Forage and Fallows

Plot Diversification II

Lead Organization:

Grupo Yanapai

Partner Organizations:

Colorado State University, Universidad Nacional de Huancavelica, Vecinos Mundiales, Comunidad Campesina de Jacaspampa, Comunidad Campesina de Quilcas

Community of Practice:







Soils play a fundamental role in supporting agricultural productivity and other key ecosystem services. As demands on agriculture continue to increase, new strategies of smallholder farmers must ensure sustainable intensification of cropping and grazing practices, while promoting resilience to climate change and maintaining soil quality and function. Fallows are an important feature of Andean agriculture that are critical to soil regeneration. It is necessary to develop effective strategies to maintain their regenerative effects on soils and contribute to farmer livelihoods, particularly as fallows are shortened as part of cropping system intensification. Previous research has provided some understanding of soil fertility restoration dynamics in Andean fallows, but this work is limited in scope and largely represents only the Altiplano or the humid northern Andes. Recent findings from the Andes also emphasize the importance of soil gradients in determining the productivity of introduced species (legumes in particular) in improved fallows. These findings suggest great potential to capitalize on multi-purpose fallows, but also highlight the need to better understand fallow performance across a range of Andean biophysical conditions, and within the context of household livelihood portfolios and whole communities.This project will extend findings from the previous phase on forage/fallow options and will amplify the impact of the work using multi-environment experimentation of fallows and forages in new regions. It  will draw from local education efforts and regional/global information networks for agroecological intensification (AEI). Socioeconomic drivers of adaptation will be explicitly addressed by exploring how fallows can best complement other livelihood options and be managed by farmers/households within different livelihood typologies. The landscape inventory of soil health and ecosystem services from our first phase will be further refined by examining the impacts of heterogeneity, resource gradients, and edge effects within agricultural fields.  This effort will allow for better prediction of the impact of land use on ecosystem service production at multiple scales.The project will promote scaling out this learning and will support future agroecological innovation by working with local schools to develop new strategies and curriculum for promoting the responsible management of soils and agroecosystems. Local universities will be supported via seminars and training in soil quality assessment, as well as involving scientists and undergraduate students in on-farm research activities. The (Andean Landscapes) website will continue to be developed to share information about the project and foster collaboration and information exchange across the Andean region. Furthermore, a highly promising technician who has grown professionally during the first phase of this project would pursue formal training in a Master’s program at Colorado State University.

Grant Aims:

Capacity building of farmers: Greater management capacity by smallholders for sustainable management of mixed crop and livestock systems under intensification.Productivity of soils and landscapes: Improved soil productivity in landscapes under intensified management in the central Andes.Research and technical capacity: Greater research and technical capacity among development professionals and researchers of the region.Scaling up ecosystem services: Improved planning capacity and knowledge of outcomes concerning ecosystem services within local and regional governments of the region.Disseminating sustainable farming: Greater consciousness among current and future land managers and policy makers regarding the importance of sustainable farming methods and the important role of sustainable farming in the future of society.

Outputs and Outcomes:

Contextual understandingInitial results from participatory mapping for landscape assessment in 3 communities in 3 countries reveal that there are distinct gradients in land use intensity associated with changing altitude and land ownership. Land use tends to decrease with altitude at the sites in Peru and Bolivia but increase with elevation at the project site (2600-3200masl) in Ecuador probably because the site in Peru and Bolivia are already at a much higher altitude (3800-4100 masl), nearing the agriculture frontier and the lower lands in Ecuador have already been degraded through overuse.According to 45 families surveyed in the community of Carrillo, Ecuador, use of conservation measures is either correlated to the labour availability or to the total cultivated land area. The additional income from migration alone is never correlated to the adoption of conservation practices. Families who have more labor than income tend to cultivate more alfalfa for milk production — which requires daily work, but not much, and little monetary investment– and those that have more income from migration than labor focus on potato — which requires high monetary investment and peaks of labor. The third type of farmers that emerged from a cluster analysis have both labor and income and tend to balance milk and potato production.Shortening fallows are a principal challenge to soil health and ecosystem services in the high Andes. Increased frequency of planting/tillage, as well as soil erosion in mountainous landscapes, exposes fields with shortened fallows to soil degradation, while grazing on the limited land base of most communities may become insufficient to meet forage needs and to support the return of nutrients to soils via manure on more extensive cropped areas. This was evident from the stated need for forage in the planning among community members in the inception phase of research, and also in the fact that in many areas closer to the communities, fallows range from zero years (continuous cropping) to one or two years. Farther from the community, fallows are longer and input rates are lower, a common pattern of near and far fields that has been seen both in the Andes and in Africa.Pot studiesIn a semi controlled field environment in Quilcas, Peru, a pot study was undertaken to screen 70 potential forage/fallow plants using two soils representative of Quilcas and Castillapata for varieties that could quickly provide ground cover. The seed came from local and commercial varieties. There was no relationship between the source and the performance. There was more than 20-fold variability in total biomass production (above and below ground) between the best and worst performing varieties.  Some varieties of Vicia (vetch) and Festulolium performed well in both soils tested, many species seem to greatly prefer one soil over another, with the Castillapata soil generally yielding better (likely due to improved infiltration and higher pH). The perennials were predictably more slow growing and not included in a 2nd round of testingA second round of pot tests using soil from 2 different local sites in Peru used forage/fallow variaties that performed well in the first round (n=70) to study overyields, or if any in particular grass-legume combinations produced more together than they did in monocultures. A number of the mixtures yielded higher in N uptake than the average of their sole crop components, especially those with vetch and red clover as part of the mixture (p<0.05 in means testing using statistical contrasts). Interestingly, many mixtures grasses tended to dominate over legumes in biomass, even as total N uptake was enhanced with respect to the sole crops. Ryegrass and Festulolium were especially promising components of grass/legume mixtures.Field studies     biomass      – Peru forage-based fallows were tested in 58 collaborating farmers’ fields over 2 years. Fallow interventions were selected in community planning workshops and were comprised of predominantly grass/legume mixtures of annuals and perennials with an emphasis on forage provision. As expected, annuals produced considerably more biomass in the first year relative to perennials. Across the trials, seeded fallows exceeded the biomass and nutrient uptake of the unseeded, status quo, controls. In a mid-elevation zone (3400-3800 masl), biomass increases ranged from 160 to 390% of the unseeded control (p<0.05), while in another higher-elevation (3800-4100 masl) community biomass was approximately 8x the control. However, in another high-elevation site only 2 treatments somewhat exceeded the control, and the local unseeded control was more competitive in establishing cover within the first 2 months. Thus, the biomass and soil cover advantages were most notable in the middle elevation zone which has the greatest tendency towards fallows intensification (2-3 year fallows).      – A second year of field studies was done in 2 sites in Peru informed by the results of pot experiments (years 1 and 2), as well as feedback from focal group evaluations of the year 1 field trials. The results show that a highly diverse mixture (8 species across grasses and legumes and annuals/perennials) produced the highest biomass in the Quilcas middle zone (3400-3800 masl). These fallows might also be a promising way to establish short-cycle perennials that would serve important soil regeneration functions during a shortened fallow of 2-3 years, which is now quite common in the intensified sectors of Andean communities      nutrition      – Results from seeded fallow experiments in 58 farmers’ fields in Peru suggest that forage resources provided by managed fallows are of higher quality (higher protein, lower NDF, higher caloric content) and quantity in comparison to grazing of unseeded fallows, with the potential to augment the manure resources available for maintaining soil fertility during cropped phases of the rotation.      legumes      – Results from seeded fallow experiments using grass/legume and annual/perennial mixtures in 58 farmers’ fields at one landscape in Peru show legumes such as vetch, lupin, and red clover don’t appear to perform well at higher elevations.  Fallow performance was associated with site soil fertility and the success of legumes within mixtures. The annual legumes Vicia sativa and Lupinus mutabilis responded differently to soil pH gradients, indicating a need to consider soil contexts when selecting fallow options. Farmers noted high density of grasses seeded or late seeding date of fallows with respect to onset of rains can also negatively affect legume biomass. Those site with greater legume proportion had dramatically higher P and N concentrations in fallow biomass (p<0.05) Annual legumes (which will tend to dominate the biomass in this first sampling) were especially unproductive in the second year of the experiments, especially at high elevations. However, a perennial legume like alfalfa improved greatly in performance over the multiyear experiment.      Farmer evaluation     – Farmer evaluations included focus groups with participants (n=30) and surveys of entire communities (n=101) of seeded forage experiments at 2 sites in Peru. Over 80% of males and females surveyed said they would continue to use oats/vetch and oats/vetch/perennials again. Reasons for the former included it produced high levels of biomass, fed a wide variety of animals, vetch is valued for increasing milk production, crowds out weed, and good for non-irrigated plots. Downsides mentioned include that elevation and fertility determine success and doesn’t regrow. In general, forage production by the fallows, either this year and in future years, was ranked highly, and soil improvement functions were less apparent in the ratings. Unmanaged fallow was less work, but might produce less forage. This emphasis on forage benefits instead of soil could lead to overall soil degradation/mining if manure is not returned to the soil and forage is overharvested/ pastured. The higher ratings by men vs. women in the survey data to lupine-based fallows could suggest a greater value towards crop production for men, who might see in a fallow that starts with lupine as in effect extending the crop rotation with an additional year of pulse production; this would support the hypothesis that men are more interested in grain versus forage production in these communities.      – As a result of seeing trial results and with collaboration from Grupo Yanapai in Peru, the local project partner, a community, seeded several large areas to intercrops of lupine with oats, ryegrass, and clover in different combinations, in order to expand the results to a larger scale and test the use of lupine at a high elevation site.      Drier year      – A second year of field studies on seed fallows was done in 3 sites in Peru, under drier conditions. Across different soil contexts, predictors of biomass results in the experiment showed some interesting contrasts to the first round of field trials. The effects of soil pH were far less important in this round of trials, and the effect of soil fertility somewhat less important, than in the first round, although P, which was positively correlated with fallow biomass in year 1 in all 3 sites, continued being the strongest soil fertility predictor of biomass at two of the sites. Organic matter was associated with increased OM the lowest elevation site, while lacking any significant association to biomass at the other sites. Results are consistent with performance of fallows in a drier year: water is more limiting than nutrients, so the effects of nutrients decline, especially at the warmest low-elevation site; organic matter and clay content of soils become important in water retention and increased yields at lower elevation, but not at higher elevation sites dominated by light textured, high organic matter soils.LandscapeParticipatory mapping of 3 landscapes, in Peru (3400-3800 masl), Bolivia (3800-4100 masl) and Ecuador (2600-3400), led to the identification of key landscape types by altitude and use (crops, forests, hedgerows, pastures – with subcategories). The chemical, physical and biological properties of soil of the different areas showed that degraded pastures are the most soil infertile land types in the landscape and have high levels of erosion, followed by cropped systems (including irrigated forage) and then the best soil health was in forests (PE)/high elevation pastures /hedgerows depending on site and variable. This shows that a focus on pastures helps address the best and worse land management practices. The different land uses displayed significant differences among the 3 variables, but that they are tightly linked, so changing one will likely effect the other two. In all sites elevation was the most important explanatory variable with higher levels of soil organic matter, and available P due to slow degradation/cooler temps.Participatory mapping of 3 landscapes, in Peru (3400-3800 masl), Bolivia (3800-4100 masl) and Ecuador (2600-3400), led to the identification of key landscape types by altitude and use (crops, forests, hedgerows, pastures – with subcategories). The chemical, physical and biological properties of soil of the different areas showed that forested systems (alder, eucalyptus and mixed) exhibit the highest diversity of soil macrofauna in the Peruvian site but the low for eucalyptus in EC and BO (maybe because of extreme slope). At all sites these forests have high levels of C and biomass, except Eucalyptus in BO. At the Ecuador site, hedgerows were important for C storage and biodiversity conservation and there is a general gradient of soil fertility and ecosystem service provision that increases from low to high elevations, suggesting that both past management and environmental factors have led to an overall degradation at lower elevations, mostly from agriculture use.In community workshops in 3 sites in 3 countries, results from sampling the chemical, physical and biological properties of soil in different land use areas within a landscape were looked at to inform future planning activities that revolved around OxC. In all 3 communities, the whole community was convened, and in QUilcas a good percentage of the members attended the workshops. In EC and Bolivia it was more a subset of the community that participated more in the research. The community in Ecuador (n=15) had no problem imagining many future scenarios based on how they managed the land. In Quilcas (n=70) there were many ideas such as planting more seeded forages in irrigated degraded areas, planting alder forests near to water storage areas, and to rent out fields less because renters tend to degrade soil. The meeting in Bolivia (n=24) was less formal and the participants did not want to engage in too much planning in this stage, but were very interested in the data and bar graphs and engaged in a lot of interpretation of the data vis a vis their activities. In Peru, the data was also shared for more local political influence (regional govt Quilcas working tables) and through civil society participation in the directorate of the Area de Conservación Regional Huaytapallana, which seeks to found a landscape conservancy that would protect the Huaytapallana mountain range in order to protect urban and rural water resources for the region and the extensive areas of high Andean pasture and wild vegetation. They sponsored a forum to discuss the results in the region with 170 participants representing 21 institutions.