H. Volden

Feed analyses and digestion methods

Feed characteristics are determined via chemical analyses and digestion methods. SpecificNorFor methods for determining parameters such as DM, sCP, iNDF and the in sacco methods are fully described in this chapter. Tables 5.1 and 5.2 present an overview of the feed analysis and digestion methods, respectively.

Estimation of rumen outflow in dairy cows fed grass silage-based diets by use of reticular sampling as an alternative to sampling from the omasal canal

A study was conducted to compare nutrient flows determined by a reticular sampling technique with those made by sampling digesta from the omasal canal. Six lactating dairy cows fitted with ruminal cannulas were used in a design with a 3 x 2 factorial arrangement of treatments and 4 periods. Treatments were 3 grass silages differing mainly in neutral detergent fiber (NDF) concentrations: 412, 530, or 639 g/kg of dry matter, each combined with 1 of 2 levels of concentrate feed. Digesta was collected from the reticulum and the omasal canal to represent a 24-h feeding cycle. Nutrient flow was calculated using the reconstitution system based on 3 markers (Co, Yb, and indigestible NDF) and using (15)N as a microbial marker. Large and small particles and the fluid phase were recovered from digesta collected at both sampling sites. Bacterial samples from the reticulum and the omasum were separated into liquid- and particle-associated bacteria. Reticular samples were sieved through a 1-mm sieve before isolation of digesta phases and bacteria. Composition of the large particle phase differed mainly in fiber content of the digesta obtained from the 2 sampling sites. Sampling site did not affect marker concentration in any of the phases with which the markers were primarily associated. The (15)N enrichment of bacterial samples did not differ between sampling sites. The reticular and omasal canal sampling techniques gave similar estimates of marker concentrations in reconstituted digesta, estimates of ruminal flow, and ruminal digestibility values for dry matter, organic matter, starch, and N. Sampling site x diet interactions were also not significant. Concentration of NDF was 2.2% higher in reconstituted omasal digesta than in reconstituted reticular digesta. Ruminal NDF digestibility was 2.7% higher when estimated by sampling the reticulum than by sampling the omasal canal. The higher estimate of ruminal NDF digestibility with the reticular sampling technique was due to differences in NDF concentration of reconstituted digesta. This study shows that nutrient and microbial protein outflow from the rumen can be measured using a reticular sampling technique. The reticular sampling technique provides a promising alternative to sampling from the omasal canal because there is less interference with the animal and it does not require advanced sampling equipment.

Digestion and metabolism in the gastrointestinal tract

In NorFor the digestion and metabolism are simulated in three compartments: (1) the rumen, (2) the small intestine and (3) the large intestine. This chapter describes the modelling of the digestion processes in the three compartments and the microbial OM synthesis in the rumen and large intestine. Most of the equations in this chapter could have been presented in a simpler form, but we present them here in the format they are implemented in the computer program since we believe this makes it easier to follow their biological rationale.

Energy and metabolizable protein supply

Several systems are used to measure energy (Van Es, 1978; Moller et al., 1983; INRA, 1989; NRC, 1989; AFRC, 1993) and metabolizable protein (MP) (Madsen, 1985; NRC, 1985; Verite et al., 1987; AFRC, 1992; Tamminga et al., 1994) supplies in cattle. Feed evaluation for cattle is in continues development and several new systems have been proposed (Sniffen et al., 1992; NRC, 2001; Thomas, 2004). The aim of the new systems is to refine and more carefully describe the interactions between the animal, its feed and the environment to predict performance. Also in the development of NorFor an important objective is to obtain improved estimates of the animal’s dietary energy and protein supplies by taking into account the effects of feeding level and diet composition on both ration digestibility and microbial activity.

Prediction of voluntary feed intake

Describing the nutrient variables in a diet and their interactions is an important part of ration formulation as productivity of the dairy cow is sensitive to the profile the nutrients absorbed. However, the prediction of feed intake is probably the most important determinant of production. Feed intake is primarily influencedby animal and feed characteristics. The most important animal characteristics are BW and physiological state, including stage of lactation, milk production, stage of gestation, live weight gain and body condition score. Feed characteristics such as digestibility and fibrecontent have both a strong influenceon rumen filland, hence, feed intake (Kristensen, 1983). However, several studies (Rinne et al., 2002; Garmo et al., 2007) have shown that cows may stop eating before the fillcapacity of the rumen is reached. This has been attributed to metabolic regulation (MR), which is also an important factor to consider when predicting feed intake. Physical intake regulation is related to the ruminal NDF pool (Bosch et al., 1992; Rinne et al., 2002) and is partly an indirect effect of the energy concentration of the diet because DMI declines in a curvilinear manner with increasing energy density of the diet (Mertens, 1994).

Animal requirements and recommendations

One of the challenges in developing the NorFor feed evaluation system has been to combine our sub-models with existing feed evaluation systems, i.e. combining NorFor’s digestion model with systems that predict energy requirements from digested nutrients. NorFor has used existing Dutch equations to predict energy requirements for milk production, maintenance and gestation for dairy cows, while French equations have been used to predict energy demand for growth. In contrast to energy requirements, NorFor has developed its own equations to determine the AATN requirements for milk production and growth. The recommendations for minerals and vitamins are mainly implemented from NRC.

Feed calculations in NorFor

NorFor has an extensive feed characterization program. In addition to the feed constituents described in Chapter 4, several feed characteristics are calculated from information based on chemical fractions, and their degradation and digestion characteristics. Nevertheless, efficientuse of a feed evaluation system in practice requires commercial feed analyses that are reliable and can be performed at a low cost. Therefore, several feed characteristics are predicted using either in vitro methods or NIRS.

Feed fraction characteristics

Absorbed nutrients are provided from fermentation and digestion in the gastrointestinal tract, and the predictions are sensitive to the nutrient profiles of the feed. NorFor has an extensive feed table (www.norfor.info) that lists chemical composition and digestion characteristics of typical Nordic feedstuffs. The feed table values are continuously updated as new information becomes available.

Relationship between passage rate and extrinsic diet characteristics derived from rumen evacuation studies performed with dairy cows

A meta-analysis based on experiments with dairy cows was conducted to study the effects of extrinsic diet characteristics on passage rate (Kp) of indigestible neutral detergent fibre (iNDF). A data set was collected that included 108 dietary treatment means from 29 studies. Dietary treatments consisted of different forages supplemented with varying amounts of concentrate feed. Minimum prerequisite for an experiment to be included in the analysis was that Kp was calculated using the flux/compartmental pool method, and that live weight (LW), total and forage dry matter intake (DMI), neutral detergent fibre (NDF) and iNDF diet concentrations were given or could be estimated. Total DMI and intake of NDF (NDFI), proportion of concentrate in diet (CProp), and milk yield in the data varied from: 8.71 to 39.0 g/kg LW, 3.76 to 16.6 g/kg LW, 0 to 0.693, and 0 to 35 kg/d, respectively. A mixed model regression was used to analyse responses of fixed factors on Kp. Root mean square error (RMSE) for the models are given. The 3 best equations were: (1) Kp = 0.0159 + 0.000790 × NDFI (g/kg LW) (RMSE = 0.00295/h); (2) Kp = 0.0129 + 0.000755 × NDFI (g/kg LW) + 0.0189 × CProp (NDF basis) (RMSE = 0.0031/h); and (3) Kp = 0.00661 + 0.000938 × NDFI (g/kg LW) + 0.0220 × CProp (NDF basis) + 0.0114 × iNDF/NDF (RMSE = 0.00328/h). Equation 3 expressed on DM basis was: Kp = 0.00816 + 0.000400 × DMI (g/kg LW) + 0.00401 × CProp + 0.00908 × iNDF/NDF (RMSE = 0.00379/h). Model performance suggested that Kp is more closely related to NDFI than DMI. The positive coefficients for CProp on NDF basis and iNDF/NDF is related to the faster Kp of concentrate particles compared with forage particles and that Kp increases when NDF potential digestibility decreases.

Overall model description

The NorFor system is a semi-mechanistic, static and science-based model, which predicts nutrient supply and requirements for maintenance, milk production, growth and pregnancy in cattle. The model can be divided into five parts: (1) an input section describing characteristics of the animal and feeds available; (2) a module simulating processes in the digestive tract and the intermediary metabolism, termed the feed ration calculator (FRC); (3) a module predicting feed intake; (4) a module predicting the physical structure of the diet; and (5) an output section describing nutrient supply, nutrient balances and production responses (Figure 2.1).