.J., Harmon, J.D., Jacobson, L.D., Noll, S.L., 2008. Effects of bird activity, ventilation rate and humidity on concentration and emission rate of a turkey barn, Livestock Environment VIII - Proceedings of the 8th International Symposium, Iguassu Falls, pp. 111-116. MWPS-32, 1990. Mechanical ventilating systems for livestock housing. Midwest Plan Service, Iowa State University, Ames p. p72. Nicholas, P., McGahan, E., Watts, P., 2002. Application of emission data in modelling and assessment. FSA Environmental, Australian Pork Ltd, Toowoomba, Queensland. Nicolas, J., Delva, J., Cobut, P., Romain, A.-C., 2008. Development and validating procedure of a formula to calculate a minimum separation distance from piggeries and poultry facilities to sensitive receptors. Atmos. Environ. 42, 7087-7095. Schauberger, G., Lim, T.T., Ni, J.Q., Bundy, D.S., Haymore, B.L., Diehl, C.A., Duggirala, R.K., Heber, A.J., 2013a. Empirical model of odor emission from deep-pit swine finishing barns to derive a standardized odor emission factor. Atmospheric Environment 66, 84-90. Schauberger, G., Piringer, M., Baumann-Stanzer, K., Knauder, W., Petz, E., 2013b. Use of a Monte Carlo technique to complete a fragmented set of H2S emission rates from a waste water treatment plant Journal of Hazardous Materials 263, 694-701. Schauberger, G., Piringer, M., Heber, A.J., 2014. Odour emission scenarios for fattening pigs as input for dispersion models: A step from an annual mean value to time series. Agriculture, Ecosystems & Environment 193, 108-116. Schauberger, G., Piringer, M., Petz, E., 2000a. Diurnal and annual variation of the sensation distance of odour emitted by livestock buildings calculated by the Austrian odour dispersion model (AODM). Atmospheric Environment 34, 4839-4851. Schauberger, G., Piringer, M., Petz, E., 2000b. Steady-state balance model to calculate the indoor climate of livestock buildings, demonstrated for fattening pigs. International Journal of Biometeorology 43, 154- 162. Schauberger, G., Piringer, M., Petz, E., 2001. Separation distance to avoid odour nuisance due to livestock calculated by the Austrian odour dispersion model (AODM). Agriculture, Ecosystems and Environment 87, 13-28. Sun, G., Guo, H., Peterson, J., 2010. Seasonal odor, ammonia, hydrogen sulfide, and carbon dioxide concentrations and emissions from swine grower-finisher rooms. J. Air Waste Manag. Assoc. 60, 471- 480. VDI 3894 Part 1, 2011. Emissions and immissions from animal husbandry - Housing systems and emissions - Pigs, cattle, poultry, horses, in: Verein Deutscher Ingenieure (Ed.). Beuth, Berlin. Wilks, D.S., 2011. Statistical methods in the atmospheric sciences: an introduction. Acad. Press, San Diego, CA. Yu, Z., Guo, H., Xing, Y., Laguë, C., 2009. Setting acceptable odour criteria using steady-state and annual hourly weather data. Biosystems Engineering 103, 329-337. 54 Controfacciata.pdf Pagina vuota Pagina vuota Pagina vuota