Given their high nitrogen (N) concentration and low costs, sweet potato vine silage (SPVS) and urea-molasses blocks (UMB) are recommended supplements for tropical regions; therefore, they were investigated in this study. Six heifers were allocated to three diets: the roughage diet (R) consisted of wheat straw (0.61) and Rhodes grass hay (0.39; on dry matter (DM) basis); R + SPVS combined R (0.81) and SPVS (0.19); and with R + UMB animals had access to UMB. During two experimental periods, feed intake, feces and urine excretion, digesta passage, and rumen microbial protein synthesis were determined during seven days and methane emissions during three days. There was no treatment effect (p > 0.05) on DM and N intake. Apparent DM digestibility of R + SPVS (510 g/kg) was higher (p < 0.05) than of R (474 g/kg). Digesta passage and duodenal microbial N flow were similar for all diets (p > 0.05), while N retention was highest with R + SPVS (p > 0.05). Methane emissions per unit of digested feed (g CH₄/kg dDM) were lower (p < 0.05) for R + SPVS (55.2) than for R (64.7). Hence, SPVS supplementation to poor–quality roughage has the potential to increase diet digestibility and N retention while reducing CH₄ emissions.

In Africa, cattle are often fed low quality tropical roughages resulting in low-yielding animals with high methane (CH₄) emission intensity (EI, g CH₄/per unit of product). Supplementation with protein is known to improve the nutritive value of the otherwise low-quality diets. However, animal nutrition studies in East Africa that are accompanied by CH₄ emission measurements are lacking. Thus, an animal experiment was conducted to quantify the effect of supplementing cattle fed mainly on low-quality Urochloa brizantha hay (control diet; CON; crude protein (CP) = 7.4%) or supplemented with either a tannin-rich leguminous fodder, Desmodium intortum hay (DES) or a commercial dairy concentrate (CUBES) on voluntary dry matter intake (DMI), nutrient apparent total tract digestibility, nitrogen (N) retention, enteric CH₄ production and animal performance (milk and average daily gain). Twelve mid-lactating crossbred (Friesian × Boran) cows (initial liveweight = 335 kg) were used in a 3×3 (Period × Diet) Latin square design with each period running for four weeks. Compared to CON, DES decreased nutrient (DM, OM, CP) intake, apparent total tract digestibility and daily milk yield. In contrast, CUBES increased nutrient intake and animal performance compared to CON, while nutrients’ apparent total tract digestibility was not different, except for CP digestibility that increased. Compared to CON, DES and CUBES improved overall N retention by the animals as a proportion of N intake. The DES diet compared with CON and CUBES, shifted the proportion of N excretion via urine to the fecal route, likely because of its tannin content. Both DES and CUBES, compared to CON, reduced methane yield (MY, g CH₄/kg DMI) by 15% and 9%, respectively. The DES diet reduced absolute enteric CH₄ emissions by 26% while CUBES increased emissions by 11% compared to CON. Based on the present findings, high supplementation levels (>50%) of Desmodium intortum hay is not recommended especially when the basal diet is low in CP content. Supplementation with lower levels of better managed Desmodium intortum forage however, need to be investigated to establish optimal inclusion levels that will improve animal productivity and reduce environmental impact of livestock in smallholder tropical contexts.

Effects of feeding levels below maintenance requirements of metabolizable energy (MER) and of feed supplementation on fecal nutrient and microbial C concentrations were evaluated. In experiment 1, Rhodes grass hay only was offered to Boran steers at 80%, 60%, and 40% of individual MER, while steers at 100% MER additionally received a concentrated mixture. This reduction in MER decreased N, increased fungal C but did not affect bacterial C concentrations in feces. In experiment 2, Holstein × Boran heifers were offered a poor-quality roughage diet without supplement, with sweet potato vine silage or with a urea-molasses block. These two supplements did not affect the fecal chemical composition or fungal C but increased bacterial C concentrations in feces. Across all data, the fungal C/bacterial C ratio was positively related to N and negatively to neutral detergent fiber concentrations in feces, indicating diet-induced shifts in the fecal microbial community.