Vol.3, No.5, 768-779 (2012) Agricultural Sciences
Integrated management of natural resources in the
Ecuador Highlands
Víctor Hugo Barrera1*, Luis Orlando Escudero1, Jeffrey Alwang2, Robert Andrade2
1Instituto Nacional Autonomo de Investigaciones Agropecuarias, Quito, Ecuador;
*Corresponding Author: victor.barrera@iniap.gob.ec
2Department of Agricultural and Applied Economics, Virginia Polytechnic Institute and State University, Blacksburg, USA;
Received 18 March 2012; revised 27 April 2012; accepted 9 May 2012
The Andean region of Ecuador is characterized by
extreme poverty caused by low agricultural pro-
ductivity, limited off-farm opportunities, and lack
of access to markets. Poverty is related to degra-
dation of natural resources as lagging agricultural
productivity leads to incursions into fragile areas
and use of eros ive farming tec hniques on steeply
sloped hillsides. Food production in fragile areas
degrades s oil and w ater resources, c ontributes to
deforest ation and loss of biodiv ersity, and r educes
productive potential over time. This article dis-
cusses an agricultural development project de-
signed to reduce the long-term downward de-
velopment spiral in a watershed in Bolivar, Ec-
uador. The applied resea rch program began w ith
analysis of the state of soil resources, w ater, and
biodiversity in the Chimbo sub-watershed. This
information was used to design a plan with the
input of local stake-holders to introduce envi-
ronmentally friendly farming practices, soil and
water conservation techniques, and various in-
stitutional innovations to promote resource con-
servation. This adaptive management program
has been a solid success. This article describes
the project, the challenges it faced, and how the
process of adaptive management led to consen-
sus among stakeholders about the appropriate-
ness of sustainable management practices. We
show how implementation of enhanced man-
agement practices contribute to reduced envi-
ronment al vulnerability and improved welfare.
Keywords: Component; Natural Capital;
Micro-Watershed; Systems Approach;
Adaptive and Integrated Watershed Management
The South American Andes are rife with environmen-
tal problems related to human activities in fragile eco-
systems. Andean populations are among the poorest in
South America and often depend on rain-fed agriculture.
The Andes form the headwaters of many of the great
river systems of South America, and runoff and agricul-
ture-related pollution can have negative consequences far
from their sources. Humans are encroaching into fragile
high plains as population pressures at lower elevations
extend the agricultural frontier. Strategies to address
these problems include more environmentally benign
agricultural technologies in fragile areas, intensified pro-
duction in less-fragile areas to reduce pressure on more
fragile areas, and raising income-earning potential
through less land-intensive activities. A key is to alter
human behavior. Adaptive management processes show
promise as means of altering behavior to attain agreed-
upon goals.
A watershed approach to natural resource management
has been tried in different settings with varying degrees
of success. Watersheds define natural linkages between
human populations and their environments [1]. Water-
shed management is consistent with decentralized gov-
ernance, which is gaining favor in Andean countries [2].
However, modern watershed management techniques
require digitized data that are of limited availability in
high mountain areas, and watershed management often
requires the cooperation of competing and overlapping
levels of local and regional government. A watershed
management approach faces many challenges.
Any watershed approach must begin with the notion
that watershed-level outcomes are products of individual
decisions on fields spread across the catchment’s area.
These decisions reflect household livelihood strategies of
allocating their physical, human, natural, and other assets
to earn livings, increase well-being, and manage multiple
risks [3]. Individual decisions have compound effects
and impacts on aggregate economic and environmental
outcomes result from a complex mosaic of economic,
social, and physical networks that characterize all water-
Copyright © 2012 SciRes. OPEN ACCESS
V. H. Barrera et al. / Agricultural Sciences 3 (2012) 768-779 769
sheds. The driving factor is human decision making. Ef-
fective management must identify mechanisms for chang-
ing human activities and introduce options to raise incomes
while mitigating negative environmental consequences.
Integrated adaptive watershed management is a rela-
tively new concept in Ecuador, but it provides hope that
some environmental problems can be addressed through
consensus building. The 1970s-era focus of tops-down
watershed management has evolved over time and newer
concepts recognize the holistic nature of the relationship
between land use, agricultural production, natural re-
source conservation, and reduction of contaminants. It
also recognizes that watershed outcomes result from hu-
man decisions [4].
Our integrated watershed management program in the
Chimbo sub-watershed in Bolivar Province is guided by
four concepts: 1) Agricultural intensification can be con-
sistent with sustainable natural resource management [5];
2) Sustainable agricultural practices can contribute to
preservation of bio-diversity [6]; 3) Increased bio-diver-
sity can contribute to household food security by diver-
sifying diets and reducing risks of crop failure [7]; and 4)
Even the poorest of the poor are interested in and capable
of adopting environmentally friendly technologies [6].
Evidence shows that these arguments are valid in the
Ecuadorean highlands [8]. Ecuador’s National Autono-
mous Agricultural Research Institute (INIAP) has en-
gaged farmers in the Chimbo for many years and has
found farmers to be receptive to solutions to natural re-
source problems [8]. Over time, INIAP has created im-
portant strategic alliances and generated broad support
for integrated adaptive watershed management. INIAP
now combines an integrated adaptive management ap-
proach with a livelihoods focus, recognizing that any
effort to improve environmental conditions must also
create economic space (i.e. provide sufficient incomes to
maintain a family) for conservation actions.
The objectives of this paper are to describe the adap-
tive watershed management process, obstacles overcome
during its implementation, and provide a preliminary
assessment of program impacts. We describe the site,
present our research methods, and identify specific in-
novations attributable to the research. We then discuss
research findings with respect to returns to management
practices and describe how the recommended practices
have spread over time. The paper concludes by discuss-
ing lessons learned and how the adaptive management
process can be applied to other areas.
The Chimbo River sub-watershed covers approxima-
tely 3.635 km2 (Figure 1); our program focuses on two
micro-watersheds: Illangama and Alumbre. The Illan-
gama micro-watershed covers 131 km2 and extends from
Figure 1. Location of study. The Chimbo River sub-watershed-
a latitude of 1˚23'55.30''S through 34'4.80''S and from
78˚50'39.38''W to 78˚58'29.52''W. The Alumbre covers 65
km2 and extends from 1˚54'29.14''S to 2˚1'36.90''S and
from 79˚0'22.20''W to 79° 6'4.41''W [9]. The Illangama is
between 2800 and 4500 masl, with agricultural activity
found between 2800 and 3600 masl. The Alumbre area
ranges from 2000 to 2800 masl with agriculture through-
hout [10].
The watershed is characterized by social and economic
conditions that threaten environmental sustainability and
create long-term risks to human populations [11]. The
area is among the poorest in Ecuador [12]. The river sys-
tem flowing through the watershed provides about 40%
of the total flow to the Guayas River, the largest system
in Coastal Ecuador. Water quantity and quality has de-
clined in recent years, partly due to upstream erosion,
deforestation and expansion of the agricultural frontier
into fragile highlands [13]. The highest (páramo) areas
are reservoirs of clean water, and incursions into them
have major downstream effects [14].
Households in the area depend on agriculture; more
than 60% of the economically active population in Boli-
var Province is dedicated to agriculture. Agriculture is
characterized by small holdings, low productivity and
environmental degradation (Table 1). Steep slopes, ir-
regular and sudden rainfall, and infrequent use of cover
Copyright © 2012 SciRes. OPEN ACCESS
V. H. Barrera et al. / Agricultural Sciences 3 (2012) 768-779
Copyright © 2012 SciRes. OPEN ACCESS
Table 1. Conditions in Chimbo sub-watershed.
Agro-ecological conditions Productive activi ties
Region: Páramo and Andean mesa
Life zones: Subalpine or boreal, montain, low mountain
and cool temperate
Temperature ˚C: 7 - 13
Altitude m: 2800 - 5000
Annual rainfall: 500 - 1300 mm
Agriculture—potato (Solanum tuberosum ), pasture, quinoa (Chenopodium quinoa),
faba (Vicia faba), chocho (Lupinus mutabilis) and barley (Hordeum vulgare);
Animal production—cattle, swine, sheep and guinea pigs;
Tourism, artisan production, commerce, cheese production and sales.
Region: Andean mesa and subtropical
Life zone: Low mountain and pre mountain
Temperature ˚C: 15 - 19
Altitude m: 2000 - 2800
Annual rainfall: 750 - 1400 mm
Agriculture—maize (Zea mays), beans (Phaseolus vulgaris), peas (Pisum sativum),
blackberry (Rubus glaucus), tree tomato (Ciphom andrea betacea), vine tomatoes
(Lycopersicum esculentum);
Animal production—poultry, swine;
Agro-industry—medicinal plants, cacao (Theobroma cacao), organic coffee
(Coffea arabica);
Tourism, small-scale commerce, artisanal production.
crops and other means of conserving soils cause severe
soil erosion.
The program was structured around an adaptive water-
shed management conceptual framework. This frame-
work begins with the watershed as a geographic entity
and recognizes that actors within the watershed make
decisions that affect the entire watershed. The adaptive
management framework is well-known [15] but has
rarely been applied in a developing country context. It
begins with an assessment of conditions and identifica-
tion of problems faced by actors in the watershed.
Stakeholders are engaged in goal-setting, and research is
designed to address obstacles to achieving goals. Re-
search findings are then used in a participatory process
with stakeholders to produce watershed plans. These
plans are implemented and outcomes are monitored.
Monitoring could lead to changes in plans over time, and
the adaptive cycle begins again. We introduce two inno-
vations to this framework: Plans are adapted on a regular
basis as the research base and acceptance of it grows, and
the land-use plans include consideration of household
decision making and how decisions create impact across
multiple systems within the watershed.
The household decision process reflects livelihood
choices. A livelihood refers to the capabilities, assets
(stores, resources, claims, access), and activities required
for a means of living [16], or how labor, land, and other
assets are distributed among productive and reproductive
activities. The decision to adopt a livelihood is based on
the household asset base; available alternatives; institu-
tional, policy, and social environments; access to infor-
mation; and the natural environment. Asset allocation
decisions have effects on household wellbeing, the abil-
ity to save and invest, and the natural environment. For
example, adoption of a maize technology affects labor
and land allocations, income, risk exposure, and may
affect erosion, runoff, and future soil quality. All these
outcomes were identified as important during implemen-
tation of the adaptive management process.
The management program was built on four dimen-
sions: communication, coordination, compromise and co-
operation. The project facilitated movement along these
dimensions through regular community meetings and a
process of participatory research. Interactions helped ge-
nerate consensus about key problems and solutions most
likely to be successful.
Our assessment began with a participatory rural ap-
praisal (PRA) to identify productive activities, assets,
and perceptions about environmental conditions. The
PRA was followed by a statistically representative house-
hold survey that collected information from 286 families.
The survey covered household demographics, assets,
sources of income, agricultural practices and others.
These data were used to categorize households into live-
lihood typologies and conduct analysis of household de-
cision-making processes. Survey observations were geo-
referenced, which allowed us to overlay survey informa-
tion with agro-ecological, soils, infrastructure and other
information in a GIS. The GIS was used to create the-
matic maps for the community engagement process, and
to inform and structure research.
Emphasis was placed on identifying alternatives and
evaluating them through hands-on research. For example,
conversion of lands to permanent pasture or reversion to
woodlands was not initially viewed as desirable. An as-
sessment of biodiversity, together with research on alter-
natives to reduce erosion on productive lands, helped
convince stakeholders that a combination of reversion
together with adoption of erosion control practices in the
V. H. Barrera et al. / Agricultural Sciences 3 (2012) 768-779 771
most erosion-sensitive areas would help meet objectives
about which consensus had been reached. Similar re-
search efforts were undertaken to help find more effec-
tive soil fertility management regimes, more environ-
mentally benign pest control methods, etc.
Livelihoods and their diversity: The baseline survey
and information from the PRA were used to identify live-
lihood clusters. A quantitative hierarchical (data based)
clustering method [17,18] was combined with expert
opinion to create these clusters. The livelihood clusters
can be thought of as groupings of households with simi-
lar asset bases and different means of combining them to
earn incomes. Some clusters were exclusively agriculture,
others rely on off-farm incomes, and others on remit-
tances from outside the area [19].
Water quality analysis: Early in the process, stake-
holders decided that water quality should be monitored.
Monitoring results were used to evaluate impacts of
land-use changes on water quality. We measured bio-
indicators (macro-invertebrates), physical-chemical com-
positions, and micro-biological parameters [20]. This
monitoring helped engage community members and built
ownership of the research. Key macro-invertebrates were
identified in exercises with local school children during
2006 and 2007; subsequent monitoring was incorporated
into the local curriculum [20]. Monthly chemical analy-
sis begins with samples being extracted by community
members and sent to Quito for detailed analysis. Nitrate,
phosphorus, total solids, temperature, pH, conductivity,
fecal coliform, and total coliform are all measured. Data
on rainfall and stream flows are being collected and used
to calibrate our watershed models (mainly the Soil and
Water Assessment Tool-SWAT).
Biodiversity assessment: The PRA indicated that stake-
holders were not aware of biodiversity or its importance.
Early on, an assessment activity evaluated the richness
and diversity of plant and animal species. The focus was
on remaining natural woodlands and areas where water
recharge occurs (mainly at upper elevations). The as-
sessment incorporated local knowledge about the value,
uses and abundance of native plant and animal species.
Stakeholders helped transect the study area, and collect
photographic and physical evidence. Evidence was clas-
sified and categorized at the National Herbarium in
Quito. Strategic transects were also undertaken in rem-
nant woodlands and areas of high vulnerability [21].
Physical and environmental vulnerability: We strati-
fied our on-farm agronomic research (on pilot farms)
according to an index of physical vulnerability which
included six parameters: slope, vegetative coverage, rain-
fall frequency and intensity, wind intensity, seasonal va-
riability and soil texture. These indicators were selected
following focus group discussions with technicians and
local farmers. The index takes a value between 0 and 1,
with 0 signifying no vulnerability and 1 representing
areas of extreme vulnerability. The index was especially
useful in helping producers understand linkages between
farming practices, soil loss, and subsequent off-farm
damages. Farmers had their parcels classified and the
index values and information on actual land use were
incorporated into the GIS. This information was used to
identify environmental hot-spots and to inform subse-
quent land use plans.
Design of environmentally friendly alternatives: The
program selected 13 production systems for research on
more sustainable practices. All the practices were con-
sistent with livelihood clusters. Illangama systems re-
volved around a well-established potato-dairy rotation,
while in Alumbre maize-beans predominate (Table 2).
Trials were established on pilot farms to evaluate impacts
on income, labor use, environmental degradation, etc. of
these practices. Best Management Practices (BMP) were
targeted for implementation in high-vulnerability areas
(Ta b le 2). Farm and watershed-wide plans were created
following consensus-building exercises with stakehold-
Four livelihood clusters were identified in the micro-
watersheds (Table 3). These reflect diversity in asset
bases, use of productive resources, and impacts on the
environment across the clusters [22].
In Illangama, most family incomes are based on agri-
cultural production and work in agriculture off the farm.
In contrast, Alumbre households use a combination of
agricultural (own-farm) and diversified off-farm activi-
ties (Table 4). Households that are more dependent on
agriculture and livestock possess and use more natural
capital (mainly land) and physical capital (farm equip-
ment and implements). Those that depend more on off-
farm incomes have higher levels of human capital (re-
flected through education of adult members and chil-
dren’s participation in school). They also possess spe-
cialized skills, such as carpentry, masonry, etc. House-
holds with greater quantities and qualities of natural and
physical capital are most closely linked to agricultural
markets and only infrequently participate in non-agri-
cultural income-generating activities. They specialize in
agriculture. Households with diversified off-farm in-
comes sources tend to be wealthiest and have fewer food
security challenges.
4.1. Water Quality Analysis
Water quality analysis confirmed perceptions from the
PRA that water quality is degraded in both watersheds.
In Illangama, only two areas (Culebrillas and Quindigua)
had water quality suitable for livestock consumption and
Copyright © 2012 SciRes. OPEN ACCESS
V. H. Barrera et al. / Agricultural Sciences 3 (2012) 768-779
Copyright © 2012 SciRes. OPEN ACCESS
Table 2. Sustainable agricultural production practices evaluated.
Illangama Alumbre
Deviation ditches with milín grass (Phalaris tuberosa) and native
Improved rotations: Natural pasture and improved varieties of
potato-barley—faba, and quinoa;
Live barriers with native species (yagual, tilo, romerillo, piquil,
chachacoma, aliso, higuerón, tilo);
Chocho associated with improved pasture;
Improved planting and harvest schedule (to manage water and
Cultivation in belts (wheat, barley) with improved varieties from
Improved pastures with forage mixes of annual rye grass,
bluegrass, white and red clover, biannual and perennial rye grass.
Belt/strip cultivation (maize and climbing beans);
Live barriers with native species (nogal, alisos, siete cueros and
Bench terraces and horticultural production;
Fruit trees on contours to form live barriers (chirimoya, lemon,
orange, avocado and blackberry);
Reduced tillage of beans and peas;
Improved pastures with forage mixes of annual rye grass, bluegrass,
white and red clover, biannual and perennial rye grass;
Crop rotations (improved varieties of): Hard maize—climbing/bush
beans. Evaluation of promising germplasm;
Contour planting and introduction of alfalfa (Pennisetum sp.) in
Table 3. Livelihood clusters in sub-watershed Chimbo River-Ecuador.
Livelihoods Percent Households Members
Diversified households (A) 27 78 432
Engaged in agricultural markets (B) 37 105 576
Rural non-farm economy (C) 17 50 218
Agricultural consumption and wage work (D) 19 53 241
Total 100 286 1,467
Source: Original analysis using household survey (Andrade, 2008).
Table 4. Summary statistics for main variables and livelihoods in Alumbre.
Va riables
Micro-watershed alumbre % 46 37 98 85 0.00***
Land holding (ha) 3.82 6.79 3.59 3.64 0.00***
Irrigation access % 23 33 6 9 0.00***
Value physical assets $ 2008 2348 856 496 0.00***
Distance to closest river (km) 1.12 0.86 2.05 1.58 0.00***
Distance to closest city (km) 7.21 7.58 3.61 5.17 0.00***
Participation in civil societies % 60 55 26 38 0.00***
Family members that migrate % 71 39 54 13 0.00***
Mestizo households % 31 25 64 53 0.00***
Household size 5.54 5.49 4.36 4.55 0.00***
Household head male % 88 90 82 72 0.02**
Secondary education or plus % 65 65 66 45 0.09*
Income per capita annually $ 653 785 839 288 0.00***
Expenditures per capita annually $ 254 252 252 184 0.03**
Source: Original analysis using household survey (Andrade, 2008). ***Significant at less than 1% level; **Significant at less than 5% level; *Significant at less
an 10% level. th
V. H. Barrera et al. / Agricultural Sciences 3 (2012) 768-779 773
no sampling site had quality suitable for human con-
sumption (Table 5). Fecal coliform contamination is a
severe problem throughout the watershed; in 100% of the
samples, we detected fecal coliform (E. coli). Highest
values were found in Paltabamba and Quindigua and
were related to the presence of trout lagoons near the
sampling site and human waste. In other areas, livestock
grazing in upstream water sources and agricultural runoff
are key contributors to water quality degradation. Even at
very high elevations in near-pristine environments, water
quality is a problem.
In the lower-elevation Alumbre, physical chemical pa-
rameters fall within normal levels and are below na-
tional limits. Indicators of total solids and turbidity indi-
cate significant siltation from erosion. Measures of fecal
and total coliform indicate severe contamination (Table
5). Coliforms (E. coli and others) were detected in all the
samples. The highest concentration of fecal coliform was
detected in Chillanes (2240 UFC/100 cc), because it is
found at the confluence of two smaller rivers carrying
human wastewaters from urban centers. The second
highest coliform count was found in Pacay, mainly be-
cause of high concentrations of cattle and swine near the
river and human wastewaters.
4.2. Biodiversity Assessment
The micro-watersheds are distinguished by major dif-
ferences in flora biodiversity. We identified around 162
tree and bush species in the area. In Illangama and
Alumbre we identified 13 and 32 species, respectively,
unique to that micro-watershed. Only 17 families of spe-
cies were common to both micro-watersheds. Biodiver-
sity is far richer in Alumbre, where the warmer climate is
more conducive to species diversification.
Tree and bush species form a significant part of live-
lihood systems, particularly in Illangama. In Illangama,
families have strong interest in species that can be used
as animal forage, firewood and charcoal, and varieties
with medicinal properties. They also use trees as live
barriers in soil conservation structures, to extract dyes,
and fibers for artisanal products. Alumbre residents are
less aware of the uses of tree species, are unfamiliar with
local names, and are mainly interested in trees for the
exploitable wood they can produce.
The water quality and the biodiversity assessments
were designed to highlight the fragility in high-altitude.
The initial assessments showed that farmers’ voice con-
cerns about environmental quality, recognize that their
productive practices can create environmental damage,
and seek alternatives to resource-mining activities. The
strongest concern for the environment was voiced by
community members in the upper watershed who recog-
nize environmental change such as variable rainfall pat-
terns, less water availability and others. The assessments
also strengthened linkages between the research team
and community members; the participatory means of con-
ducting them and open sharing of findings built owner-
ship of the adaptive management process.
4.3. Physical and Environmental
In order to prioritize interventions, stakeholders need
information on vulnerability and its variability over
space. The GIS was combined with watershed modeling
to generate a map of vulnerability to runoff. Results
showed that about 4000 ha in Illangama and 2000 ha in
Alumbre are extremely vulnerable to environmental da-
mage. These areas needed special attention during the
planning and management phases.
The vulnerability mapping exercise uncovered evi-
dence of conflicts between ideal and actual land uses:
Some of the most environmentally vulnerable lands are
currently under intensive crop production. These areas
should be reserved for conservation or reforested and
managed sustainably. Research thus focused on the phy-
sical and economic/social consequences of less intensive
Table 5. Analysis of variance for microbiological indicators of water quality in sub-watershed Chimbo River-Ecuador.
Illangama Alumbre
Collection sites UFC
E. coli /100 cc UFC colif.
total/100 cc Collection sites UFC
E. coli/100 cc UFC colif.
total/100 cc
Culebrillas (3495 m) 550 c 4243 b Chillanes (2274 m) 2240 a 14,926 a
Quindigua (2930 m) 1075 ab 6793 a Pacay (2240 m) 840 ab 5746 b
Quindigua (2886 m) 600 bc 4462 b Guayabal (2193 m) 487 b 4346 b
Paltabamba (2723 m) 1244 a 6512 a
Meang 867 5503
Meang 1189 8340
P 0.0136 0.0025 P 0.0155 0.0001
ource: INIAP-SANREM CRSP-SENACYT, 2009. Letters indicate statistical significance (P 0.05).
Copyright © 2012 SciRes. OPEN ACCESS
V. H. Barrera et al. / Agricultural Sciences 3 (2012) 768-779
uses on vulnerable lands. Such activities are especially
critical in areas of water recharge. Two main challenges
constrain efforts at conservation in these areas: Lack of
finance to ensure that households can survive during the
transition from intensive to extensive production (such as
forestry or agro-forestry systems), and low rates of return
in extensive production systems. Few own sufficient land
resources to sustain a family on forestry production.
Part of the problem is institutional; farmers have no
means of capturing the off-farm benefits from less in-
tensive land use. The team began negotiations with down-
stream govern ments to examine if these governments
were willing to pay farmers to avoid downstream dam-
ages. These negotiations are ongoing, but downstream
siltation is increasingly associated with costly flooding
and there is strong interest in finding low-cost ways to
avoid these damages.
4.4. Design of Environmentall y Friendly
Farming Alternatives
The research identified several environmentally friendly
agricultural production options for farms in the Chimbo
(Table 2). These alternatives increase productivity, en-
hance soil retention and improve soil health. They were
tested on model farms, where farmers participated in site
preparation, cultivation, and evaluation. Field days de-
monstrated the practices to farmers. Substantial adoption
has occurred already and, given the success of the alter-
natives, we expect more widespread adoption as infor-
mation becomes more widespread.
The pilot sites were established on farms with average
sizes of 7.5 ha. In 2006, production systems included small
pine forests, natural pasture, small areas of improved pas-
ture, and potato production together with mashua (Trop-
aeolum tuberosum). At that time, the most vulnerable
areas had been devoted to crop production. The research
team designed a farm use plan incorporating improved
cropping systems and farming practices, pastures and
woodlands (Figure 2).
We tested and subsequently recommended use of im-
proved potato varieties, faba beans, barley, quinoa and cho-
cho. Conservation agriculture practices such as improved
igure 2. Land use map for model production system in Illangama, 2006.
Copyright © 2012 SciRes. OPEN ACCESS
V. H. Barrera et al. / Agricultural Sciences 3 (2012) 768-779 775
rotations, reduced tillage, and increased groundcover
were included (Table 2 ). We also recommended in-situ
conservation of native Andean tubers such as native po-
tatoes, oca (Oxalis tuberosa), melloco (Ullucus tubero-
sus), mashua and carrot (Daucus caraota).
As of 2010, the project had been functioning for 5
years and it was possible to evaluate its impacts. This
was done by transecting the sub-watersheds to measure
the extent of adoption of the practices and computing
changes in farm incomes associated with the practices.
Tab le 6 shows an assessment of uptake of BMPs in Il-
langama and Alumbre.
In Illangama, net economic benefits have risen to
about $ 1921 per hectare per year, an increase of about
65% compared to 2006. Improvements have resulted
from incremental increases in yields of potatoes, faba
beans, chocho, barley, quinoa and improved pasture. Soil
management has changed dramatically as ground cover
is more widespread throughout the year. Part of this
change was caused by changes in relative prices; potato
and other crop prices have become increasingly variable
and farmers are moving toward dairy production with
continuous pasture and other more environmentally
suitable crops. Potato net profits have, however, grown
by as much as 50%, due to improved rotations and re-
duced pesticide use (a major cost of production). Use of
late blight-resistant potato varieties, improved soil ferti-
lity and use of better-quality seeds help lower variable
Table 6. Changes in farming practices and uptake of BMPs, 2006-2010.
Micro-watershed BMP 2006 2010
Area under potatoes (ha) 1.02 (100) 0.85 (87)
Area under faba beans (ha) 0.38 (13) 0.45 (48)
Area under chocho (ha) 0.60 (4) 0.50 (35)
Area under barley (ha) 0.59 (5) 0.65 (65)
Area under quinoa (ha) 0 0.15 (40)
Area under natural grass (ha) 2.13 (48) 1.80 (35)
Area under improved pasture (ha) 1.51 (71) 2.25 (85)
Milk production (l/day) 18 50
Potato yields (t/ha) 8.35 13.09
Faba beans yields (t/ha) 0.45 0.95
Chocho yields (t/ha) 0.60 1.15
Barley yields (t/ha) 0.73 1.30
Quinoa yields (t/ha) N/A 1.20
Pesticide used in potato production ($/ha) 289 186
Net benefits ($/year) 1163 1921
Area under white maize (ha) 1.85 (72) 1.85 (85)
Area under yellow maize (ha) 1.18 (5) 1.18 (15)
Area under maize/beans (ha) 4.48 (27) 3.00 (20)
Area under beans (ha) 2.17 (20) 1.50 (60)
White maize yields (t/ha) 0.44 1.10
Yellow maize yields (t/ha) 0.41 0.97
Maize/beans yields (t/ha) 0.57 0.91
Beans yields (t/ha) 0.40 0.88
Net benefits ($/year) 898 1629
Source: INIAP-SANREM CRSP-SENACYT, 2006-2010. Illangama: Percentage of farmers with the crop in each year (sample size: 117 in 2006; 250 in 2010).
lumbre: Percentage of farmers with the crop in each year (sample size: 169 in 2006; 80 in 2010). A
Copyright © 2012 SciRes. OPEN ACCESS
V. H. Barrera et al. / Agricultural Sciences 3 (2012) 768-779
costs. Milk production per land unit grew by 122% due
to improved forages, and better sanitation and feeding
practices. Food security has also improved. Diversified
grain sources broaden the dietary base, reduce risks from
dependence on single crops, and increase energy and
protein intake.
The data indicate impressive trends toward more di-
versified production, with increases in relatively new (to
the area) products such as quinoa. Quinoa production has
emerged, and the crop provides nutrition for home con-
sumption and high prices in the market. As a result of all
these changes, erosion is being reduced and water quality
is improving.
In Alumbre, net benefits from agricultural production
increased by 81% to $1629 per hectare per year in 2010.
This increment was a product of increased yields of
white maize, yellow maize and beans, resulting from
improved management practices. The main engine was
introduction of improved varieties, and more intensive
management concentrated in less vulnerable and more
productive areas. Planting densities have increased and
integrated pest management practices have reduced input
costs. Increased agro-diversity and lower profit risks
(due to fewer purchased inputs) have also increased food
Use of vulnerability maps to guide land use planning
has reduced production on most vulnerable lands and
improved ecosystem services. Indicators of biodiversity,
soil retention and water quality have improved alongside
improvements in agricultural profitability. Farmers now
concentrate productive activities on the most fertile and
least vulnerable lands. Yield improvements and cost re-
ductions allow farmers to earn higher incomes and si-
multaneously improve environmental conditions. Ability
to observe farming practices on the pilot farms has built
confidence in the new practices and they have naturally
spread throughout the watershed. Concurrently, the study
of biodiversity raised consciousness about the value of
native species and led to planting and maintenance of
these potential sources of biodiversity. These actions
have improved environmental conditions and water avail-
Prior to 2006, conservation practices in the area did
not exist. Now, various practices are widely found, such
as improved crop rotations, strip cultivation, deviation
ditches, contour plowing, and use of live barriers. An
indigenous innovation has led to the protection of devia-
tion ditches with various local species. These include
milín grass and native plants such as Quishuar, Yagual,
Chachacoma, Romerillo, Aliso, Pumamaqui, Lupinus,
Piquil. Contour cultivation is also widely practiced now
in both watersheds, irrigation water management has
improved and actions have been taken to protect areas of
water recharge. This protection has involved replanting
many of the native plants metioned above.
Table 7 summarizes results of the 2010 evaluation.
Farmers in the Illangama watershed were more likely to
apply all natural resource management methods, except
for green fertilizer. Differences over time of use of con-
servation methods are statistically significant.
4.5. Participatory Planning
Our team identified local stakeholders, institutions and
government and non-government partners to engage in
participatory planning. Participants identified research
themes and designed research activities and collaborated
in on-farm trials. The process included meetings, work-
shops and information exchanges. Stakeholders immedi-
ately recognized the need for coordinated cross-sectorial
actions and institutional change to increase the value of
natural resources. A regular meeting of a project steering
group was held; the group identified and promoted the
Table 7. Adoption of improved management practices, 2010.
Alumbre Illangama
Method % Using S.E. % Using S.E. p-value
Strip cultivation* 3.77 1.23 21.25 4.60 0.0004
Deviation ditches* 5.02 1.42 31.25 5.21 0.0000
Contour plowing* 6.28 1.57 22.50 4.70 0.0015
Crop rotation* 59.41 3.18 92.50 2.96 0.0000
Live barrier* 24.69 2.79 63.75 5.41 0.0000
Reduced till* 26.36 2.86 76.25 4.79 0.0000
Green fertilizer 7.11 1.67 6.25 2.72 0.7873
No. respondents 239 80
*Denotes statistically significant differences between the watersheds (0.01 level).
Copyright © 2012 SciRes. OPEN ACCESS
V. H. Barrera et al. / Agricultural Sciences 3 (2012) 768-779 777
idea of integrated adaptive management. This group en-
gaged local and Provincial Governments who are full
partners in the process. The Provincial Government cre-
ated a new unit for environmental management and link-
ages across government units has facilitated coordi-
nated actions; our technical team has trained the Gov-
ernment’s technical team and this strategic alliance has
been strengthened over time. The alliance is important
because the Provincial Government bears responsibility
for creating and enforcing the regulatory and legal re-
Our research agenda was arrived at after an arduous
process of building consensus among stakeholders. Pro-
bably the most valuable research output at the start of the
process was to help stakeholders understand and appre-
ciate the value of their natural resources. This new-
found appreciation of value has strengthened incentives
for actions to promote soil retention and health, and to
use native species as a contributor to this conservation.
Native species of trees and bushes have been widely in-
corporated into live barriers to reduce water and wind
erosion, and as lining biomass for deviation ditches. The
team also helped identify a major source of reduced wa-
ter supply and quality: Incursions into the upper páramo
areas. As a result, we have built support for increased
intensification at lower elevations and a sense of com-
munity-wide disapproval for those who exploit the pris-
tine higher-elevation areas of the Illangama. Social pres-
sure is having an effect.
Social capital has been strengthened in many ways.
The participatory planning process is strengthening so-
cial networks in the region. In addition, training in bio-
diversity, natural resource valuation, and natural resource
management has built networks of activists in both mi-
cro-watersheds. Efforts to understand the potential bene-
fits of higher-valued market chains and obstacles to par-
ticipation in them have helped identify how networks of
producers can have more effects than individual actors.
Subsequent efforts to build these networks have also re-
inforced local social capital.
The participatory land-use planning process led to a
functioning watershed planning model. A key component
of success was investment in agricultural and other re-
search to increase incomes. Through this research, we are
creating the economic space to address longer-term pro-
blems associated with natural resource degradation. The
participatory process has built confidence among stake-
holders who now largely buy into our larger program.
Exposure to new technologies to raise incomes has help-
ed build participation.
The nature of the watershed and the variety of stake-
holders pushed us toward casting a large net; we made
major efforts to involve institutions with any presence in
the watersheds. Some of the less-recognized assets of our
program: technical expertise, knowledge of successes
and failures from elsewhere in Ecuador helped build
bridges to institutions that in other cases might be less
receptive to innovative ideas. For example, our ability to
provide training for units of local governments helped
legitimize our presence in the eyes of this important
Our adaptive watershed management project is less a
political process and more a process of social learning
and empowering community actions. A key barrier to
effective local action was the Balkanization that pre-
dominated prior to our project: Different actors and
stakeholders did not communicate and were even less
likely to undertake coordinated actions. By casting a
wide net and strongly encouraging participation, the pro-
ject broke many of the barriers to collective action. Our
training and participatory research efforts helped bring
down these barriers and created a common consensual
set of knowledge about actions. These programs brought
down barriers and increased the capability to effect
Our most important lesson was the necessity to build
consensus and engage stakeholder groups. The effort
required to reach this point was substantial and involved
tireless exercises in outreach, networking and stake-
holder engagement. This process is long and one that
may not show immediate results. However, the impact it
eventually created was worth the effort. Communities in
the watersheds now actively participate.
A first recommendation is to continue participatory
consensus building. Efforts to engage community mem-
bers in field experiments, in water quality data collection,
and in the biodiversity assessment were especially help-
ful in gaining local ownership and building credibility
for the entire project. We began with small steps and a
massive amount of participation, and built in complexity
as the process evolved. Wide participation also lowers
labor costs associated with project activities.
A second recommendation is to incorporate a multi-
sectorial approach. The primary objective of the project
was to create sustainable means of natural resource
management through adaptive watershed management.
However, we documented the importance of research for
more profitable technologies, for increased returns to
producers through higher-valued chains (such as dairy
production in the Illangama watershed), and for en-
hanced social and institutional capital. All these actions
Copyright © 2012 SciRes. OPEN ACCESS
V. H. Barrera et al. / Agricultural Sciences 3 (2012) 768-779
created space for and consensus about the need to con-
serve natural resources. Without them, the approach
would not have been successful. We are still seeking a
means of increasing farmer capture of value from off-
farm benefits to on-farm investments (reduced erosion,
enhanced water quality, enhanced biodiversity).
A final recommendation is that researchers should take
risks. We began the project with skepticism about the
adaptive management approach. We were concerned
about the ability to generate data and make use of high-
tech tools such as GIS and SWAT. We wondered whether
the community would accept research results from exotic
tools. We found that although the data base is still in-
adequate and the SWAT results do not reflect reality as
well as they might, the results are being used. GIS is
probably the most effective tool in our arsenal for pre-
senting results to stakeholders; they understand and can
effectively interpret the maps we produce. These tools,
while exotic, are not beyond the reach of a modest pro-
gram in highland Ecuador.
This project was part of the SANREM CRSP, supported by the
United States Agency for International Development and the generous
support of the American people through Cooperative Agreement No.
EPP-A-00-04-00013-00. We thank the people of the communities of the
Alumbre and Illangama sub-watersheds for their participation in the
[1] Doolette, J. and McGrath, W. (1990) Strategic issues in
watershed development. In: Doolette, J. and McGrath, W.,
Eds., Watershed Development in Asia, World Bank Tech-
nical Paper No. 127, Washington DC.
[2] Guerra-García, G. and Sample, K. (2007) Policy and pov-
erty in Andean countries. International IDEA, Stockholm.
[3] Siegel, P. and Alwang, J. (2004) An asset-based approach
to social risk management: A conceptual framework. So-
cial Protection Discussion Paper 9926. Social Protection
Unit, Human Development Network, World Bank, Wash-
ington DC.
[4] Dourojeanni, A. and Jouravlev, A. (2001) Crisis in govern-
ability and water management: Challenges to implementa-
tion of agreed-upon recommendations. Economic commis-
sion for Latin America and the Caribbean (CEPAL). Series
on Natural Resources and Infrastructure, Santiago.
[5] Scherr, S. and McNeely, J. (2004) Reconciling agriculture
and wild biodiversity conservation: Policy and research
challenges. In: Conservation and Sustainable Use of Ag-
ricultural Biodiversity: A Sourcebook, CIP-UPWARD,
Lima, 46-55.
[6] Scherr, S. and Downward, A. (2000) “Spiral? Recent
evidence on the relationship between poverty and natural
resource degradation. Food Policy, 5, 479-498.
[7] De Marco, J. and Monteiro Coelho, F. (2004) Services
performed by the ecosystem: Forest remnants influence
agricultural cultures’ pollination and production. Biodi-
versity and Conservation, 13, 1245-1255.
[8] Barrera, V., León-Velarde, C., Grijalva, J. and Chamorro,
F. (2004) Management of the potato-pasture production
system in Andean Ecuador: Technology options. Editorial
ABYA, Technical Bulletin INIAP-CIP-PROMSA, Quito,
196 p.
[9] SIGAGRO (2008) Information on micro-watersheds in
Alumbre and Illangama, Ecuador. Geographical Informa-
tion System, Quito.
[10] INIAP (2008) Geographical Information System for the
Chimbo River, Bolivar-Ecuador. Climate Monitoring. INIAP,
[11] Barrera, V., Alwang, J. and Cruz, E. (2008) Integrated
management of natural resources for small-scale agriculture
in the Chimbo river, Ecuador-lessons learned. INIAP
SENACYT Information Bulletin No. 339, Quito.
[12] INEC-MAG (2002) III National Census of Agriculture:
Resultados nacionales, provinciales y cantonales, instituto
nacional de estadísticas y censos y ministerio de agri-
cultura y ganadería. Quito.
[13] GPB (2004) Strategic plan for development, provincial
government of Bolivar. Planning office, Guaranda, 224 p.
[14] Gallardo, G. (2000) Final report for the integrated man-
agement of natural resources in Watersheds. Quito, 220 p.
[15] Salafsky, N., Margoluis, R. and Redford, K. (2001) Adap-
tive management: A tool for conservation practitiones.
Biodiversity Support Program, Washington DC.
[16] Chambers, R. and Conway, G. (1992) Sustainable rural
livelihoods: Practical concepts for the 21st century. IDS
Discussion Paper 296. Institute for Development Studies,
[17] Aldenderfer, M. and Blashfield, R. (1984) Cluster analy-
sis; Series: Quantitative applications in the social science.
SAGE University Paper, Beverly Hills.
[18] Ward, H. (1963) Hierarchical grouping to optimize and
objective function. Journal of the American Statistical
Association, 58, 236-244.
[19] Alwang, J., Barrera, V., Andrade, R., Hamilton, S. and
Norton, G. (2009) Adaptive watershed management in the
South American highlands: Learning and teaching on the
Fly. In: SWCS, Ed., The Sciences and Art of Adaptive
Management: Innovating for Sustainable Agriculture and
Natural Resource Management. Ankenny, 209-227.
[20] Calles, J. (2007) Bioindicadores terrestres y acuáticos
para las microcuencas de los ríos Illangama y Alumbre,
provincia Bolívar. EcoCiencia, Quito, 30 p.
[21] Cruz, E. (2009) Study of biodiversity in microwater- sheds
of Illangama and Alumbre, Ecuador. INIAP, Quito, 26 p.
Copyright © 2012 SciRes. OPEN ACCESS
V. H. Barrera et al. / Agricultural Sciences 3 (2012) 768-779
Copyright © 2012 SciRes. OPEN ACCESS
[22] Andrade, R. (2008) Household assets, Livelihood deci-
sions and well-being in chimbo ecuador. Master of Sci-
ence Thesis, Department of Agriculture and Applied
Economics, Virginia Tec