Introduction

World population is now 7 billion 200 million; by 2050 it will be 9 billion 100 million (UNFPA, 2013). Mexico has 112 336 538 inhabitants and a growth rate of 1.8 % (^{INEGI, 2010}). Food production increases due to continuous population growth. Moreover, the need for food from animals would duplicate to satisfy the demand of a larger population, urbanization, cultural changes and higher incomes (^{Steinfeld et al., 2009}). For this reason, intensive livestock production systems were developed and they contribute 65 % of the domestic dairy production; the remaining 35 % is produced in small-scale systems with herds of fewer than 100 cows (^{SIAP, 2011}).

Mexico holds 14^th place in world milk production with 10 724.30 million L and a population of 2 382 440 dairy cows on an area of 1 031 660 km² (^{FAO, 2011}). San Luis Potosí has 16 023 cows that produce 128.77 million L and holds 17 ^th place in national milk production. In the Valley of San Luis Potosí, 2300 cows produce 17-79 million L of milk (^{SIAP, 2011}).

Livestock production systems produce global impact because they emit greenhouse gases (GHG). Of these gases, 9 % is carbon dioxide (CO₂) from cattle's respiration; 37 % is methane (CH₄), whose effect on global warming is 25 times greater than that of CO₂, from enteric fermentation and from handling manure; and 65 % is nitrous oxide (N₂0) resulting from the process of nitrification and denitrification during aerobic decomposition of manure. Nitrous oxide potentially contributes 298 times more to global warming than CO₂ (^{IPCC, 2014}; ^{Steinfeld et al., 2009}).

Cattle production of CH₄ is a loss of 2 to 12 % of the feed energy (^{Johnson and Johnson, 1995}) and correlates with dry matter intake, digestibility and motility of the diet. Methane is generated mainly by ruminai fermentation of the feed and the hydrogen (H₂) surpluses are used by methanogenic bacteria to reduce CO₂ to CH₄, which is emitted when the ruminants belch. Therefore, the objective of this study was to characterize Holstein cows production systems of San Luis Potosí Valley, Mexico, and identify the differences among them in terms of milk production, feed conversion efficiency and methane emissions.

Materials and Methods

The study area is located west of the Sierra Madre Oriental between 21° 57' and 22° 40' N, and 100° 44' and 101° 11' w, where it forms a shallow closed basin with an extension of 1980 km² (^{CNA, 2002}). Mean annual temperature and precipitation are 16.9 °С and 416 mm, respectively (^{SMN, 2014}).

Information was collected from 15 of 35 production units selected at random. Herds varied from 20 to 100 Holstein milking cows in confinement. The analyzed variables were herd size, number of producing cows, average milk production, percentage of protein and fat in milk, live weight, dry matter (DM) intake, forage:concentrate ratio and manure management. To calculate the factor enteric methane CH₄ emission and manure management, feed use efficiency was defined as liters of milk produced per kg DM ingested per day. Methane emission intensity was calculated as the amount of enteric CH₄ emitted and methane emitted by manure management per liter of milk produced (^{Leslie et al., 2008}), for which milk production per year was adjusted to 305 d of lactation. The production systems were classified by descriptive statistics and cluster analysis with the variables.

Dry matter intake (kg d ^-1 cow ^-1) and total energy (Mcal d ^-1 cow ^-1) was calculated with ^{NRC (2001)} software.

Methane emission was determined based on Intergovernmental Panel on climate change, 2006, level 2 (^{Dong et al, 2006}). Equation (1) was used for CH₄ emission from enteric fermentation, equation (2) for CH₄ emission from manure management, and equation (3) to estimate the quantity of volatile solids excreted:

where FE _e CH₄ is the enteric CH₄ emission factor kg cow ^-1 year ^-1; EB is total energy intake, MJ d ^-1 cow ^-1; Y _m is the factor of conversion into CH₄, percentage of total energy of feed converted to CH₄, 6.5 % ± 1 %; the factor 55.65 (MJ kg^-1 CH₄) is the energy content of CH₄.

where FE _h CH₄ is the manure management CH₄ emission factor, kg cow ^-1 year^-1 SV are volatile solids excreted per day; B ₀ is the maximum production of CH₄ from manure, 0.188 m³ CH₄/SV, 0.67 is the conversion factor of m³ CH₄ to kg СН₄; MCF _S,k are factors of CH₄ conversion for each system of manure management (s) per climate region (k), =1.5 %; MS _(S,k) is the fraction of manure managed in the system (5) in climate region (k), dimensionless = 0.1.

where SV are volatile solids excreted per day; EB is total energy per day (MJ d ^-1 cow ^-1).

Live weight was calculated following the Quetelet methodology (^{Ávila and Gutiérrez, 2010}) with body length and thoracic perimeter (equation 4) :

where PV is the animal live weight; PT is thoracic perimeter; L is body length; C is the constant for females (87.5); constant for males (99).

Results and Discussion

The structure of the 15 dairy cattle herds in San Luis Potosí Valley was 49 % milking cows, 8 % dry cows and 43 % replacement cows and others (Table 1).

Table 1 Herd structure and methane emission in dairy milk production systems in San Luis Potosí Valley, México. (n=15). 

PV: Live weight, obtained with the equation PV = (PT) ² *L*C (^{Ávila and Gutiérrez, 2010}).

For milking cows, eb of the ingested feed was 17-25±3-94 Meal d ^-1 per cow, excreted volatile solids was 3-24±0-74 kg d^-1 per cow, dm intake was 13-45 ±1.76 kg d ^-1 per cow, and the enteric fermentation CH4 emission factor was 51-27± 11.71 kg year ¹ per cow. The last factor was lower (121 kg year ^-1 per cow) than that indicated by the IPCC (2006) level 1 for North America, where milk production (8400 kg year ^-1 per cow) is higher than production in the study area (3774.07± 1392-46 kg year ^-1 per cow). This is due to feed digestibility and the forage:concentrate ratio, which is higher for the diet in North America. Moreover, dm intake and feed conversion efficiency is higher and genetic quality of the cows is better.

In the enteric methane emission factors, there were differences (p≤0.05) in function of the physiological state of the cows. This was attributed to the higher energy requirements of these cows leading to an increase in dm intake and excretion of volatile solids (Table 2). This shows that current inventories of CH₄ are overestimated since they use a single enteric CH₄ emission factor (72 kg year ^-1 per cow) for all cows without taking into account their physiological stage.

Table 2 Factors of enteric methane emission and manille management in dairy milk production systems in San Luis Potosí Valley, México. (n=15). 

IMS: DM intake; ЕВ: total energy; FE _e . Factor of enteric CH₄ emission, calculated with the equation FE _e CH₄ = EB * (y _m /100) * 365 / (55.65);FE _h : Factor of CH₄ emission from manure, calculated with the equation FE _h CH₄ = (SV * 365) * [B ₀ * 0.67 kg m^-3 * Σ _s,k MCF _s,k , / 100 * MS _(s,k) ]; SV: Volatile solids, calculated with the equation SV = (ЕВ * 0.5 + (0.04 * EB)) * (0.92 /18.45) (^{Dong et al, 2006}).

The CH₄ emission factor (Table 2) for manure management was 0.22±0.05 kg year ^-1 per cow, which was lower than that established by IPCC, level 1, for North America (71 kg year ^-1 per cow) and Latin America (1 kg year ¹ per cow). The FE _h CH₄ found coincides with findings of ^{González and Ruiz (2007}) in Mexico (0.21 kg year ^-1 per cow).

Current FAO CH₄ inventories consider an enteric CH₄ emission factor of 72 kg year ^-1 per cow, which results in an emission of 67 392 kg for the 936 cows of the 15 production herds of the study area, and comparing that amount with the 35 090.42±6668.27 kg obtained in our study, there is an overestimation of 60 to 240 %. According to the CH₄ factor for management of the 936 kg, compared with 151-09±28.24 kg, it is five to seven times higher than that obtained in our study (Table 1).

The cluster analysis defined three groups (Table 3). Among these groups, there were no significant differences in herd size, milking cows, live weight or protein and fat content in milk. For the low feed efficiency cows (Table 3), milk production was lower (6.25±2.19 L) and CH₄ emission intensity was higher (19-59±4.10 g L ^-1). In these systems, milking is done with portable equipment and family labor. Feed production area is small (1.5±2.12 ha), diet is a 95:5 forage:concentrate ratio. Forage consists of fresh alfalfa (Medicago sativa), maguey (Agave salmiana) and sorghum (Sorghum bicolor L.), and the concentrate is a mixture of 40 % poultry manure, 20 % maize grain, 20 % bran and 20 % commercial concentrate.

Table 3 Variables for the cluster analysis of dairy milk production systems in San Luis Potosí Valley, México. (n=15). 

*Low (0.55±0.13). Intermediate (0.87±0.18). High (1.21±0.12). IMS: DM intake. Means with different letters in a row are statistically different (Tukey, p≤0.05).

Mean feed efficiency (Table 3) is defined by a milk production of 11.58±2.69 L and a CH₄ emission intensity of 14.44±2.3 g L ^-1. There were no differences (p>0.05) between the low feed efficiency group and the intermediate feed efficiency in DM intake, forage:concentrate ratio, EB consumed by producing cows or the enteric emission factor. Likewise, the intermediate feed efficiency group was not different (p>0.05) from high feed efficiency group in terms of the forage: concentrate ratio, but it had a lower (p≤0.05) milk production and CH₄ emission intensity than the high feed efficiency group. The intermediate efficiency group was characterized by a diet based on fresh alfalfa and commercial concentrate, with a forage:concentrate ratio of 66:34 and a higher (p≤0.05) DM intake, 13.20≤0.83 kg d ^-1 per cow ^-1, as compared to the low feed efficiency group. The group with the highest feed efficiency had the lowest (p<0.05) CH4 emission intensity (11-74±1.51 g L ^-1), the diet with the highest digestibility (p≤0.05) and a forage:concentrate ratio of 55:45; the forage consisting of maize silage, sorghum silage, fresh alfalfa and alfalfa hay, and the concentrate is a mixture of 75 % commercial concentrate and 25 % maize grain. Dry matter intake was 15·87±1·72 kg d ¹ per cow; milk production increased, and CH₄ per liter of milk produced decreased.

The Pearson correlation analysis showed that CH₄ emission intensity correlates negatively (r=-0.92) with feed efficiency; that is, CH₄ emission intensity increases and feed efficiency decreases.

The simple linear regression analysis (equation 5) shows:

where IECH4 is the methane emission intensity; EA is feed efficiency.

The increment of dairy cow productivity may be the most successful strategy for reducing CH₄ emission intensity. To this end, feeding factors that affect DM intake and milk production should be taken into account. Thus, diet composition, physical form, moisture, forage:concentrate ratio and genetic quality of the cows can increase or decrease milk production. ^{Hristov (2013}) stated that diets with more than 40 % concentrate decrease enteric CH₄ emissions; this was observed in our study: the high efficiency cows had a 65 % lower CH₄ emission intensity than the low efficiency cows.

The increase in forage digestibility and intake of digestible forage reduce CH₄ emissions from enteric fermentation and manure management (^{Afshar and Naser, 2011}). Moreover, according to ^{Hassanat et al. (2012}) and ^{Lettat et al. (2013}), substitution of grass or alfalfa silage for maize silage reduces methane production; when the concentration of propionate increases, pH and concentrations of acetate and butyrate decrease, and protozoa populations decrease, producing hydrogen and maintaining a symbiotic relationship with the methanogenic bacteria.

Conclusions

The factors enteric methane emission and manure management differ in function of the physiological state of the cows. Methane emission intensity correlates negatively with feed efficiency and therefore, the current methane inventories do not reflect its real environmental impact. The methane emission inventories should thus be carried out in accord with the structure of the herds and the physiological state of the cows.

Duplication of total energy intake, increasing dry matter intake relative to live weight, as well as the concentrate in diets for low feed efficiency cows, could improve milk production and decrease negative impacts of methane emission per liter of milk produced.