https://orcid.org/0000-0002-7436-4013
Figures
Abstract
This study aimed to evaluate the nutritional quality and in vitro total gas production of Piptocoma discolor (Kunth) Pruski forage collected from two contrasting landscapes (alluvial plain and hill) in the Colombian Amazon piedmont. To achieve this, leaf and non-lignified stem samples were collected from scattered trees in each landscape and analyzed for their nutritional composition, degradability, and gas production. Crude protein (CP) levels differed significantly between landscapes (p < 0.05), with higher concentrations in the alluvial plain compared to the hill (14.6% vs. 9.5%, respectively). A similar trend was observed for ash, phosphorus, and nitrogen content (p < 0.05). However, neutral detergent fiber (NDF) content was higher in the hill landscape. The concentrations of magnesium, potassium, calcium, sodium, zinc, manganese, copper, and iron did not differ significantly between landscapes (p < 0.05). Regardless of the landscape, P. discolor exhibited moderately acceptable degradability (72%). Cumulative gas production (CGP) at 96 h was significantly higher (p < 0.05) for forage collected in the hill landscape compared to the alluvial plain (66 vs. 53 ml CGP/g dry matter incubated, respectively). These results confirm the potential of P. discolor to improve ruminant nutrition and support sustainable silvopastoral practices in the region.
Citation: Álvarez F, Ríos-Parra PA, Casanoves F, Sterling A, Molina-Botero IC (2026) Nutritional quality and in vitro gas production of Piptocoma discolor (Kunth) Pruski forage across contrasting landscapes in the Colombian Amazon Piedmont. PLoS One 21(3): e0345567. https://doi.org/10.1371/journal.pone.0345567
Editor: Vitor Hugo Rodrigues Paiva, MARE – Marine and Environmental Sciences Centre, PORTUGAL
Received: March 7, 2025; Accepted: March 6, 2026; Published: March 23, 2026
Copyright: © 2026 Álvarez et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its supplementary information files.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that there are no conflicts of interest.
Introduction
Cattle ranching is one of the main causes of tropical forest fragmentation worldwide, altering the composition, configuration and ecological function of the biome. This transformation alters [1,2] biological cycles, biodiversity and the overall stability of ecosystems, contributing to increased climate variability [3,4]. According to Ordoñez et al. [5] the department of Caquetá, in the Colombian Amazon, is particularly affected by this problem, since several geoforms in the region have suffered important physical [6], chemical [7] and biological [8] alterations. These changes are largely due to the conversion of native forests to grasslands, which has resulted in a 19.6% decrease in soil fertility [9]. The hilly landscape, comprising 67.9% of the region, has been disproportionately affected (70% conversion) compared to the floodplain landscape (9.7%) [10]. As of 2020, the livestock population in Caquetá was approximately 2 million head of cattle [11], primarily managed within dual-purpose dairy and fattening systems. Traditionally, these production systems rely on extensive grazing over large areas of natural and improved pastures, where relatively few animals are dispersed across vast landscapes [12]. The predominant forage base consists of grasses with low nutritional value, i.e., 6.6% CP, 75% NDF and 65% digestibility, as reported in regional studies, limiting animal productivity in the Amazonian piedmont [13].
In this context, the integration of tree and shrub species into livestock diets presents a viable strategy for improving forage quality. These species offer higher crude protein content and greater biomass yield than traditional pasture grasses [14,15]. Additionally, multipurpose trees provide fruits that can substitute for grain-based feed in livestock diets [16,17], often surpassing commercial concentrates in energy content [18]. Promising species such as Tithonia diversifolia (Hemsl.) A. Gray, Gmelina arborea Roxb. ex Sm, and Acacia mangium Willd exhibit high biomass production, favorable nutritional profiles, and adaptability to diverse soil and climatic conditions, making them well-suited for ruminant supplementation [19]. In addition, these species contain secondary compounds such as tannins that help mitigate ammonium and methane emissions in the rumen [20] thus contributing to improved environmental sustainability in livestock production [21]. Their inclusion in forage systems promotes an efficient nutrient cycle, while adapting to the unique characteristics of Amazonian soils [22]
A particularly promising native species for the Amazon region is P. discolor (Kunth) Pruski [22], which thrives in secondary forests and is readily consumed by cattle grazing in pastures [23]. It should be noted that this species grows rapidly, even in degraded soils [24], tolerates high levels of precipitation typical of the Amazon region where tropical rainforest prevails [25], and is easily adaptable with a high production capacity in areas of natural regeneration, with yields of 180.56 g dry mater/plant [23], making it a valuable component of silvopastoral and agroforestry systems in the department of Guaviare, Colombia [25]. The incorporation of native trees and shrubs into livestock systems offers multiple benefits, such as increased milk production and animal weight [18], improved biodiversity conservation in agricultural landscapes, and improved resilience to climate change.
In addition, soil organisms such as earthworms and fungi improve nutrient availability and combat land degradation, supporting sustainability by linking soil health to forage quality [26]; these systems mitigate environmental impact and contribute to sustainable land management.
The wide diversity of forage species and their high nutritional value for cattle feeding in the livestock systems of the Colombian Amazon, along with their potential to reduce greenhouse gas emissions, represents an important opportunity for improving animal nutrition and mitigating the effects of climate change. These advantages contribute to enhancing both sustainability and productivity of livestock systems. However, environmental factors such as landscape type may have a direct impact on the nutritional composition of forage and the gas emissions resulting from the consumption of forage species.
In this context, hill-dominated landscapes have soils derived from highly weathered claystones and mudstones, characterized by strong acidity, high aluminum saturation, and low natural fertility due to minimal concentrations of essential nutrients such as calcium, magnesium, potassium, sodium, and phosphorus. In contrast, the soils of the alluvial plain landscapes originate from alluvial deposits, exhibit variable granulometry, and are generally shallow, poorly drained, and classified as having low to moderate natural fertility [10].
To address these gaps, we hypothesized that the nutritional characteristics of P. discolor forage are higher in alluvial plain landscapes, which are characterized by soils with greater natural fertility, whereas its consumption by cattle results in lower total gas production compared to hill landscapes. To test this hypothesis, this study aimed to characterize the nutritional quality and in vitro total gas production of P. discolor in two contrasting landscapes (alluvial plain and hill) in the Colombian Amazon piedmont.
Materials and methods
Study area
This study was conducted in the Municipality of Florencia, covering an area of 2,262 km2. The region is situated in the piedmont between the Eastern Cordillera and the Amazon rainforest, within the Department of Caquetá [10]. The area is characterized by an average annual rainfall of 3,759 mm, an average temperature of 25.8 °C, and a relative humidity of 81%, classified as Köppen Af climate (tropical rainforest) in Colombia [24].
Two contrasting landscapes dominated by cattle grazing areas and presence of scattered P. discolor trees were considered [27] (i) a hill landscape with an undulating topography, featuring elevations below 300 m, broad, rounded, and elongated hilltops, and slopes ranging from 8% to 16%. The soils in this landscape are clayey, of low fertility, with variable drainage, and prone to compaction and erosion; (ii) an alluvial plain landscape, characterized by an elongated and flat topography, with slopes ranging from 0% to slightly inclined, including high and medium terraces as well as flood-prone areas. The soils in this landscape originate from alluvial-colluvial sediment deposits, exhibit moderate fertility, and have poor drainage.
The laboratory phase was carried out at the Amazonian Research Center (CIMAZ – Macagual) “Cesar Augusto Estrada González” (1°37’ N, 75°36’ W), located at an elevation of 300 m a.s.l., approximately 22 km south of Florencia, in the Department of Caquetá [28].
Future research could complement field-based assessments with remote sensing approaches, which offer powerful methodological tools to analyze vegetation and environmental dynamics at the landscape scale [29,30]. Integrating such approaches would provide valuable insights to expand forage studies and strengthen sustainable land-use strategies in the region.
Sampling design and forage collection
The sample size was calculated with a 5% sampling error and a 95% confidence level using Cochran’s formula [31]. Since a stratified sampling design with optimal allocation was applied, the allocation considered the variance of forage quality attributes estimated from pilot measurements (variance of 2.1 in the hill and 1.3 in the alluvial plain). This led to a greater sampling effort in the hill stratum (36 trees) compared to the alluvial (23 trees), for a total of 59 trees. Within each landscape unit, trees were selected based on the following criteria: (i) mature trees (8–10 years old) with a height between 16 and 20 m; (ii) trees with well-defined morphological characteristics, taking into account architecture, shape, height, and crown, without alterations due to burning or cultural practices; (iii) diameter at breast height (DBH) ≥ 10 cm; and (iv) trees free of pests and diseases.
After selection of P. discolor trees, leaf and non-lignified stem samples (approximately 2 kg per tree) were collected, placed in labeled paper bags, and transported to the Forage Evaluation and Animal Nutrition Laboratory at the Centro de Investigaciones CIMAZ Macagual “Cesar Augusto Estrada González.”
Nutritional composition
Laboratory samples were dried in a DHD-9030® oven (Zenith Lab, Jiangsu, China) at 105 °C for 6 h or until a constant weight was reached to determine the dry matter (DM) content, calculated as the percentage of remaining weight (AOAC, 2005; Method 950.46). Subsequently, the samples were ground using a Willey® mill (60 Hz) equipped with a 1 mm sieve. From each sample, 200 g were selected for the determination of crude protein (CP) via the Kjeldahl method (984.13) [32], acid detergent fiber (ADF), neutral detergent fiber (NDF) [33], ash content [32], and organic carbon was measured on a dry basis (Cdb) and on a wet basis (Cwb) using the Walkley and Black method [34]. The laboratory samples were dried in a DHD-9030® oven (Zenith Lab, Jiangsu, China) at 105°C for 6 hours or until a constant weight was achieved to determine the dry matter (DM) content, following the Association of Official Analytical Chemists (AOAC, 2005; Method 950.46). Subsequently, the samples were ground using a Willey® mill (60 Hz) equipped with a 1 mm sieve. From each sample, 200 g were selected for the determination of crude protein (CP) via the Kjeldahl method (984.13) [35]. Mineral quantification, including Ca, Mg, K, Na, Zn, Mn, Cu, and Fe, was performed using atomic absorption spectrophotometry with an AA7000® spectrophotometer (Shimadzu, Kyoto, Japan), while P was analyzed via UV-Vis spectrophotometry (UV 1800®, Shimadzu, Kyoto, Japan) within a wavelength range of 200–1000 nm.
In vitro dry matter degradability
A total of 180 crucibles were selected, thoroughly washed, and placed in an oven for 12 hours. Subsequently, they were transferred to a desiccator for 5 min to equilibrate. After 96 h of incubation, the rumen kinetic residue was filtered, and the samples were dried in an oven for 24 h. Upon completion of the drying process, the samples were placed in a desiccator to reach room temperature before being weighed. The final weight of the crucible and sample was recorded to determine the in vitro dry matter digestibility [36,37].
In vitro gas production
Gas production was evaluated following the protocol established by Theodorou et al. [38]. Rumen fluid was obtained from animals processed by Compañía de Ferias y Mataderos de Caquetá (COFEMA S.A.), the authorized municipal slaughterhouse in Florencia, Colombia, which operates under national animal health and welfare standards (resolution 240 of 2013).
This study did not perform any type of animal handling, experimentation or slaughter, and all samples were taken from animals that had already been slaughtered as part of routine meat processing.
The ruminal fluid was filtered through gauze and transported to the laboratory in insulated thermos flasks to maintain temperature stability. For each incubation, 0.625 g of sample was inoculated with 53.1 ml of digestion medium and 2 ml of a reducing agent in 120 ml amber glass bottles, with three replicates per treatment. Forage samples (non-lignified leaves and stems) were collected from 10 trees per landscape type, yielding a total of 180 incubation bottles (10 trees × 2 landscapes × 3 replicates × 3 ruminal fluid sources), plus 15 blank bottles (without forage). The flasks were preheated to 39 °C in a B120 digital thermostatic water bath (OVAN®, Barcelona, Spain) before adding 6.25 ml of ruminal fluid. This incubation temperature (39 °C) was maintained throughout the experiment and readings were recorded every 3 h [38]. The total gas volume was calculated from the recorded pressure values using the following equation:
where Y represents the gas volume produced (ml) per unit of pressure x
The gas production data were then fitted to the nonlinear model Gompertz [39]:
where Y denotes the cumulative gas production at time x, a (> 0) represents the maximum gas production, b (> 0) is the difference between initial and final gas volume at time x, and c (> 0) describes the specific gas accumulation rate. The biological interpretation of these parameters includes: (i) TIP: Time to the inflection point (h); (ii) GIP: Gas volume at the inflection point (ml); and (iii) MRGP: Maximum rate of gas production (ml/h).
Data analysis
To compare the means between landscape types, a t-test was conducted for each nutritional quality variable. For variables exhibiting heterogeneous variances between landscapes, Satterthwaite’s correction was applied. To examine the relationships between nutritional variables, a Pearson correlation analysis was conducted [40]. The correlation matrix displaying significant relationships was visualized using the circlize package [41] in R v. 4.3.3 language [42]. Principal Component Analysis (PCA) was performed to assess associations between nutritional variables and landscapes, with results visualized using a biplot [40]. Statistical analyses related to nutritional quality were conducted using InfoStat software, v. 2019 [43].
Dry matter degradation, gas kinetics, and Gompertz model parameters were analyzed using a randomized complete block design in SAS® software, version 9.4 [44]. Each sample was incubated in triplicate across three different inoculants, with the inoculant serving as the blocking variable. Mean comparisons were performed using Tukey’s test [45]. Data normality was assessed using the Shapiro-Wilk test [46] on the raw residuals within the PROC GLM procedure [47].
Results
Forage nutritional quality
The nutritional and mineral composition of P. discolor forage in two Amazonian foothill landscapes shows that the floodplain forage showed significantly higher DM content than that of the hilly landscape (36.1% vs. 30.9%, P = 0.0458).
A similar trend was observed for CP, with P. discolor forage from the alluvial plain containing 1.54 times more CP than that from the hill landscape (14.6% vs. 9.5%). In contrast, NDF content was higher in the hill landscape compared to the alluvial plain (52.9% vs. 49.5%, respectively). The ADF values did not differ significantly between landscapes (p > 0.05).
Regarding mineral composition, ash, nitrogen, and phosphorus contents were significantly higher in the alluvial plain (9.66%, 2.34%, and 1.29%, respectively) than in the hill landscape (7.15%, 1.51%, and 0.93%, respectively), with differences ranging from 0.36% to 2.5%. However, the concentrations of Mg, K, Ca, Na, Zn, Mn, Cu, and Fe did not differ significantly between landscapes (p > 0.05) (Table 1).
Pearson’s correlation analysis revealed the strongest significant positive correlations between CP and N, Cdb and Cwb, as well as Mg with C, Mn, and Ca (all, r > 0.6; p < 0.05) (Fig 1). In contrast, Pearson’s analysis showed strong positive correlations between CP and N (r > 0.6, p < 0.05) and Mg with Ca, Mn (r > 0.6, p < 0.05), and negative correlations between NDF and DM, Cwb (r < −0.6, p < 0.05) (Fig 1).
Coloration and width of the ribbons indicate the direction and magnitude of the correlation, with darker blue for stronger positive (r > 0.4) and darker red for stronger negative (r < −0.3).
The PCA conducted on the nutritional and mineral quality variables, considering landscape type, accounted for 43% of the total variability (Fig. 2). PC1 explained 27.5% of the variation, primarily differentiating nutritional variables such as CP, nitrogen, DM, carbon on a wet basis (Cwb), and carbon content on a dry basis (Cdb), which were positively associated with PC1 and showed higher values in the alluvial plain landscape. In contrast, NDF, OC, and M were more strongly associated with the hill. PC2 accounted for 17.9% of the variation, distinguishing variables such as OM, ADF, and mineral elements (Mn, Fe, Ca, K, Mg, Cu, and Na), which showed similar values across both landscapes.
Dry Matter (DM), Organic Matter (OM), carbon on a wet basis (Cwb), carbon on a dry basis (Cdb), organic carbon (CO), moisture (M), Crude Protein (CP), Neutral Detergent Fiber (NDF, Acid Detergent Fiber (ADF), Nitrogen (N), Phosphorus (P), Magnesium (Mg), Potassium (K), Calcium (Ca), Sodium (Na), Zinc (Zn), Manganese (Mn), Copper (Cu) and Iron (Fe).
In vitro gas production
Cumulative gas production (CGP) per incubated dry matter (IDM), degradation, and their relationship are presented in Fig 2 and Table 2. At 12 h, CGP was higher in the alluvial plain landscape, reaching 24.4 ml/g IDM, while the hilly landscape yielded 19.9 ml/g IDM, 4.5 ml/g IDM less than the floodplain (p < 0.05). At 24 h, no significant differences were observed between landscapes (p > 0.05). However, at 48, 72, and 96 h post-incubation, forages from the hill landscape exhibited higher CGP values of 42, 55.9, and 65.8 ml/g IDM, respectively, compared to 35.4, 43.9, and 52.5 ml/g IDM in the alluvial plain landscape. The disparity between landscapes increased over time. On average, dry matter degradation (DMD) at 96 h was 72.1% (p > 0.05). CGP per degraded organic matter (DOM) was significantly higher in the hill landscape compared to the alluvial plain (102.7 vs. 78.99 ml/g).
When fitting the gas production data to the nonlinear Gompertz model, significant differences were observed (p < 0.05) in the cumulative gas production rate at a specific time and the maximum production rate, both of which were higher for forages from the hill landscape (Table 3). However, no significant differences (p > 0.05) were found between landscapes for other parameters, including maximum gas production, the difference between initial and final gas at time x, time at the inflection point, and gas production at the inflection point.
Discussion
CP content (9.48% hill vs. 14.62% floodplain) was lower than 21% [21] and 20.4% [23], probably due to regrowth age and environmental factors. These results highlight the potential of this species in silvopastoral systems, due to its good growth, volume, and protein content, which exceeds that of tropical grasses, which tend to have low PC levels (≤ 7% PC) [13]. In addition, it is considered to have potential for animal feed due to its high biomass supply and nutritional composition [48] and as a promising alternative for improving the energy-protein ratio in ruminant diets.
The differences in moisture content and, consequently, DM between the hill and alluvial plain landscapes reflect the influence of landscape type on forage plants. According to Suarez et al. [49], these variations may be directly related to the species’ adaptation to its environment or to potential water stress caused by soil surface compaction in the alluvial plain, leading to higher DM values in this landscape.
The NDF content was higher in samples collected from the hill than in those from the alluvial plain (52.9% vs. 49.5%, respectively). The NDF levels in this study were comparable to those reported by Riascos et al. [21] for this species (46.60% NDF) but exceeded the values found by Maciej et al. [50] who reported NDF contents ranging from 23.4% to 36% for Acacia decurrens Willd. and Sambucus nigra L., two high-altitude forage woody species. Conversely, for lowland forage woody species, our results fall within the range (31–67% NDF) reported by Pinto-Ruiz et al. [51] for ten species (31 and 67% NDF). Regarding ADF content, the values observed in this study were 9% higher than those reported by Guayara [23] (36.1% ADF); however, ADF values as high as 27.25% have been documented in other regions, such as the Putumayo Department in Colombia. It is important to emphasize that the nutritional composition of compounds such as CP and structural carbohydrates in forage woody plants varies according to agroclimatic conditions and the number of regrowth days [52,53].
The colonization of fiber by ruminal microorganisms, their adherence to the cell wall, and the initiation of catabolic activity depend on the rumen environment (e.g., moisture, pH, temperature) and the NDF content of the diet. The degradability of forages in the rumen is closely linked to the proportion and lignification of plant cell walls [19,54].
Ash content, an indicator of total mineral content, was higher in the alluvial plain landscape than in the hill. Both values were within the range reported for other shrubs [55]. Nitrogen levels in P. discolor followed the same trend as ash content. Across both landscapes, mineral contents (P, N, K, Ca, Mn, Zn, Fe, Cu) were higher than those found in grasses and shrub leaves [56], suggesting that P. discolor could serve as a valuable mineral supplement for ruminants.
Phosphorus content in P. discolor was higher than that of several tropical grasses, including Pennisetum purpureum Schumach (0.26–0.48%), Cenchrus ciliaris cv. Biloela L. (0.27–0.48%), Urochloa decumbens (Stapf) R.D. Webster (0. 19–0.22%), Urochloa humidicola (Rendle) Morrone & Zuloaga (0.11–0.17%), Urochloa arrecta (Hack. ex T. Durand & Schinz) Morrone & Zuloaga (0.14–0.22%), Panicum maximum Jacq. (0.25%), and Dichantium aristatum (Poir.) C.E.Hubb. (0.11%) [57,58]. Similarly, Ca levels were higher than those reported for tropical grasses, such as P. maximum (0.41%) and D. aristatum (0.29%) [59]. These findings are significant, as the consumption of P. discolor forage could help meet the mineral requirements of ruminants fed on grass-based diets. Minerals such as phosphorus and calcium are essential for ruminant health, as they play crucial roles in cellular reactions, bone and muscle formation, milk production, osmoregulation, amino acid and protein metabolism, and enzyme activity [60].
The differences in DM, ash, NDF, CP, N, and P reflect soil fertility, enhanced by biota due to the presence of earthworms and fungi, which contribute to soil health by contributing to the nutrient cycle and plant growth [26], through the degradation of organic matter, the formation of stable structures that increase soil porosity, and rapid absorption of these [61]. This continuous supply of nutrients through sediment deposition and the presence of soil fauna contrasts with the upland area, where nutrient loss is more pronounced due to leaching and suboptimal soil management, reducing the availability of N and P [62]. Laycock and Price [63] highlighted that environmental factors such as water availability, solar radiation, temperature, and soil properties significantly influence the chemical composition of forage by reducing PC content. However, Nelson and Moser [64] pointed out that each plant species has unique morphological and physiological traits that determine its adaptation, growth, and forage quality.
In this study, gas production was higher in the hill landscape than in the alluvial plain landscape, possibly due to soil characteristics influencing the nutritional composition of P. discolor forage. Variability in gas production and DM degradation highlights the potential of certain forages to support microbial biomass production by providing favorable fiber and protein compositions [65]. These factors enhance fermentation rates and overall forage quality [66]. Regarding CGP, P. discolor showed a substantial increase between 48 and 96 hours in the hill landscape. This pattern contrasts with the findings of Quintanilla et al. [67], who observed a decline in CGP over time for Moringa oleifera Lam. Studies suggest that forages with high cell wall content generally exhibit low digestibility and limited energy availability [68,69]. However, diets with high CP content can enhance microbial activity and improve feed intake in ruminants [70]. During the first 24 h of incubation, gas accumulation per unit of IDM did not differ between treatments. These results align with those reported by Molina-Botero et al. [59], who indicated that fermentation time influences degradation rates, thereby affecting gas production kinetics.
The DMD in P. discolor was higher in the alluvial plain (73.6%) than in the hill (70.7%). These values were lower than those reported by Riascos et al. [19], who found an in situ degradability of 81.5% for this species. The observed differences may stem from variations in analytical techniques. Nonetheless, P. discolor remains a highly degradable species, benefiting from microbial activity influenced by its fiber and CP content [71].
Conclusions
Nutritional analysis of P. discolor from the floodplain suggests favorable properties for ruminant supplementation, with characteristics superior to those observed in the mountainous landscape. This is mainly due to its higher content of essential nutrients, such as crude protein, nitrogen, and phosphorus, along with a lower content of neutral detergent fiber, which are directly related to the production of gases in the rumen produced by the fermentation kinetics of the food consumed, the resistance time of the food in the rumen, and its degradation rate. In other words, the higher the volume of gas production, the higher the digestibility in the first few hours, demonstrating, regardless of the type of landscape, its moderate and acceptable degradation capacity.
Based on these findings, further research is recommended to determine the optimal inclusion levels of P. discolor in cattle diets. Such studies could assess its potential to reduce greenhouse gas emissions, thereby enhancing its overall benefits and facilitating its integration into silvopastoral systems in the Colombian Amazonian piedmont region.
Additionally, future research could incorporate landscape-scale approaches, such as remote sensing, to evaluate the adaptability of P. discolor within silvopastoral systems and expand the applicability of these findings across broader environmental contexts.
Acknowledgments
The authors would like to thank the University of Amazonia and COFEMA S.A., especially Mr. Milton Chávez López, for their valuable support.
References
- 1. Lavelle P, Dolédec S, de Sartre XA, Decaëns T, Gond V, Grimaldi M, et al. Unsustainable landscapes of deforested Amazonia: An analysis of the relationships among landscapes and the social, economic and environmental profiles of farms at different ages following deforestation. Global Environmental Change. 2016;40:137–55.
- 2. FAO, PNUMA. The State of the World’s Forests 2020. FAO and UNEP. 2020.
- 3. Murgueitio E, Chará J, Barahona R, Rivera JE. Avances en ganadería sostenible con sistemas silvopastoriles en América Latina. Cuban Journal of Agricultural Science. 2019;53:65–71.
- 4. Rodríguez L, Suárez JC, Rodriguez W, Artunduaga KJ, Lavelle P. Agroforestry systems impact soil macroaggregation and enhance carbon storage in Colombian deforested Amazonia. Geoderma. 2021.
- 5.
Ordoñez L, Perilla L, Pérez W. Evaluación del nivel de degradación de suelo y. Momentos de Ciencia. 2014.
- 6. Hermelin M. Geomorphological landscapes and landforms of Colombia. World geomorphological landscapes. 2016. p. 1–21.
- 7.
Venegas C, Vanegas G. Dinámica de suelos amazónicos procesos de degradación y alternativas para su recuperación. 2010.
- 8.
Zerbino M. Evaluación de la macrofauna del suelo en rotaciones cultivos-pasturas con laboreo convencional.
- 9.
Peña-Venegas CP, Mendoza ER, Rodríguez CH, Betancourt B, Garzón MT, Cardona GI. Depositos de carbono en suelos bajo relictos de bosques y praderas de fincas del departamento del Caquetá [accessed 2025 September 23]. Available from: https://www.sinchi.org.co/files/publicaciones/revista/pdf/8/3%20depsitos%20de%20carbono%20en%20el%20suelo%20bajo%20los%20relictos%20de%20bosque%20y%20praderas%20de%20fincas%20del%20departamento%20de%20caquet%20colombia.pdf
- 10.
IGAC. Estudio general de suelos y zonificación de tierras: departamento de Caquetá, escala 1/100.000. Instituto Geográfico Agustín Codazzi. 2014. Available from: https://scholar.google.com.mx/scholar?hl=es&as_sdt=0%2C5&q=IGAC%20%282014%29.%20Estudio%20general%20de%20suelos%20y%20zonificaci%C3%B3n%20de%20tierras%3A%20departamento%20de%20Caquet%C3%A1%2C%20escala%201%2F100.000.%20Subdirecci%C3%B3n%20de%20agrolog%C3%ADa.%20Instituto%20Geogr%C3%A1fico%20Agust%C3%ADn%20Codazzi.%20410.&btnG=
- 11.
Torrijos R. Cifras de contexto ganadero caquetá. 2021 [accessed 2023 November 7]. Available from: https://issuu.com/rafaeltorrijos/docs/contexto_2021
- 12. Ramirez-Contreras NE, Fontanilla-Díaz CA, Pardo LE, Delgado T, Munar-Florez D, Wicke B, et al. Integral analysis of environmental and economic performance of combined agricultural intensification & bioenergy production in the Orinoquia region. J Environ Manage. 2022;303:114137. pmid:34847366
- 13. Enciso Valencia KJ, Rincón Castillo Á, Ruden DA, Burkart S. Risk reduction and productivity increase through integrating arachis pintoi in cattle production systems in the colombian orinoquía. Front Sustain Food Syst. 2021;5.
- 14.
Vélez OM, Noguera RR, Posada SL, Gaona RC, Guerrero HS. Identification of in vitro antimethanogenic potential of native plants from the floodable savanna adapted to dry conditions in the Colombian Orinoco [accessed 2024 June 18]. LRRD. Available from: https://scholar.google.com.mx/scholar?hl=es&as_sdt=0%2C5&q=V%C3%A9lez%20O%20M%2C%20Noguera%20R%20R%2C%20Posada%20S%20L%2C%20Gaona%20R%20C%2C%20Guerrero%20H%20S.%20%282015%29.%20Identification%20of%20in%20vitro%20antimethanogenic%20potential%20of%20native%20plants%20from%20the%20floodable%20savanna%20adapted%20to%20dry%20conditions%20in%20the%20Colombian%20Orinoco.%20LRRD%2C%2027%285%29&btnG=
- 15. Ramírez-Restrepo C, Vera R. Bodyweight performance, estimated carcass traits and methane emissions of beef-cattle categories grazing Andropogon gayanus, Melinis minutiflora and Stylosanthes capitata mixed swards and Brachiaria humidicola pasture. Anim Prod Sci. 2019;59:729–40.
- 16. Zamora S, García J, Bonilla G, Aguilar H, Harvey C, Ibrahim M. ¿Cómo utilizar los frutos de guanacaste (Enterolobium cyclocarpum), guácimo (Guazuma ulmifolia), genízaro (Pithecellobium saman) y jícaro (Crescentia alata) en alimentación animal?. Agroforestería en las Américas. 2001;8:45–9.
- 17. Molina-Botero I, Arroyave-Jaramillo J, Valencia-Salazar S, Barahona-Rosales R, Aguilar-Pérez C, Burgos A. Efectos de los taninos y saponinas contenidos en el follaje de Gliricidia sepium y las vainas de Enterolobium cyclocarpum sobre la fermentación, las emisiones de metano y el rumen. Ciencia y tecnología de la alimentación animal. 2019;251:1–11.
- 18.
Benavides JE. Agroforestería para la producción animal en Latinoamérica árboles y arbustos forrajeros: una alternativa agroforestal para la ganadería. 1999.
- 19. Riascos Vallejos A, Reyes Gonzales J, Valenciaga Gutiérrez D, Apraez Guerrero J. Degradabilidad ruminal de suplementos a base de tres forrajeras nativas y adaptadas al pie de monte amazónico colombiano. Cuban Journal of Agricultural Science. 2020;54:193–207.
- 20.
Gómez FGTGCLSAC, SJ. Dominancia, composición química-nutritiva de especies forrajearas y fitomasa potencial en una selva secundaria. 2017.
- 21. Riascos Vallejos AR, Reyes González JJ, Mendoza LAA. Nutritional characterization of trees from the Amazonian piedmont, Putumayo department, Colombia. Cuban Journal of Agricultural Science. 2020;54.
- 22. Vivas Arturo W, Vera Arteaga D, Alpizar Muni J. Determinación in vitro de la calidad nutricional de tres leguminosas forrajeras. Revista de las Agrociencias. 2017;(17):43.
- 23.
Guayara A. Plantas forrajeras nativas identificadas. 2010 [accessed 2023 November 7]. Available from: https://scholar.google.es/scholar?hl=es&as_sdt=0%2C5&q=Guayara%20a.%20plantas%20forrajeras%20nativas%20identificadas%20por%20campesinos.%20-google%20acad%C3%A9mico%20%5Binternet%5D.%20amazon%20conservation%20team%20corpoamazonia
- 24.
Javier N, Ardila M, Gonzalo U, García M, Néstor S, Martínez J. Ministerio De Ambiente, Vivienda Y Desarrollo. Territorial Instituto de Hidrología, Meteorología y Estudios Ambientales-Ideam. 2010. Available from: https://www.academia.edu/download/35137766/Publicacion_Leyenda_nacional_de_coberturas_de_la_tierra_Ene_20.pdf
- 25. Castillo AR, Rienzo JAD, Cavallini D. Effect of a mix of condense and hydrolysable tannins feed additive on lactating dairy cows’ services per conception and days open. Vet Anim Sci. 2025;27:100434. pmid:40103688
- 26. Narváez-Herrera JP, Angulo-Arizala J, Barragán-Hernández W, Mahecha-Ledesma L. Arbustivas y arbóreas del piedemonte amazónico colombiano: potencial nutricional y ambiental en sistemas ganaderos. Agron Mesoam. 2023;:52442.
- 27.
Villota H. Geomorfología aplicada a levantamientos edafológicos y zonificación física de tierras. Instituto Geográfico Agustín Codazzi- IGAC. 2005. Available from: https://scholar.google.es/scholar?hl=es&as_sdt=0%2C5&q=+Geomorfolog%C3%ADa+aplicada+a+levantamientos+edafol%C3%B3gicos+y+zonificaci%C3%B3n+f%C3%ADsica+de+tierras.+Instituto+Geogr%C3%A1fico+Agust%C3%ADn+Codazzi+%E2%80%93+IGAC+%28Departamento+Nacional+de+Estad%C3%ADstica%29.&btnG=
- 28.
Estrada-Gonzales C, Rosas-Patiño G. Proyecto: centro de investigaciones amazónicas Macagual. 2007. Available from: https://scholar.google.es/scholar?hl=es&as_sdt=0%2C5&q=+Proyecto%3A+centro+de+investigaciones+amaz%C3%B3nicas+Macagual+%28CIMAZ%29.&btnG=
- 29. Sharma P, Bhardwaj DR, Singh MK, Nigam R, Pala NA, Kumar A, et al. Geospatial technology in agroforestry: status, prospects, and constraints. Environ Sci Pollut Res Int. 2023;30(55):116459–87. pmid:35449327
- 30. Bhattacharjee R, Gaur S, Ohri A, Das N, Chaturvedi S, Kumar R. Reconnaissance of spatio-temporal variability in ecological stability with local adaptability based on multi-temporal LANDSAT archive in ganga river basin. Physics and Chemistry of the Earth, Parts A/B/C. 2024;135:103667.
- 31.
Cochran W. Técnicas de muestreo. 1ra ed. México, DF: Compañía Editorial Continental; 1986.
- 32. Books in brief. J AOAC Int. 1997;80(6):127A-128A.
- 33. Van Soest PJ, Robertson JB, Lewis BA. Methods for dietary fiber, neutral detergent fiber, and nonstarch polysaccharides in relation to animal nutrition. J Dairy Sci. 1991;74(10):3583–97. pmid:1660498
- 34. Walkley A, Black IA. An examination of the Degtjareff method for determining soil organic matter, and a proposed modification of the chromic acid titration method. Soil Sci. 1934. Available from: https://journals.lww.com/soilsci/citation/1934/01000/AN_EXAMINATION_OF_THE_DEGTJAREFF_METHOD_FOR.3.aspx
- 35. Cunniff P, Washington D. Métodos oficiales de análisis de la AOAC internacional. J AOAC Int. 1997. Available from: https://academic.oup.com/jaoac/article-abstract/80/6/127A/5684423
- 36. Palmonari A, Cavallini D, Sniffen CJ, Fernandes L, Holder P, Fusaro I. In vitro evaluation of sugar digestibility in molasses. Ital J Anim Sci. 2021;20:571–7.
- 37. Theodorou MK, Williams BA, Dhanoa MS, McAllan AB, France J. A simple gas production method using a pressure transducer to determine the fermentation kinetics of ruminant feeds. Animal Feed Science and Technology. 1994;48(3–4):185–97.
- 38. Theodorou MK, Williams BA, Dhanoa MS, Mcallan AB, France J. A simple gas production method using a pressure transducer to determine the fermentation kinetics of ruminant feeds. Elsevier. 1994;48:185–97.
- 39. Lavrenčič A, Mills C, Stefanon B. Aplicación del modelo de Gompertz para describir las características fermentativas de componentes químicos en forrajes. Ciencia animal. 1998.
- 40.
Johnson RA, Wichern DW. Applied Multivariate Statistical Analysis. New Jersey: Pearson Prentice Hall. 2007.
- 41.
Gu Z. Package: “circlize”: Circular Visualization Package Version 0.4.12. 2021.
- 42.
R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. 2024.
- 43.
Di Rienzo J, Casanoves F, Balzarini M, Gonzales L, Tablada M, Robledo C. InfoStat, versión 2018. Centro de Transferencia; 2020. Available from: https://scholar.google.es/scholar?hl=es&as_sdt=0%2C5&q=InfoStat%2C+versi%C3%B3n+2021.+Centro+de+Transferencia+InfoStat%2C+FCA%2C+Universidad+Nacional+de+C%C3%B3rdoba%2C+Argentina&btnG=
- 44.
Wicklin Rick, SAS Institute. Simulating data with SAS. 2013
- 45. Masebo NT, Marliani G, Shannon Del Re F, Abram L, Cavallini D, Di Pietro M, et al. Evaluation of antimicrobial and non-steroidal anti-inflammatory treatments for BRD on health and welfare in fattening bulls: a cross-sectional study. Vet Q. 2024;44(1):1–11. pmid:38711265
- 46. Cabrera Orrego R, Ríos-Osorio LA, Keynan Y, Rueda ZV, Gutiérrez LA. Molecular detection of Coxiella burnetii in livestock farmers and cattle from Magdalena Medio in Antioquia, Colombia. PLoS One. 2020;15(6):e0234360. pmid:32520958
- 47. Raspa F, Schiavone A, Pattono D, Galaverna D, Cavallini D, Vinassa M, et al. Pet feeding habits and the microbiological contamination of dog food bowls: effect of feed type, cleaning method and bowl material. BMC Vet Res. 2023;19(1):261. pmid:38062425
- 48. Rodríguez R, González N, Alonso J, Domínguez M, Sarduy L. Valor nutritivo de harinas de follaje de cuatro especies arbóreas tropicales para rumiantes. Revista Cubana de Ciencia Agrícola. 2014;48.
- 49. Suárez J, Ramírez B, Velásquez J. Comportamiento agronómico de cinco especies forrajeras bajo el sistema de corte y acarreo en suelos de terraza y mesón en el piedemonte amazónico. Zootec Trop. 2008;26:347–50.
- 50. Serda M, Becker FG, Cleary M, Team RM, Holtermann H, The D. Composición nutritiva y aceptabilidad del ensilaje de avena forrajera (Avena sativa), enriquecido con arbustivas: acacia (Acacia decurrens), chilca (Braccharis latifolia) y sauco (Sambucus nigra) en ovinos. Revista Veterinaria y Zootecnia (On Line). 2012;6:25–35.
- 51. Pinto-Ruiz R, Hernández D, Gómez H, Cobos MA, Quiroga R, Pezo D. Árboles forrajeros de tres regiones ganaderas de Chiapas, México: usos y características nutricionales. Universidad y Ciencia. 2010;26:19–31.
- 52. Ortega-Gómez R, Castillo-Gallegos E, Jarillo-Rodríguez J, Escobar-Hernández R, Ocaña-Zavaleta E, De La Mora B. Nutritive quality of ten grasses during the rainy season in a hot-humid climate and ultisol soil. Tropical and Subtropical Agroecosystems. 2011;13:481–91.
- 53. Chino Velasquez LB, Molina-Botero IC, Moscoso Muñoz JE, Gómez Bravo C. Relationship between Chemical Composition and In Vitro Methane Production of High Andean Grasses. Animals (Basel). 2022;12(18):2348. pmid:36139207
- 54. Quintero-Anzueta S, Molina-Botero IC, Ramirez-Navas JS, Rao I, Chirinda N, Barahona-Rosales R, et al. Nutritional Evaluation of Tropical Forage Grass Alone and Grass-Legume Diets to Reduce in vitro Methane Production. Front Sustain Food Syst. 2021;5.
- 55. García D, Medina M, Clavero T, Humbría J, Baldizán A, Dominguez C. Preferencia de árboles forrajeros por cabras en la zona baja de los andes Venezolanos. Revista Cientifica. 2008;18:549–55.
- 56. Vivas E, Rosado Rubio J, Castellanos Ruales A, Heredia y Aguilar M, Cabrera-Torres E. Contenido mineral de forrajes en predios de ovinocultores del estado de Yucatán. Revista mexicana de ciencias pecuarias. 2011;2:465–75.
- 57. Combellas J, Gonzales E. Rendimientos y valor nutritivo de forrages tropicales:. Agron, Tropical. 1972;:623–34.
- 58. Homen M, Entrena I, Arriojas L. Biomasa y valor nutritivo de tres gramíneas forrajeras en diferentes períodos del año en la zona de bosque húmedo tropical, Barlovento, estado Miranda. Zootec Trop. 2010;28:115–27.
- 59. Molina Botero I, Cantet J, Montoya S, Correa G, Barahona R. Producción de metano in vitro de dos gramíneas tropicales solas y mezcladas con Leucaena leucocephala o Gliricidia sepium. Medicina Veterinaria y Zoootecnia. 2013;8:15–31.
- 60.
Soto H, Garmendia J. Uso de fuente nacional de fósforo en la suplementación. 1997 [accessed 2023 November 8]. Available from: https://scholar.google.es/scholar?hl=es&as_sdt=0%2C5&q=Uso+de+fuente+nacional+de+f%C3%B3sforo+en+la+suplementaci%C3%B3n+de+novillas+Brahman%3B+%5BThe+use+national+sources+of+P+in+the+mineral+supplementation+of+Brahman+Heifers%5D.+&btnG=
- 61. Bhadauria T, Saxena KG. Role of earthworms in soil fertility maintenance through the production of biogenic structures. Applied and Environmental Soil Science. 2010;2010:1–7.
- 62. Ahmed N, Al-Mutairi KA. Earthworms effect on microbial population and soil fertility as well as their interaction with agriculture practices. Sustainability. 2022;14(13):7803.
- 63. Moore KJ, Lenssen AW, Fales SL. Factors affecting forage quality. Forages. Wiley. 2020:701–17.
- 64. Nelson CJ, Moser LE. Plant factors affecting forage quality. Forage Quality, Evaluation, and Utilization. 2015. p. 115–54.
- 65. Juárez Reyes M, Saúl A, Soto C, Andrea M, Ornelas G, Treviño R. Estimación del valor nutricional de pastos tropicales a partir de análisis convencionales y de la producción de gas in vitro. Técn pecu Méx. 2009.
- 66. Mould FL, Morgan R, Kliem KE, Krystallidou E. A review and simplification of the in vitro incubation medium. Animal Feed Science and Technology. 2005;123–124:155–72.
- 67. Quintanilla-Medina JJ, López-Aguirre D, Martínez-González JC, Limas-Martínez AG, Lucero-Magaña FA, Ruíz-García S, et al. Digestibilidad in vitro de dietas con diferentes niveles de inclusión de moringa (Moringa oleifera) para corderos en crecimiento. Rev investig vet Perú. 2020;31(3):e16840.
- 68. Elghandour M, Kholif A, López S, Mendoza G, Odongo N, Salem A. In vitro gas, methane, and carbon dioxide productions of high fibrous diet incubated with fecal inocula from horses in response to the supplementation with. Journal of Equine Veterinary Science. 2016.
- 69.
Ferro M, Castro W, Zanine A, Souza E. Cinética de fermentação ruminal in vitro de silagem de cana-de-açúcar com resíduo de cervejaria desidratado. 2017. Available from: https://gurupi.ufma.br/jspui/handle/123456789/676
- 70. Babayemi OJ, Demeyer D, Fievez V. Nutritive value and qualitative assessment of secondary compounds in seeds of eight tropical browse, shrub and pulse legumes. Commun Agric Appl Biol Sci. 2004;69(1):103–10. pmid:15560266
- 71. Nuñez-Torres O, Rodriguez-Barros M. Subproductos agrícolas, una alternativa en la alimentación de rumiantes ante el cambio climático. Scielo. 2019;6:24–37.