3.1 Dairy systems description
The small cattle dairy systems studied were exclusively focused on milk production. In all farms, the cattle herd was predominantly composed of cows, accounting for 69 to 87% of the total animals, followed by female calves retained for replacing culled cows. Additionally, two farms had bulls for reproductive purposes (Table 3). The farms lacked indoor facilities and only had outdoor pens and/or chutes for cattle handling procedures. Calves were weaned and sold a few days after birth, while female calves were retained to replace culled cows and/or expand the herd. However, no change in the stock size on the farm was assumed, and the animal inventory remained the same in both the baseline and improvement scenarios. Key variables describing the farms are presented in Table 3. Cattle excretions (urine and dung) were not managed by the farmers and were directly deposited on pastures. Consequently, manure was not considered a farm co-product in this study, enabling the use of the IDF biophysical allocation method to assign GHGE between milk and LWG. Fertilization was practiced on all farms, while soil amendments were applied only on farms number two and three. All farms were situated in hilly and undulating terrain.
Table 3
Dairy systems description - cattle inventory, production rates, feeding supplementation, land uses, and fertilizers application rates in 3 provinces of Cundinamarca Department
|
Dairy cattle farms (n = 4)
|
|
Baseline scenario
|
Improvement scenario
|
|
Farm 1
|
Farm 2
|
Farm 3
|
Farm 4
|
Farm 1
|
Farm 2
|
Farm 3
|
Farm 4
|
Municipality
|
Cajibío
|
El Tambo
|
Sotará
|
Timbío
|
Cajibío
|
El Tambo
|
Sotará
|
Timbío
|
Herd (% of category in herd)
|
|
|
|
|
|
|
|
|
Cows, no
|
4 (87%)
|
5 (69%)
|
4 (86%)
|
4 (68%)
|
4 (87%)
|
5 (69%)
|
4 (86%)
|
4 (68%)
|
Female calves (0–1 year), no
|
2 (13%)
|
3 (13%)
|
2 (14%)
|
2 (10%)
|
2 (13%)
|
3 (13%)
|
2 (14%)
|
2 (10%)
|
Bulls, no
|
0
|
1 (19%)
|
0
|
1 (22%)
|
0
|
1 (19%)
|
0
|
1 (22%)
|
Milk (increase of milk production), kg FPCMa cow− 1 year− 1
|
1508.2
|
2450.8
|
1508.2
|
2262.3
|
2073.7 (+ 37.5%)
|
3393.4 (+ 38.5%)
|
2262.3 (+ 50%)
|
3016.4 (+ 33.3%)
|
Commercial concentrate, kg AU− 1b yr− 1
|
730
|
256
|
1095
|
183
|
584
|
255.5
|
365
|
182.5
|
Cereal bran, kg AU− 1 yr− 1
|
365
|
256
|
183
|
0
|
365
|
255.5
|
182.5
|
182.5
|
Molasses, kg AU− 1 yr− 1
|
---
|
65
|
---
|
---
|
---
|
65
|
---
|
---
|
Mineral salt, kg AU− 1 yr− 1
|
380.9
|
375.4
|
59.7
|
0.0
|
405.6
|
375.4
|
298.6
|
402.7
|
Gross energy, MJ day− 1 AU− 1
|
266.6
|
276.6
|
259.4
|
265.7
|
227.6
|
214.2
|
272.6
|
251.4
|
Dry Matter Intake, T DMI AU− 1 yr− 1
|
5.3
|
5.5
|
5.1
|
5.3
|
4.5
|
4.2
|
5.4
|
5.0
|
Land
|
|
|
|
|
|
|
|
|
Area, ha
|
5,5
|
2,29
|
11.0
|
20,9
|
5,5
|
2,29
|
11.0
|
20,9
|
IP, % of area
|
36.0
|
13.0
|
2.0
|
0.0
|
82.0
|
98.0
|
30.0
|
53.0
|
SPS, % of area
|
0.0
|
0.0
|
0.0
|
0.0
|
18.0
|
2.0
|
18.0
|
46.0
|
Natural pastures, % of area
|
64.0
|
87.0
|
98.0
|
100.0
|
0.0
|
0.0
|
52.0
|
1.0
|
Ureac, kg ha− 1 yr− 1
|
0.0
|
0.0
|
0.0
|
0.0
|
0.0
|
0.0
|
150.0
|
0.0
|
Fertilizerd, kg ha− 1 yr− 1
|
0.0
|
90.0
|
150.0
|
0.0
|
144.0
|
200.0
|
150.0
|
150.0
|
Dolomite lime, kg ha− 1 yr− 1
|
0.0
|
0.0
|
30.0
|
0.0
|
0.0
|
100.0
|
30.0
|
0.0
|
Farm
|
|
|
|
|
|
|
|
|
Stocking rate, AU per ha
|
0.8
|
3.1
|
0.4
|
0.3
|
0.8
|
3.1
|
0.4
|
0.3
|
Diesel, L ha− 1 yr− 1
|
82.6
|
59.5
|
20.6
|
0.7
|
82.6
|
59.5
|
20.6
|
0.7
|
a FPCM: Fat Protein Corrected Milk (3.5% fat, 3.3% protein)
b AU: Animal Unit (1 AU being either 1 cow, or 3.3 female and male calves less than 1 year, or 1.7 female and male calves 1–2 year, or 1.3 heifers 2–3 year, or 1.3 steers 1–2 year, or 0.8 bulls)
c Urea: 46(N)
d Fertilzer: 31(N): 8(P): 8(K)
|
3.2 Management measurements implemented in the improvement scenario
The Cauca Department is one of the departments of Colombian with a high proportion of the area with agricultural vocation dedicated to pastures, which shows that cattle farming constitutes one of the main economic activities in the region. Despite this important participation, in this region, serious limitations threaten its sustainability, such as loss of productive capacity of soils due to degradation processes from inadequate management practices, climate change, and natural disasters. Cattle farmers in this department do not have an adequate quality diet for animals, have received little technical training, and have no adequate infrastructure to carry out proper cattle husbandry practices. This is reflected in the low milk production rates in the baseline, from 1,508 to 2,451 l cow− 1 yr− 1 (Table 3), when compared to the national average milk production figures for specialized dairy systems which range from 3,600 to 4,300 l cow− 1 yr− 1 (Carulla and Ortega 2016).
Training sessions about pasture management and reproductive practices such as rotational grazing, knowledge on grass growth stages to identify the grazing periods, fertilization practices, artificial insemination, and reproductive control on cows and bulls were performed. The success of these training sessions was evident in the improvement scenario, where farmers adopted better cattle husbandry practices than in the baseline. Thereby, in this scenario, all farmers kept productive records, performed rotational grazing and weed control by using different methods (manual, mechanical, and/or chemical), adopted higher areas of IP, and optimized the fertilizer application rates. Burning of grazing plots as a renewal practice was not reported by the farmers. Milking was mainly performed manually, and only farm number four had mechanical milking implemented. These good cattle husbandry practices were reflected in the higher milk production achieved in the improvement scenario, with increases of about 33% when compared to the baseline (Table 3).
The Cauca Department is a region in Colombia with a significant proportion of its land dedicated to pastures, making cattle farming one of its main economic activities. However, the sustainability of this sector is threatened by several factors, including soil degradation due to inadequate management practices, climate change, and natural disasters. Additionally, cattle farmers in this department often face challenges such as poor-quality diets for their animals, limited technical training, and inadequate infrastructure for proper cattle husbandry. This is reflected in the low baseline milk production rates, ranging from 1,508 to 2,451 liters per cow per year (Table 3), which are significantly lower than the national average for specialized dairy systems, which ranges from 3,600 to 4,300 liters per cow per year (Carulla and Ortega 2016). To address these issues, a series of training sessions on pasture management and reproductive practices were conducted under the frame of the “Development of the Dairy Chain to Improve the Quality of Life in Families from Cauca Department” project. These sessions covered topics such as rotational grazing, understanding grass growth stages to identify optimal grazing periods, fertilization practices, artificial insemination, and reproductive control for cows and bulls. The effectiveness of these training sessions was demonstrated in the improvement scenario, where farmers adopted better cattle husbandry practices compared to the baseline.
In the improvement scenario, all farmers maintained productive records, practiced rotational grazing, and controlled weeds using various methods (manual, mechanical, and chemical). They also expanded the area of improved pastures (IP) and optimized fertilizer application rates. Notably, the burning of grazing plots as a renewal practice was not reported by the farmers. Milking was predominantly performed manually, with only farm number four implementing mechanical milking. These improved cattle husbandry practices led to a significant increase in milk production in the improvement scenario, with an approximate 33% increase compared to the baseline (Table 3).
3.3 Animal diet composition
In both the baseline and improvement scenarios, the animal diet was predominantly based on year-round grazing, with minimal inclusion of external feed inputs such as commercial concentrate, cereal bran, molasses, and mineralized salt (Table 3). This feeding strategy is commonly reported for small dairy systems in Colombia (González-Quintero et al. 2020). In the improvement scenario, the proportion of IP was higher than in the baseline scenario. This strategy was implemented to enhance forage productivity and quality (crude protein and digestibility), thereby increasing milk yields. The use of IP in pasture-based systems has been proposed as an intervention in several LCA studies for cattle systems in Latin America (Mazzetto et al. 2015; Cardoso et al. 2016; Gaitán et al. 2016; Ribeiro-Filho et al. 2020; González-Quintero et al. 2021a, b). This approach not only has the potential to increase animal yields but also offers opportunities to reduce greenhouse gas emission (GHGE) intensities. Consequently, a comparative estimation of GHGE intensities for the baseline and improvement scenarios, using the LCA methodology, would be highly beneficial in identifying potential GHGE mitigation benefits following the implementation of IP and SPS in the studied dairy farms. To this end, an attributional LCA was performed to calculate the GHGE of the studied farms, using real farm data for both the baseline and improvement scenarios. This LCA enabled the identification of the main GHGE hotspots, potential reductions in milk and meat CFs, and increases in milk yields before and after the implementation of the intervention scenario.
3.4 Nitrogen balance
N surpluses are attributed to N losses from emissions of NH3, N2O, nitrogen oxides (NOx), runoff, leaching of nitrate (NO3−) into surface and groundwater, and changes in soil N stock (Penati et al. 2011). In the baseline scenario, the primary N input was from atmospheric deposition (15 kg N ha− 1 yr− 1), followed by purchased external feed (11.8 kg N ha− 1 yr− 1). In the improvement scenario, the primary N input shifted to fertilizers (58.6 kg N ha− 1 yr− 1), followed by atmospheric deposition (15 kg N ha− 1 yr− 1). For N outputs, the main output in both scenarios was N in milk, with values of 9.73 kg N ha− 1 yr− 1 in the baseline and 14.67 kg N ha− 1 yr− 1 in the improvement scenario. The N surplus was higher in the improvement scenario (42.5 kg N ha− 1 yr− 1) compared to the baseline scenario (18.8 kg N ha− 1 yr− 1). This increase is attributed to the higher application of fertilizers in the improvement scenario due to the higher proportion of IP. Direct and indirect N emissions were calculated to be 18.8 kg N ha− 1 yr− 1 in the baseline and 34.13 kg N ha− 1 yr− 1 in the improvement scenario. The small difference between the N surplus at the farm level and the N lost through emissions indicates that there was negligible N available for leaching. Consequently, indirect N emissions from leaching were considered negligible and were not included in the LCA of the farms, consistent with the findings of Viglizzo et al. (2006) and Modernel et al (2013). Our findings align with those reported in CF studies of low-input pasture-based dairy farms in Colombia and Italy, where N surpluses of 6.4 and 15 kg N ha− 1 yr− 1 were observed, and leaching emissions were similarly excluded from the CF calculations (Penati et al. 2011; González-Quintero et al. 2021b).
3.5 Total GHGE and milk and meat CFs of dairy production systems in Cauca
GHGE from a cradle-to-farm gate perspective were 96.4 and 89.5 tons of CO2eq for the baseline and improvement scenarios, respectively (Table 4). The total GHGE per farm ranged between 20.5 and 31.7 tons CO2eq for the baseline scenario and between 18.6 and 25.4 tons CO2eq for the improvement scenario (Table 4). GHGE from on- and off-farm activities, carbon sequestration in biomass, and the CFs of milk and meat for each farm are detailed in Table 4. In both scenarios, farms displayed a similar pattern of emissions concerning the contribution of direct and indirect sources to total GHGE (Fig. 2). In the baseline scenario, CH4 emissions from animals (enteric fermentation and excretions on pastures) accounted for an average of 79% of total farm GHGE. This was followed by CO2 emissions from the manufacturing of external feed inputs and agrochemicals (11.3%), direct and indirect N2O emissions from excreta deposited on pastures and fertilizer application (7.3%), CO2 from fossil fuel combustion (2.1%), and CO2 from the transportation of inputs to the farm (0.3%). In the improvement scenario, CH4 emissions from animals contributed 71.6% of total GHGE. CO2 emissions from feed and agrochemicals manufacturing were the second-largest source at 14.1%, followed by N2O emissions from excreta deposited on the field and fertilizer application at 11.9%, and CO2 from fuel combustion and transportation at 2.4%. The increased use of inputs in the improvement scenario resulted in a higher proportion of emissions allocated to the manufacturing of inputs and fertilizer application compared to the baseline.
Table 4
Total GHGE from different emission sources, and milk and meat CFs for studied farms: baseline and improvement scenarios.
|
On-farm activities
|
Off-farm activities
|
|
|
|
|
|
|
Farm number
|
Animals
|
Pastures (excretions and agrochemicals)
|
Burning of fossil fuels
|
Carbon capture in biomass
|
Manufacturing of inputs (feeds and agrochemicals)
|
Transport
|
Total emissions
|
|
Milk carbon footprint AR5 (reductions from baseline)
|
Meat carbon footprint AR5 (reductions from baseline)
|
Milk carbon footprint AR6 (reductions from baseline)
|
Meat carbon footprint AR6 (reductions from baseline)
|
Milk production (reductions from baseline)
|
|
Ton CO2eq
|
|
KgCO2eq kgFPCM− 1
|
KgCO2eq kgLWG− 1
|
KgCO2eq kgFPCM− 1
|
KgCO2eq kgLWG− 1
|
kgFPCM cow− 1 yr− 1
|
Baseline scenario
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Farm 1
|
15.9
|
1.6
|
1.0
|
0.00
|
2.6
|
0.1
|
21.3
|
|
3.2
|
22.1
|
3.1
|
21.6
|
1508.2
|
Farm 2
|
25.2
|
2.1
|
0.3
|
0.00
|
3.9
|
0.2
|
31.7
|
|
2.4
|
16.2
|
2.4
|
15.8
|
2450.8
|
Farm 3
|
14.8
|
1.6
|
0.5
|
0.00
|
3.5
|
0.1
|
20.5
|
|
3.0
|
21.3
|
3.0
|
20.9
|
1508.2
|
Farm 4
|
20.4
|
1.7
|
0.0
|
0.00
|
0.8
|
0.0
|
22.9
|
|
2.4
|
15.8
|
2.3
|
15.4
|
2262.3
|
Improvement scenario
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Farm 1
|
10.6
|
2.3
|
1.0
|
-0.01
|
4.4
|
0.0
|
18.3
|
|
2.7 (-16%)
|
18.4 (-16%)
|
2.6 (-16%)
|
17.9 (-16%)
|
2073.7 (+ 38%)
|
Farm 2
|
19.4
|
2.8
|
0.3
|
-0.01
|
2.7
|
0.0
|
25.4
|
|
1.4 (-40%)
|
9.4 (-42%)
|
1.4 (-40%)
|
9.0 (-42%)
|
3393.4 (+ 38%)
|
Farm 3
|
15.6
|
2.6
|
0.5
|
-0.02
|
2.5
|
0.0
|
21.2
|
|
2.2 (-28%)
|
14.6 (-31%)
|
2.2 (-28%)
|
14.2 (-31%)
|
2262.3 (+ 3%)
|
Farm 4
|
19.3
|
2.9
|
0.0
|
-0.02
|
2.4
|
0.0
|
24.6
|
|
2.0 (-17%)
|
12.8 (-19%)
|
1.9 (-17%)
|
12.4 (-19%)
|
3016.4 (+ 8%)
|
Methane emissions were the dominant GHG, contributing 79% and 71.6% of total GHGE in the baseline and improvement scenarios, respectively. Enteric fermentation was the primary source of CH4 emissions, averaging 2.8 kg CO2eq per kg FPCM in the baseline scenario and 1.5 kg CO2eq per kg FPCM in the improvement scenario. The high contribution of CH4 emissions to the milk CF is consistent with other studies on pasture-based cattle systems, where extended grazing seasons reduce reliance on farm inputs such as external feed and fertilizers (Gilardino et al. 2020; D’aurea et al. 2021; Reyes-Palomo et al. 2022; Herron et al. 2022). The emission of CH4 from enteric fermentation is driven by the ruminant digestion process, which is influenced by feed quality. Given these constraints, it is crucial to provide additional support to producers beyond the introduction of IP and SPS to fully leverage the potential of these technologies. Options include favorable agricultural credit terms and government programs aimed at emissions mitigation in dairy production. Another significant aspect of support involves realizing the environmental benefits of GHGE mitigation through these technologies by exploring participation in Payments for Ecosystem Services or Carbon Markets. Finally, it is important to consider that the farms analyzed in this study are small-scale operations with limited economies of scale. Larger farms may benefit from better economies of scale, making positive economic outcomes more attainable.
3.6 Economic analyses
The main results for each scenario can be consulted in Table 5. As observed in Table 5, the economic indicators consistently improved under each scenario, i.e., when factoring in the environmental values. It is important to note that the current state is hypothetical, with potential realization in the future if producers engage in initiatives like Payments for Ecosystem Services or Carbon Markets. Scenario IV demonstrates the highest environmental value, mitigating 163 tons of CO2eq, valued at US$ 27,716 among the four farms assessed. Farm 2 in Tambo emerges as the frontrunner in performance, yielding positive outcomes in four out of the five scenarios. Conversely, Farm 1 in Cajibío exhibits the least favorable results across all scenarios. This disparity is due to lower animal productivity in Farm 1, where losses are minimized at a stocking rate of 2.6 animals per hectare and increase again beyond 2.7 animals per hectare. Furthermore, we also evaluated the economic value attributed to microclimatic regulation within the farm area where the SPS was implemented. On each farm, a total of 620 Alnus acuminata trees were planted, creating a calculated shaded area of 6,510 m2. This microclimatic regulation holds an economic value of US$ 6,951. These figures are consistent with other studies, such as Sandoval et al. (2023), who estimated the economic-environmental values of implementing SPS with Leucaena leucocephala in Colombia. They conclude that such SPS arrangements (i) increase productivity and improve the economic indicators of the farm, such as the Net Present Value, Internal Rate of Return, and Benefit-Cost Ratio; (ii) can generate shade valued at US$ 2,000 per hectare; and (iii) reduce GHGE valued at US$ 6 per beef cattle and year. Likewise, Enciso et al. (2019) estimated improved economic indicators when Leucaena diversifolia is included in SPS in Colombia.
Table 5. Economic and environmental evaluation
Farm
|
Municipality
|
Stocking rate
|
Carbon footprint reduction (Ton CO2 eq/year)
|
NPV carbon footprint reduction (US/year)
|
Economic results
|
Economic and environmental results
|
NPV (US/year)
|
IRR
|
B/C
|
NPV (US/year)
|
IRR
|
B/C
|
Baseline scenario
|
|
|
|
|
|
|
|
|
|
1
|
Cajibío
|
0.8
|
0
|
0
|
-32,531
|
losses every years
|
0.37
|
|
|
|
2
|
Tambo
|
3.1
|
0
|
0
|
-8,885
|
losses every years
|
0.81
|
|
|
|
3
|
Sotará
|
0.4
|
0
|
0
|
-39,874
|
losses every years
|
0.33
|
|
|
|
4
|
Timbío
|
0.3
|
0
|
0
|
-32,941
|
losses every years
|
0.48
|
|
|
|
Real improved scenario
|
|
|
|
|
|
|
|
|
1
|
Cajibío
|
0.8
|
4.6
|
784
|
-33,590
|
losses every years
|
0.4
|
-32,806
|
losses every years
|
0.42
|
2
|
Tambo
|
3.1
|
24.1
|
4,139
|
-1,225
|
7%
|
0.97
|
2,914
|
43%
|
1.06
|
3
|
Sotará
|
0.4
|
8
|
1,368
|
-33,736
|
losses every years
|
0.42
|
-32,368
|
losses every years
|
0.42
|
4
|
Timbío
|
0.3
|
7.6
|
16,270
|
-53,334
|
losses every years
|
0.41
|
-37,064
|
losses every years
|
0.59
|
Improved scenario with ideal stocking rate
|
|
|
|
|
|
|
|
1
|
Cajibío
|
2.5
|
14.3
|
2,450
|
-27,958
|
losses every years
|
0.71
|
-25,509
|
losses every years
|
0.74
|
2
|
Tambo
|
5.5
|
42.7
|
6,236
|
18,731
|
619%
|
1.35
|
24,967
|
earnings every years
|
1.46
|
3
|
Sotará
|
2
|
39.8
|
6,842
|
-21,533
|
losses every years
|
0.79
|
-14,691
|
losses every years
|
0.86
|
4
|
Timbío
|
0.8
|
20.2
|
3,466
|
-31,970
|
losses every years
|
0.72
|
-28,504
|
losses every years
|
0.75
|
Improved scenario with minimum stoking rate
|
|
|
|
|
|
|
|
1
|
Cajibío
|
2.6
|
14.8
|
2,548
|
-28,686
|
losses every years
|
0.71
|
-26,139
|
losses every years
|
0.74
|
2
|
Tambo
|
1.3
|
10.1
|
1,474
|
262
|
21%
|
1.01
|
1,736
|
39%
|
1.05
|
3
|
Sotará
|
4.9
|
97.6
|
16,762
|
586
|
19%
|
1
|
17,348
|
47%
|
1.09
|
4
|
Timbío
|
1.6
|
40.3
|
6,933
|
2,212
|
20%
|
1.01
|
9,144
|
29%
|
1.06
|
All farm area with IP
|
|
|
|
|
|
|
|
1
|
Cajibío
|
3.4
|
19.4
|
3,332
|
-23,982
|
losses every years
|
0.8
|
-20,650
|
losses every years
|
0.83
|
2
|
Tambo
|
5.5
|
42.7
|
7,344
|
45,309
|
earnings every years
|
1.72
|
52,653
|
earnings every years
|
1.84
|
3
|
Sotará
|
2.7
|
53.8
|
9,236
|
27,733
|
76%
|
1.22
|
36,969
|
105%
|
1.29
|
4
|
Timbío
|
1.2
|
30.3
|
5,200
|
-15,386
|
-2%
|
0.89
|
-10,186
|
5%
|
0.92
|
The baseline scenario (Scenario I) resulted in economic losses across all farms. Introducing IP and SPS (Scenario II) aids in curbing these losses but falls short of generating profits. Notably, positive economic results emerge as stocking rates and the area under IP are expanded (Scenarios III, IV, and V). However, small producers face challenges in capitalizing on increased stocking rates due to difficulties in acquiring more animals, thus missing out on potential additional income.
The process of adopting IP and SPS encounters multiple obstacles that, among others, encompass (i) financial impediments: limited credit access, prolonged repayment periods; (ii) knowledge gaps: restricted information availability, lack of technical assistance and extension services, different knowledge sets required for applying IF and SPS practices; (iii) socio-cultural factors: gender roles, traditional ways of doing things (i.e., extensive grazing on natural pastures); (iv) labor shortages due to competition with other (illegal) sectors; (v) unclear land tenure; (vi) market dynamics; (vii) legal restrictions; and (viii) farmers' general risk aversion. Tackling these diverse barriers demands a holistic and nuanced approach that combines focused interventions, robust policies, and comprehensive support mechanisms (Calle et al. 2013; Zapata et al. 2015; Zepeda Cancino et al. 2016; Raes et al. 2017; Puppo et al. 2018; Charry et al. 2019; Lee et al. 2020; Jara-Rojas et al. 2020; Tschopp et al. 2020, 2022; Enciso et al. 2022).
Given these constraints, it becomes crucial to extend support to producers beyond just introducing IP and SPS, to enable them to fully harness the potential of these technologies. Options include favorable agricultural credit terms and government programs aimed at emissions mitigation in dairy production. An equally significant facet of support is materializing the environmental value derived from GHGE mitigation through these technologies. This entails exploring avenues through which producers can participate in Payments for Ecosystem Services or Carbon Markets. Lastly, it is worth considering that the farms analyzed in this study are small-scale operations with limited economies of scale. Larger farms may exhibit more favorable economies of scale, making positive economic outcomes more attainable.