Full text of DOI:10.1080/02508060108686932

This article was downloaded by: [McMaster University]
On: 05 November 2014, At: 07:03
Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK
Water International
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/rwin20
Water Use and Protection in Rural Communities of the
Peruvian Amazon Basin
Michael E. McClain a , Luis Miguel Aparicio b & Carlos A. Llerena b
a IWRA, Florida International University , Miami, Florida, USA
b Universidad Nacional Agraria La Molina , Lima, Peru
Published online: 22 Jan 2009.
To cite this article: Michael E. McClain , Luis Miguel Aparicio & Carlos A. Llerena (2001) Water Use and Protection in Rural
Communities of the Peruvian Amazon Basin, Water International, 26:3, 400-410, DOI: 10.1080/02508060108686932
To link to this article: http://dx.doi.org/10.1080/02508060108686932
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

International Water Resources Association
Water International, Volume 26, Number 3, Pages 400–410, September 2001
Water Use and Protection in Rural Communities
of the Peruvian Amazon Basin
Michael E. McClain, Member, IWRA, Florida International University, Miami
Florida,USA, Luis Miguel Aparicio and Carlos A. Llerena, Universidad Nacional
Agraria La Molina, Lima Peru
Abstract: Inhabitants of the Peruvian Amazon enjoy plentiful water and other aquatic resources
that enhance their well-being in many ways. However, the intimate and generally unbuffered connection
between the region’s inhabitants and their water resources leads to complex negative feedbacks when
these resources are mismanaged. Due to severe water problems in other, more populated, parts of the
country the Peruvian government currently devotes little attention to water management in the Amazon.
Thus, organized management is mainly left to individual communities and households. This study re-
ports on the results of 351 interviews of households in the Pachitea Basin of the central Peruvian
Amazon. Our aim is to quantify the use of water and other aquatic resources among different social
groups and within different geographical settings of the region. With these data, we evaluate and
identify priorities for community-driven water management in the region. We found that 50 to 90 percent
of households take their water directly from primary sources, 35 to 94 percent transport it manually to
their homes, and 50 to 75 percent practice only the simplest form of treatment (boiling). Indigenous
households tended to rely less on water infrastructure and water treatment. Fish and an assortment of
other aquatic and riparian resources were important inputs to all social groups and in all geographic
settings. Disposal of wastes in nearby water bodies was also found to be widespread. We conclude that
water management efforts in the Pachitea Basin should focus on the protection of water quality in rivers
and streams through careful disposal of wastes away from water bodies and the preservation of natural
water purification features such as riparian forests and wetlands. We also recommend developing a
basin-wide master plan, which empowers end-users and integrates more detailed plans developed at
community and association levels.
Keywords: Water use, Peruvian Amazon, community water management, preservation.
and amoebic dysentery are transmitted to river waters
Introduction
with animal and human wastes, and insect vectors for
Water is abundant in the Peruvian Amazon and con- malaria and dengue fever thrive in the habitat that abounds
tributes to rural people’s physical, social, and economic in these humid landscapes.
well-being in multifaceted ways. When traveling through
Despite the multifaceted importance of water in the
the region, one notes that surface and ground waters sat- Peruvian Amazon, water resources management in the
isfy basic needs for human consumption, cooking, wash- region receives little attention from national agencies and
ing, waste removal, and rain waters crops. Likewise, policy makers. This inattention is in fact systemic and may
personal observations and a limited amount of previous be attributed to a geographical mismatch of people and
research show that rivers and lakes are sources of fish, water in the country. More than 90 percent of Peru’s popu-
crustaceans, and plants that contribute significantly to the lation (as well as its capital and virtually all of its large
dietary needs of the population. Local people also harvest cities) exists along the arid Pacific coast and in the high
riparian plants and hunt semi-aquatic animals and water- Andean cordillera where only two percent of the country’s
seeking animals along riverbanks. Rivers transport prod- water resources occur (INRENA, 1996). Consequently,
ucts to and from rural communities in roadless areas, they water shortages and conflicts over equitable water distri-
serve as recreational areas, and they often take on spiri- bution for irrigation and other uses dominate the psyche
tual importance, especially within indigenous groups. These of water managers and policy makers (McClain and
close and often unbuffered linkages between people and Llerena, 1998). Efforts to address these issues also con-
surrounding water resources, furthermore, have major im- sume the vast majority of funds, expertise, and labor allo-
plications for human health. Pathogens linked to cholera cated for the implementation of sound water management
400

Water Use and Protection in Rural Communities of the Peruvian Amazon Basin
401
plans and activities. Formal water management in Peru is
organized according to irrigation districts, wherein techni-
cal administrations interact with user groups to develop
and implement water-use plans. In the Amazon portion of
the country, however, irrigation districts have not been
defined and thus no coordinating bodies exist to manage
the use and protection of water resources. The task of
managing water and aquatic resources in rural portions of
the Peruvian Amazon is, therefore, left in the hands of
local communities and individual users.
In this paper, we present the results of the first house-
hold survey of water use patterns in rural portions of the
central Peruvian Amazon. Our goal is to provide basic
information and preliminary interpretations required to sup-
port emerging community-driven plans for water manage-
ment. This study was also carried out in conjunction with
an ongoing research program to investigate the hydrology
and water quality of the basin. Effort was made to distin-
guish between water-use patterns among indigenous com-
munities, colonist farmers, and townspeople, as well as
between people living in different geographical settings in
the basin.
Figure 1. Physiographic map of the Pachitea Basin indicating the
location of the basin within Peru and its main mountain ranges, val-
leys, and rivers.
Study Area
The Pachitea watershed lies in the Departments of
Pasco and Huanuco of the Selva Central of Peru and oc-
Census data from 1996 place the population of the
cupies an area of approximately 29,000 km2. The water- Pachitea at 145,966 individuals distributed among 26,876
shed consists of broad undulating lowlands covering households (INEI, 1996). This translates to a population
approximately 19,000 km2 and a series of montane and density of 4.8 people/km2 in the area of the census, which
high alpine valleys covering approximately 10,000 km2. is slightly larger than the area of the watershed. Popula-
These two regions differ in natural characteristics such as tion density is much higher (9.4 people/km2), however, in
ecosystems, topography, and climate, and they also differ the montane and alpine portion of the watershed where 65
in socio-economic and demographic terms.
percent of the population lives on 31 percent of the land.
According to estimates of the Peruvian National In- The majority of both the lowland (86 percent) and moun-
stitute for Natural Resources (INRENA, 1996), precipita- tain (79 percent) population is rural, and agriculture is the
tion across the entire watershed averages 2,809 mm/yr, of main economic activity of between 73 percent (mountain)
which 2,153 mm/yr (or 77 percent) runs off in the Pachitea and 80 percent (lowland) of the population 15-years-old
river. Precipitation tends to be highest in the lowland por- or older. The most common crops are manioc, bananas,
tion of the basin near the foot of the mountains, and pre- corn, and rice (Aparicio, 1999).
cipitation generally decreases with increasing elevation
Yanesha’ and Asháninca indigenous groups live in dis-
(SENAMHI, 1999). Elevations range from nearly 5,000 persed communities in the basin and make their livelihood
meters above sea level (masl) in the headwaters along the through farming, cattle raising, and occasional logging. The
Cordillera Oriental to slightly less than 200 masl at the Yanesha’ are small in number and live mainly within a
mouth of the Pachitea river. Three other smaller mountain communal reserve in the Palcazu valley. The Asháninca
ranges also divide the watershed into distinct subbasins are larger in number and occupy a larger area within the
(Figure 1). The strong elevational gradient of the water- lowland Pichis valley. The earliest colonists to enter the
shed translates to a similarly strong gradient in natural watershed were Austro-Germans who arrived in the 1850s
ecosystems. Eleven ecological life zones of the Holdridge and settled primarily in mountain valleys. Mestizo colo-
classification system occur in the region, ranging from very nists have been growing in numbers in the watershed since
humid tropical forest to humid steppe (ONERN, 1976). then, but the greatest increase in colonist numbers came
Humid tropical forest dominates the lowland portions of during the second half of the 20th century (Smith, 1983).
the watershed but grades quickly into humid montane for- In 1980, the Peruvian government announced an ambi-
est and cloud forest as elevations surpass 1,000 m. In the tious program of road building and colonization in the re-
highest portions of the basin, alpine steppe, known locally gion. Over the following decade, roads were constructed
as puna, predominates.
into the Palcazu valley and along the main channel of the
IWRA, Water International, Volume 26, Number 3, September 2001

402
M.E. McClain, L.M. Aparicio, and C.A. Llerena
Pachitea river, and the colonist population has grown in
Interviews were structured to distinguish water use
response. Road construction is continuing in the region, and protection between different social groups and be-
and the number of colonists is likely to continue to grow as tween inhabitants living in different geographic settings of
well.
the region. The three social groups considered were: colo-
nists living in towns (132 interviews), colonists living on
farms (122 interviews), and indigenous inhabitants (97 in-
terviews) living within villages or farms. Respondents were
Methodology
In order to better understand the relationship of the distinguished geographically according to whether they lived
region’s population to its water resources, water use and in the lower, mainly flat portion of the watershed or in the
related data were gathered through interviews of 351 higher montane valleys. In addition to questions address-
households. Each interview consisted of an identical set ing water use, a number of questions also addressed the
of 35 individual or compound questions printed on a sur- use of aquatic resources and land use practices which are
vey form, which was filled in by interviewers. Two inter- relevant to water quality.
viewers moved through the region simultaneously and met Data are reported as proportions of the total responses
each day to discuss any questions or problems that arose according to social class or geographic region. The mar-
during the day and to ensure that the interview content gin of error (ME ) of each proportion was calculated as
remained consistent. All 351 interviews were conducted follows
p
by these same two interviewers. Severe logistical chal-
lenges imposed by the size and remoteness of the region
surveyed, the dispersed locations of households, and a lack
of reliable maps of human distribution prohibited a rigor-
ous randomization of sampling sites. Interviews were con-
π ꢂꢁ−π ꢀ
ꢀ
ꢁꢂ5 =ꢁꢅꢃꢄ
where π is the proportion and N is the number of re-
ducted in eight towns, along four sections of road, five sponses. The upper and lower 95 percent confidence val-
reaches of river, and in three indigenous communities (Fig- ues may be calculated as [π + π(MEp)] and [π - π(MEp)],
ure 2). Interviews were conducted continuously over a respectively.
period of 30 days and at all times of the day.
Results
Water Use
The 351 households surveyed have an average family
size of 5.4 individuals, with little difference between social
groups or geographic regions (Table 1). Slightly smaller
average family sizes were encountered among the towns-
people interviewed (4.9), while slightly larger average fam-
ily sizes were encountered among indigenous people
interviewed (5.8). Households use an average of 82 liters
of water per day for drinking, cooking, and washing (Table
1). This amounts to 15 liters per person per day. Indig-
enous households use, on average, less water (73 l/d) than
colonist and townspeople households (89 and 84 l/d, re-
spectively). Likewise, households of the montane valleys
use, on average, less water (65 l/d) than lowland house-
holds (95 l/d).
Indigenous households take most of their water for
human consumption from streams (44.3 percent) and
springs (39.2 percent), while colonist households take most
of their water for human consumption from streams (39.3
percent) and rivers (30.3 percent) (Table 2). The largest
proportion of townspeople take their water from the pub-
lic supply system (49.2 percent), but other sources are
also important. Colonist households take more water for
human consumption from rivers than other social groups,
and townspeople make the most use of groundwater (18.1
percent) (Table 2). Outside of towns, indigenous house-
holds (9.3 percent) make more use of groundwater than
colonist households (4.9 percent). Households in the low-
Figure 2. Map of the sites where interviews where made, including
eight towns (with names shown), three indigenous communities, four
sections of road, and five reaches of river.
IWRA, Water International, Volume 26, Number 3, September 2001

Water Use and Protection in Rural Communities of the Peruvian Amazon Basin
403
Table 1. Household Water Use
Indigenous
Colonist
Townspeople
Montane
Lowland
# Family Members
lt/day/family
5.8
5.6
4.9
5.2
5.6
72.7
88.7
83.4
64.8
95.0
Reported as average of total responses.
Table 2. Water for Human Consumption
Source of Water for Human Consumption
Indigenous
Colonist
Townspeople
Montane
Lowland
River
Lake
Stream
Spring
Rain
Public Supply
Public Spigot
Well
19.6(4.2)
1(1.0)
30.3(4.8)
0(0.0)
39.3(5.1)
22.9(4.4)
10.7(3.2)
18(4.0)
0.8(0.9)
4.9(2.3)
0.8(0.9)
6.8(2.6)
0(0.0)
15.5(3.8)
0.0(0.0)
21.9(4.3)
25.6(4.6)
16.6(3.9)
47.1(5.2)
1.6(1.3)
2.7(1.7)
1.1(1.1)
22.0(4.3)
0.6(0.8)
38.4(5.1)
22.6(4.4)
6.7(2.6)
4.3(2.1)
10.4(3.2)
20.7(4.2)
1.2(1.1)
44.3(5.2)
39.2(5.1)
11.3(3.3)
8.2(2.9)
1(1.0)
9.8(3.1)
14.4(3.7)
13.6(3.6)
49.2(5.2)
13.6(3.6)
18.2(4.0)
2.3(1.6)
9.3(3.0)
0(0.0)
Other
Means of Delivering Water to Household
Pipe
Open Canal
Manual
13.4(3.6)
9.3(3.0)
82.5(4.0)
34.4(5.0)
3.3(1.9)
62.3(5.1)
55.3(5.2)
0(0.0)
48.5(5.2)
60.9(5.1)
5.3(2.3)
35.3(5.0)
8.6(2.9)
1.8(1.4)
93.9(2.5)
Reported as percentages of total responses/margin of error in parentheses.
land portions of the watershed make greater use of river These values are very near the number of total house-
and stream water (22.0 and 38.4 percent, respectively), holds that practice agriculture (Aparicio, 1999). Thus, rain
while households of the montane valleys make greater use accounts for nearly 100 percent of water applied to agri-
of rainwater and water from public supply systems (16.6 cultural fields in the watershed. Irrigation is nearly non-
and 47.1 percent, respectively). Lowland households also existent in the portions of the watershed surveyed. The
make greater use of groundwater (20.7 percent) than do majority of households that have animals responded that
montane valley households (2.7 percent) [Table 2].
their animals drink from rivers and streams (Table 4).
In the lowland portion of the watershed, greater than Animals from a significant number of households also drink
90 percent of households manually collect water for hu- from springs. In the montane valleys 4.8 percent of house-
man consumption, while only 35 percent of montane val- holds responded that their animals drink water from the
ley households manually collect water. When viewed across public supply system and 5.5 percent of lowland house-
social groups, indigenous households (82.5 percent) rely holds responded that their animals drink water taken from
most on manual delivery of water for human consumption wells (Table 4).
while townspeople households (48.5 percent) rely least on
Rivers were the main water body used for transporta-
manual delivery. Slightly greater than 60 percent of mon- tion in the watershed (Table 5). Indigenous households
tane households received water for household use via pipes, make the greatest use of rivers for transportation (54.6
while less than 10 percent of lowland households had wa- percent), and transportation of people in particular (59.8
ter piped into them (Table 2).
percent). In addition, 37.1 percent of indigenous house-
The majority of households in the region apply some holds interviewed transport products by river, and 17.5
sort of treatment to water prior to consuming it (Table 3). percent transport wood. Rivers were less important for
Nearly 87 percent of households in montane valleys boil transportation in montane valleys, and this was particu-
their water, while approximately 70 percent boil their wa- larly true in the higher Huancabamba valley (Aparicio,
ter in the lowlands. Townspeople households are most dili- 1999). Colonists and townspeople make slightly less use
gent in that slightly more than 90 percent treat water prior of water for transportation, but the percentage is still no-
to consuming it. In contrast, only 70 percent of indigenous table (Table 5). Rivers and streams are also important sites
households treat their water prior to consuming it.
of recreation for families of the watershed. Greater than
Greater than 80 percent of indigenous and colonist 50 percent of all households across all social groups and
households and 34.8 percent of townspeople households geographic regions make use of rivers for swimming (Table
rely on rain to water their agricultural fields (Table 4). 5). A significant percentage also use streams as recre-
IWRA, Water International, Volume 26, Number 3, September 2001

404
M.E. McClain, L.M. Aparicio, and C.A. Llerena
Table 3. Water Treatment Prior to Consumption
Indigenous
Colonist
Townspeople
Montane
Lowland
Decanting
Filtration
Boiling
Clorine
Other Chemical System
Other Natural System
No Treatment
5.1 (2.3)
0 (0.0)
70.1 (4.8)
13.4 (3.6)
6.2 (2.5)
0 (0.0)
6.6 (2.6)
0 (0.0)
82 (4.0)
7.4 (2.7)
3.3 (1.9)
0 (0.0)
5.3 (2.3)
1.5 (1.3)
83.3 (3.9)
15.1 (3.7)
0.8 (0.9)
1.5 (1.3)
9.1 (3.0)
2.7 (1.7)
0.5 (0.7)
86.9 (3.5)
8.0 (2.8)
2.1 (1.5)
0.0 (0.0)
14.4 (3.7)
9.2 (3.0)
0.6 (0.8)
70.1 (4.8)
16.5 (3.9)
4.2 (2.1)
1.2 (1.1)
22.0 (4.3)
30.9 (4.8)
17.2 (3.9)
Reported as percentages of total responses/margin of error in parentheses.
Table 4. Source of Water for Agriculture and Animals
Agriculture
Indigenous
Colonist
Townspeople
Montane
Lowland
River
Lake
Stream
Spring
Rain
Public Supply
Public Spigot
Well
0 (0.0)
0 (0.0)
2.1 (1.5)
1 (1.0)
83.5 (3.9)
0 (0.0)
0 (0.0)
3.3 (1.9)
0 (0.0)
0.8 (0.9)
3.3 (1.9)
82.0 (4.0)
0 (0.0)
0 (0.0)
0.8 (0.9)
0.8 (0.9)
0 (0.0)
0 (0.0)
3.8 (2.0)
1 (1.0)
34.8 (5.0)
1.5 (1.3)
0 (0.0)
0.8 (0.9)
0.8 (0.9)
0.5 (0.7)
0.0 (0.0)
2.7 (1.7)
2.2 (1.5)
59.3 (5.1)
1.0 (1.1)
0.0 (0.0)
0.0 (0.0)
0.5 (0.7)
1.8 (1.4)
0.0 (0.0)
1.8 (1.4)
1.8 (1.4)
70.7 (4.8)
0.0 (0.0)
0.0 (0.0)
1.2 (1.1)
0.6 (0.8)
0 (0.0)
0 (0.0)
Other Animals
River
Lake
Stream
Spring
Rain
Public Supply
Public Spigot
Well
19.6 (4.2)
1 (1.0)
36.1 (5.0)
13.4 (3.6)
3.1 (1.8)
1 (1.0)
25.4 (4.6)
2.5 (1.6)
45.9 (5.2)
9.8 (3.1)
3.3 (1.9)
1.6 (1.3)
0 (0.0)
12.1 (3.4)
1.5 (1.3)
18.2 (4.0)
6.1 (2.5)
2.3 (1.6)
4.5 (2.2)
1.5 (1.3)
5.3 (2.3)
1.5 (1.3)
16.8 (3.9)
1.6 (1.3)
32.1 (4.9)
9.1 (3.0)
0.5 (0.8)
4.8 (2.2)
0.5 (0.8)
0.0 (0.0)
3.7 (2.0)
20.7 (4.2)
1.8 (1.4)
33.5 (4.9)
9.8 (3.1)
5.5 (2.4)
0.0 (0.0)
0.6 (0.8)
5.5 (2.4)
0.0 (0.0)
0 (0.0)
1 (1.0)
0 (0.0)
0.8 (0.9)
4.1 (2.1)
Other
Reported as percentages of total responses/margin of error in parentheses.
Table 5. Use of Water for Transportation and Recreation
Transportation Media
Indigenous
Colonist
Townspeople
Montane
Lowland
River
Lake
54.6 (5.2)
0 (0.0)
3.1 (1.8)
37.7 (5.1)
0 (0.0)
1.6 (1.3)
42.4 (5.2)
0 (0.0)
1.5 (1.3)
27.8 (4.7)
0.0 (0.0)
1.1 (1.1)
62.8 (5.1)
0.0 (0.0)
3.0 (1.8)
Streams
Material Transported
People
59.8 (5.1)
37.1 (5.1)
17.5 (4.0)
41.8 (5.2)
28.7 (4.7)
10.7 (3.2)
48.5 (5.2)
17.4 (4.0)
9.8 (3.1)
31.5 (4.9)
15.0 (3.7)
9.0 (3.0)
69.5 (4.8)
40.2 (5.1)
15.9 (3.8)
Products
Wood
Recreation
River
Lake
Stream
Springs
61.9 (5.1)
0 (0.0)
21.6 (4.3)
2.1 (1.5)
59 (5.1)
0 (0.0)
16.4 (3.9)
0.8 (0.9)
61.4 (5.1)
0 (0.0)
9.1 (3.0)
4.5 (2.2)
66.3 (4.9)
0.0 (0.0)
5.9 (2.5)
0.5 (0.8)
54.3 (5.2)
0.0 (0.0)
25.6 (4.6)
4.9 (2.3)
Reported as percentages of total responses/margin of error in parentheses.
IWRA, Water International, Volume 26, Number 3, September 2001

Water Use and Protection in Rural Communities of the Peruvian Amazon Basin
405
ation sites. Households were asked about the use of wa- lowland households (87.2 percent) than in montane valley
ter for industrial purposes and electricity generation, but households (80.2 percent). Fishing is also a common ac-
no such use was made of water by households directly.
tivity in most households (Table 8). The percentage of in-
Indigenous households make the greatest use of wa- digenous households that fish (84.5 percent) is nearly equal
ter bodies for disposing of wastes, with 22.3 percent and to the percentage that eat fish (88.7 percent). The small-
14.4 percent responding that they dumped their wastes in est proportion of households that fish are those of towns-
rivers and streams, respectively (Table 6). Townspeople people (47 percent) and those in montane valleys (55.6
households generally make less use of water bodies for percent). In line with other statistics in this section, indig-
dumping, but 21.2 percent reported that they dump wastes enous households also consume the larger quantity of fish
into rivers. Little difference exists between the percent- per week (4.8 kg/wk). Overall, the quantity of fish con-
age of lowland versus montane valley households dump- sumed in lowlands (3.4 kg/wk) is similar to that consumed
ing wastes into water bodies.
in montane valleys (3.0 kg/wk), however, in the discussion
A majority of households have some infrastructure, it will be shown that the source of the fish is quite differ-
normally a latrine, for the disposal of human fecal waste ent. Rivers are the major source of fish caught by house-
(Table 7). Townspeople households had the largest pro- holds of the regions, followed by streams (Table 8). The
portion (96.2 percent) of such infrastructure, while colo- most common fish species consumed were boquichico
nist households had the smallest (73.8 percent.). No (Prochilodus cf nigricans), lisa (Schizodon fasciatum),
difference was noted between geographic regions in the and zungaro (Zungaro zungaro), and 55.7 percent of in-
proportion of households with latrines. Where latrines ex- digenous households noted a decrease in the abundance
ist, they lie an average of 58.3 meters from the house- of fish in recent years. Hook and line, followed by cast net
holds’ source of water (Table 7). On average, indigenous and gill net, is the most common method of fishing (Table
households position their latrines furthest from water 9). Among indigenous households harpooning and use of
sources (74 m). Only minor differences were noted be- barbasco (a natural poison) are also significant. Use of
tween geographic regions in the distance of latrines from explosives was reported infrequently.
water sources.
Households also acquire other resources from water
bodies, including aquatic plants, crustaceans, and semi-
aquatic animals that are hunted. Other animals and plants
Use of Other Aquatic Resources
Rivers, lakes, and streams are used for more than are collected in close proximity to water bodies, either
water supply, transportation, recreation, and waste disposal. because they are attracted there by the water, as in the
Between 78 percent and 89 percent of all households sur- case of animals, or because they grow in riparian habitats,
veyed eat fish regularly (Table 8). Consumption is great- as in the case of plants. Indigenous households make the
est in indigenous households; it is also slightly greater in greatest use of these resources, with 17.5 percent and
Table 6. Disposal of Wastes in Water Bodies
Indigenous
Colonist
Townspeople
Montane
Lowland
Percent of households disposing
of wastes into water bodies
River
Lake
Stream
Spring
Well
38.1 (5.1)
22.3 (4.4)
0 (0.0)
14.4 (3.7)
0 (0.0)
0 (0.0)
31.1 (4.8)
18.8 (4.1)
0 (0.0)
13.1 (3.5)
0.8 (0.9)
0 (0.0)
27.3 (4.7)
21.2 (4.3)
0.8 (0.9)
6.8 (2.6)
0 (0.0)
30.0 (4.8)
20.9 (4.3)
0.0 (0.0)
11.7 (3.4)
0.0 (0.0)
0.0 (0.0)
33.5 (4.9)
20.7 (4.2)
0.6 (0.8)
10.4 (3.2)
0.6 (0.8)
0.0 (0.0)
0 (0.0)
25 (4.5)
Other
25 (4.5)
20.5 (4.2)
24.6 (4.5)
25.6 (4.6)
Reported as percentages of total responses/margin of error in parentheses.
Table 7. Use of Systems to Evacuate Human Fecal Waste
Indigenous
Colonist
Townspeople
Montane
Lowland
Yes - use some system
No - no system used
Dist. (m) of latrine from
water source*
82.5 (4.0)
17.5 (4.0)
74
73.8 (4.6)
26.2 (4.6)
65.7
96.2 (2.0)
3.8 (2.0)
40.2
84.5 (3.8)
15.5 (3.8)
53.0
84.7 (3.8)
15.( 93.8)
60.4
Reported as percentages of total responses/margin of error in parentheses.
*Reported as average of total responses.
IWRA, Water International, Volume 26, Number 3, September 2001

406
M.E. McClain, L.M. Aparicio, and C.A. Llerena
Table 8. Consumption of Fish and Other Aquatic Flora and Fauna
Fish
Indigenous
Colonist
Townspeople
Montane
Lowland
Percent of households that eat fish 88.7 (3.3)
84.5 (3.8)
4.8
85.2 (3.7)
65.6 (5.0)
2.8
78 (4.3)
47 (5.2)
2.7
80.2 (4.2)
55.6 (5.2)
3.0
87.2 (3.5)
73.2 (4.6)
3.4
Percent of households that fish
Kg fish per week consumed *
River
Lake
Stream
Flora and Fauna
68 (4.9)
8.2 (2.9)
38.1 (5.1)
55.7 (5.2)
2.5 (1.6)
18 (4.0)
40.9 (5.1)
3 (1.8)
5.3 (2.3)
48.6 (5.2)
1.1 (1.1)
11.2 (3.3)
59.1 (5.1)
8.0 (2.8)
27.4 (4.7)
River
Lake
Stream
10.3 (3.2)
6.2 (2.5)
17.5 (4.0)
8.2 (2.9)
2.5 (1.6)
7.4 (2.7)
3 (1.8)
0 (0.0)
2.3 (1.6)
4.3 (2.1)
1.1 (1.1)
4.2 (2.1)
9.8 (3.1)
4.3 (2.1)
12.8 (3.5)
Reported as percentages of total responses/margin of error in parentheses.
*Reported as average of total responses/standard deviation in parentheses.
Table 9. Fishing Techniques
Indigenous
Colonist
Townspeople
Montane
Lowland
Harpoon
Gill net
Cast net
Bow and arrow
Hook and line
Explosives
Barbasco
Other poison
Closing off stream
Other
16.5 (3.9)
38.1 (5.1)
49.5 (5.2)
24.7 (4.5)
75.3 (4.5)
7.2 (2.7)
16.5 (3.9)
8.2 (2.9)
3.1 (1.8)
5.1 (2.3)
2.5 (1.6)
25.4 (4.6)
34.4 (5.0)
0.8 (0.9)
56.6 (5.2)
4.9 (2.3)
1.6 (1.3)
0.8 (0.9)
1.6 (1.3)
1.6 (1.3)
0.8 (0.9)
19.7 (4.2)
15.9 (3.8)
0.8 (0.9)
38.6 (5.1)
5.3 (2.3)
0.8 (0.9)
0 (0.0)
3.2 (1.8)
12.3 (3.4)
41.5 (5.2)
2.7 (1.7)
46.0 (5.2)
5.9 (2.5)
4.8 (2.2)
1.1 (1.1)
1.6 (1.3)
2.7 (1.7)
8.5 (2.9)
43.3 (5.2)
20.1 (4.2)
12.8 (3.5)
65.3 (5.0)
5.5 (2.4)
6.1 (2.5)
4.3 (2.1)
1.2 (1.1)
1.8 (1.4)
0 (0.0)
0.8 (0.9)
Reported as percentages of total responses/margin of error in parentheses.
10.3 percent responding that they collect flora and fauna ing water quality, especially water quality linked to organic
from streams and rivers, respectively. Lowland households wastes. In many instances, the source of a household’s
make more use of these resources than montane valley water for domestic use is also a depository for its solid
households (Table 8).
and liquid organic wastes. Even those households that care-
fully dispose of their wastes away from water sources
are vulnerable to contamination introduced by less respon-
sible upstream neighbors or by animals using the same
Discussion
Inhabitants of the Pachitea watershed live in intimate water sources. Households located downstream of popu-
contact with their water resources and exert direct con- lation centers are especially at risk, as towns with sewer
trol over water management and protection. As data gath- systems deposit their untreated wastes directly into nearby
ered in this study have shown quantitatively, a majority of rivers.
households take their water directly from primary sources
In the absence of any wastewater treatment infra-
(springs, streams, and rivers), transport it manually to their structure, water users in the Pachitea must rely on dilution
homes, and make use of it after only the simplest form of to lower contaminant concentrations and on natural pro-
treatment (boiling). These same sources of water provide cesses to assimilate organic wastes. Health data from the
fish and an assortment of other aquatic and riparian re- region and a small set of water quality analyses suggest,
sources. Rivers and streams are also the depositories of however, that the capacity of these processes to effec-
waste for a significant proportion of the population. Per- tively maintain water quality is being surpassed. Aparicio
sonal and/or product transportation for a significant pro- (1999) reported that greater than 50 percent of house-
portion of the inhabitants is via river, and many inhabitants holds interviewed complained of health problems linked to
swim and seek relief from the heat of the day in rivers and water. Likewise, hospitals operated in the towns of Puerto
streams.
Inca and Oxapampa recorded a total of 8,125 cases of
These multiple and direct contacts between people illness related to water between January and October 1998
and water resources hold several implications for protect- (Aparicio, 1999). Acute diarrhea was the most commonly
IWRA, Water International, Volume 26, Number 3, September 2001

Water Use and Protection in Rural Communities of the Peruvian Amazon Basin
407
reported illness (4,172 cases) along with “other” helminchic clear away the floodplain forest to make way for seasonal
infections (intestinal worms – 3,578 cases), but 62 cases crops (Salo and Kalliola, 1993). Thus, the destruction of
of viral hepatitis and 18 cases of cholera were also re- fish habitat outside of the Pachitea is likely impacting the
ported. The threat of waterborne pathogens is indicated in quality of fisheries in the basin. There are also occasional
certain water samples from this same time period collected commercial fishing vessels from Pucallpa that ply the lower
by the Environmental Unit of the Pichis-Palcazu Special reaches of the Pachitea and extract large quantities of
Project (administered through the Ministry of the Presi- fish for markets elsewhere. In past decades, dynamite was
dent). In their monitoring report published in April 1999, used as a popular form of fishing that proved especially
the unit reported positive detections of fecal coliform bac- destructive to aquatic communities. In fact, Bayley (1981)
teria in six of 14 samples collected, although concentra- reported that dynamite was the most common method of
tions never exceeded 9 MPN/100ml (PEPP, 1999). Total fishing used in the Palcazu valley at the beginning of the
coliform bacteria were detected in all samples, with con- 1980s. Our results suggest that dynamite is still used, but
centrations ranging from 4 to 240 MPN/100ml. Even higher at a reduced intensity. By far the most popular methods of
concentrations of coliform bacteria reported in other parts fishing were hook and line, cast nets, and gill nets.
of the Peruvian Amazon suggest that the problem of or-
ganic wastes is more or less ubiquitous (Gómez, 1994). Water Use among Indigenous vs. Colonist
The strong dependence of the local people on fish Households
consumption makes their health even more vulnerable to
Some of the most interesting and informative results
diminishing water quality, as productive fisheries rely on of our survey come from comparing water use patterns
clean water. Based on data collected in this study, indig- between indigenous and non-indigenous households. In
enous people consume approximately 120 g of fish per several respects, indigenous households are in more inti-
day per person. For colonists and townspeople, the per mate contact with the region’s water and thus more vul-
capita consumption rates are approximately 71 and 79 g/ nerable to the ill effects of poor water management.
day, respectively. Our values for indigenous consumption Indigenous households registered, on average, lower per
of fish compare well with the 122 g/day per person re- capita water consumption (12.5 l/day) than colonists (15.8
ported for the Shiringamazu Community of the Palcazu l/day), but they gathered a larger proportion of their water
Valley (Bayley, 1981) and the 175 g/day per person re- manually and drank it untreated more often. Indigenous
ported for the Campas Community of the Pichis Valley households eat fish and other aquatic flora and fauna in
(Gaviria, 1980). When indigenous, colonist, and townspeople larger quantities and make nearly equal use of rivers and
consumption rates are taken together, however, they are streams for recreation. Indigenous households also make
less than the estimated 136 g/day/person calculated for greater use of rivers as a transportation media.
fish consumption among mixed people of the nearby
While these use patterns seem to expose them more
Ucayali river valley in a 1966 survey (Pierret and directly to the harmful effects of poor water quality, indig-
Dourojeanni, 1967). In that survey, fish were found to con- enous people appear to better discriminate between water
stitute 62 percent, by weight, of the animal protein con- sources and waste disposal areas than do rural colonists
sumed by those interviewed. If we assume that total daily (Tables 2 and 6). Whereas rivers were a source of water
consumption of meat (222 g/day/person) is similar between for human consumption among 30 percent colonist house-
the two populations interviewed, then fish still constitute holds, indigenous households took larger percentages of
between 32 percent and 54 percent of animal protein con- their water for human consumption from streams and
sumed by local people in the Pachitea, with the remainder springs. Conversely, indigenous households deposited larger
coming from wild animals, chickens, pigs, and, to a lesser percentages of their waste in rivers, downstream of their
extent, cattle.
water sources. Indigenous households also, on average,
Productive fisheries also rely on intact fish habitat, located their latrines slightly further from water sources.
and thus human health is also linked to habitat preserva- It is tempting to attribute these seemingly more intelligent
tion. To date there has been little perturbation of the river use patterns to traditional knowledge among the indigenous
system’s natural structure and flooding patterns. No ma- communities, and this may in fact be true. Perhaps counter-
jor engineering structures such as dams, dikes, or diver- intuitively, however, indigenous households make more use
sion canals have been constructed along the main rivers of methods of fishing such as explosives and poisons (in-
of the Pachitea basin, and none are immediately planned cluding the natural poison, barbasco)(Table 9), which kill
to our knowledge. However, several species of fish eaten indiscriminately and are widely held to be unsustainable.
by local people actually migrate annually into the basin in
what the natives have traditionally called the mijano. These Water Use in Mountain Valleys vs. Lowlands
species include the boquichico (Prochilodus cf nigricans)
There are also distinct patterns in water use between
and catfish like the zungaro (Zungaro zungaro) that make households located in the mountainous portions of the
use of the extensive floodplains of the Ucayali river. These western basin and the eastern lowlands. A much larger
floodplains are under severe pressure from colonists who proportion of mountain valley households have water piped
IWRA, Water International, Volume 26, Number 3, September 2001

408
M.E. McClain, L.M. Aparicio, and C.A. Llerena
into them, and only 35 percent of households reported col-
Protection of surface water quality must be the main
lecting their water manually. This compares to 94 percent focus of water management efforts in the basin, as this is
of lowland households. Despite the convenience of water most likely to be the limiting characteristic of water for
being piped into their homes, inhabitants of the mountain both human consumption and fish productivity. Already
valleys use less water per capita (12 l/day) than do low- between 74 percent and 96 percent of households report
land inhabitants (17 l/day). A larger proportion of moun- that they use some system to evacuate fecal wastes. These
tain valley households also report treating their water prior include both latrines and, in towns, simple sewer systems.
to consumption. The relative abundance of piped water Properly designed and placed, latrines will effectively pro-
may be attributed to the longer history of colonization in tect water bodies from contamination, but sewer systems
the mountain valleys and more economically successful of the basin generally discharge directly into water bodies.
communities. The mountain valleys are also better served Towns should evaluate the option of discharging their
by roads, and thus the use of rivers for the transportation wastes into natural or artificial wetlands for treatment prior
of people and products is substantially less. The use of to discharge to the river system. When considering natu-
rivers for recreation, however, is actually higher.
ral wetlands, however, care must be taken to evaluate
An especially interesting observation is that while benefits against losses in species habitat. Proper land
fewer mountain valley households reported fishing, the management in uplands is also crucial, as mismanagement
consumption of fish (both percent of households consum- leads to increased erosion and further deterioration of water
ing and quantity consumed) is comparable between moun- quality.
tain valley and lowland households. As in the lowlands,
Across the basin, households should take care to con-
commercial fishing is nearly non-existent in the mountain struct adequate latrines, and they should also be especially
valleys, so the difference between what is caught and what diligent about boiling or otherwise treating water prior to
is consumed must be made up for by importation of fish from consumption. Moreover, they should strive to conserve
outside the basin. According to data presented by Aparicio those parts of the natural aquatic system that provide water
(1999), fish imported to the Huancabamba valley are ma- purification services (McClain, in press). Expensive wa-
rine species that come mainly from the capital city of Lima. ter purification systems are unlikely to be constructed in
the Pachitea for many years and the basin’s inhabitants
Management Implications
must rely on natural services provided by riparian forests,
The many uses described in this paper demonstrate wetlands, and hyporheic zones. The conservation of ripar-
that the relationships between local people of the central ian buffer strips is a common recommendation in develop-
Peruvian Amazon and their water resources, especially ment activities in the Peruvian Amazon
surface water, are direct and in many ways more intimate (INADE-APODESA, 1990) and it is required by law, but
than in more developed or less humid areas. Water at its still many landholders cut them.
source (river, stream, spring) provides for household needs,
Comprehensive and effective water management in
carries away wastes, waters animals, provides food, trans- the basin will require a Master Plan for aquatic resources
ports people and products, and serves recreational needs. protection and management that integrates finer-scale plans
The importance of water at its source cannot be over- of local communities (native, peasant and colonist), pro-
stated, as this fundamental issue markedly distinguishes ducers associations, municipal governments, conservation
the humid Amazon portion of Peru from the less humid areas, and other water-related plans. These plans should
cordillera and arid coast and has profound implications for be developed through a participatory methodology that
water management.
actively involves local people, as it is in their hands that
Households and communities of the basin are linked the ultimate power for implementation lies.
by the flow of streams and rivers. This is mainly an up-
stream-downstream linkage, but downstream inhabitants
Conclusions
may also significantly affect the aquatic resources of up-
stream inhabitants by altering the annual upstream migra-
The results presented in this paper have illuminated
tion of fish. In the absence of dams, reservoirs, irrigation the relationship between people and water resources in a
canals, and other significant water supply infrastructure, region of the humid tropics for which virtually no quantita-
little government control is exerted over water resources tive data previously existed. Inhabitants of the Pachitea
in the Pachitea basin. Water resources management is basin live in intimate contact with their aquatic resources,
thus left mainly to communities and individual households. and their health and dietary security are closely linked to
In light of basinwide linkages of communities and house- the quality of the basin’s waters and vitality of its fisher-
holds via river and stream flow, management efforts should ies. We have drawn the following conclusions from our
be closely coordinated on a basinwide scale. To date, how- analysis.
ever, there is little or no coordinated activity among the
basin communities or inhabitants to manage water and
aquatic resources.
•
Surface water is by far the predominant source of water
for inhabitants of the basin, irregardless of social group
IWRA, Water International, Volume 26, Number 3, September 2001

Water Use and Protection in Rural Communities of the Peruvian Amazon Basin
409
(indigenous, colonist, townspeople) or geographical lo-
cation (mountain valley, lowland). Colonists tend to take
their water most often from rivers, while indigenous
people take more water from streams and springs.
Water consumption varies from 12.5 to 17.0 l/day per
person, with the greatest difference between moun-
tain valley households, which use the lower value and
lowland households, which use the upper value.
Households in towns and mountain valleys are most
often served by public water supply and piped water,
but a majority of households also reported collecting
water manually from water sources. Greater than 90
percent of lowland households reported collecting their
water manually, with the highest single social group
percentage being registered by indigenous households.
Between 70 percent and 87 percent of all households
reported that they treat their water prior to consuming
it, with boiling ranking as the most common form of
treatment (70 to 83 percent). Indigenous households
reported consuming untreated water more often (30
percent), but it is likely that individuals from all social
groups consume untreated water on occasion when in
the countryside away from their homes.
importance of rivers and streams as sources of fish, as
avenues for transporting people and products, and as
recreational sites. These differences have strong rami-
fications for management and make the model of wa-
ter management developed in the more populated arid
regions of Peru inappropriate for the humid Amazon
region.
Water management efforts in the Pachitea basin should
focus on the protection of water quality in rivers and
streams to preserve human health and maintain fisher-
ies. In the absence of engineering controls, water qual-
ity protection will be best accomplished by careful
disposal of wastes away from water bodies and pres-
ervation of natural water purification features such as
riparian forests and wetlands.
•
•
•
•
•
Regional coordination of management activities is es-
sential given the close linkage of upstream and down-
stream water users. We recommend that a basinwide
Master Plan that empowers end-users and integrates
more detailed plans developed at community and asso-
ciation levels.
Acknowlegdements
•
Between 27 percent and 38 percent of households re-
ported dumping wastes into water bodies, but between
This research is part of the Andean Amazon Rivers
74 percent and 96 percent also reported that they have Analysis and Management (AARAM) project and was
some form of system to dispose of human fecal waste. funded through a grant from the Inter American Institute
In some cases, particularly in towns, this system may for Global Change Research (IAI). We thank Pedro Aguilar
discharge wastes into water bodies.
of Pro Naturaleza for logistical assistance in the
•
•
Irrigation is virtually non-existent in the basin and cul- Huancabamba Valley. We would also like to express our
tivators from all social groups and geographic areas appreciation to the people of the Pachitea basin who wel-
rely almost exclusively on direct rainfall to water their comed us into their homes and spoke honestly of their use
crops.
of water and concerns about its proper management.
Domestic animals also rely on surface water and very
often drink from the same sources as their owners.
Colonist households most frequently reported that their
animals drink from streams up gradient of human wa-
ter sources.
About the Authors
Michael E. McClain is an Assistant Professor in
the Department of Environmental Studies at Florida Inter-
•
•
Rivers are the most important aquatic avenue for the national University, Miami, Florida, USA. He specializes
transportation of people and products in the basin, and in the hydrology and water quality of large tropical rivers,
they also serve as the main site of water-related rec- with special emphasis on the Amazon. He is the General
reation for between 54 percent and 66 percent of Coordinator of the Andean Amazon Rivers Analysis and
households.
Fish are an important form of animal protein for be- mcclainm@fiu.edu.
tween 78 percent and 89 percent of households in the Carlos A. Llerena is a professor in the Department
Management
(AARAM)
project.
Email:
basin. Indigenous households consume the greatest of Forest Management at the Universidad Nacional Agraria
quantity of fish (120 g/day/person) and also catch fish La Molina, Lima, Peru. He specializes in watershed man-
more often (84 percent of households). Most fish are agement and forest hydrology in the Andean Amazon re-
caught in rivers, but fishing in streams is also impor- gion and is the Peruvian Coordinator of the Andean
tant. Indigenous households also make the most use of Amazon Rivers Analysis and Management (AARAM)
other aquatic and riparian resources such as medicinal project. Email: callerena@lamolina.edu.pe.
plants.
Luis Miguel Aparicio Alcázar is a Forestry Engi-
•
Water use patterns in the Pachitea basin are distinct neer who specializes in forestry management and the con-
from those in the higher Andean Cordillera and along servation of natural resources. He is currently with the
the Peruvian coast. The most important difference is Fundación peruana para la Conservación de la Naturaleza
the lack of irrigation in the Pachitea and the greater in Lima, Peru. Email: lmaparicio@yahoo.com.
IWRA, Water International, Volume 26, Number 3, September 2001

410
M.E. McClain, L.M. Aparicio, and C.A. Llerena
Discussions open until March 1, 2002.
University Press.
McClain, M.E. and C.A. Llerena Pinto. 1998. El manejo de cuencas
en la selva: de los Andes a la Amazonia. Agua y Riego 11:
11–13.
References
Aparicio Alcázar, L.M. 1999. Diagnóstico del Uso Actual de los
Recursos Hídricos de la Cuenca del Río Pachitea. Ing. The-
sis in the Facultad de Ciencia Forestales, Universidad
Nacional Agraria La Molina, Lima.
ONERN (Oficina Nacional de Evaluación de Recursos Natu-
rales). 1976. El Mapa Ecologico del Perú.
PEPP (Proyecto Especial Pichis-Palcazu). 1999. Segundo
Monitoreo de la Calidad de Agua Valles Pichis Palcazu. A
report of the Comision de Lucha Contra el Consumo de
Drogas. Ministry of the President, Peru.
Bayley, P.B. 1981. “Fish Resources of the Palcazu valley: Effects
of the Road and Colonization on Conservation and Protein
Supply.” Appendix A of the Project Synthesis Paper of the
Central Selva Natural Resources Management Project. USAID
Project No. 527–0240. Lima, Peru.
Pierret, P.V. and M.J. Dourojeanni. 1967. “Importancia de la caza
para alimentación humana en el curso inferior del Rio Ucayali,
Peru.” Revista Forestal del Peru 1:10–21.
Gaviria, A. 1980. Consumo de carne de monto por los Campas Salo, J.R. and Kalliola. 1993. “Necesidades para el futuro en la
del Pichis. Puerto Bermudez. CIPA/COTESU Project Report.
Gómez, R. 1994. Contaminación Abiental en la Amazonia
Peruana. Informe Técnico de Avance of the Instituto de
Investigaciones de la Amazonia Peruana. Iquitos.
selva baja peruana: conservación e investigación cientifica.”
R. Kalliola, M. Puhakka, and W. Danjoy, eds. Amazonia
Peruana – Vegetación Húmeda Tropical en el Llano
Subandino. Jyväsklä: Paut and Onern: 239–249.
SENAMHI (Servicio Nacional de Meteorologia e Hidrologia).
1999. Precipitation Datasheets for Stations in the Pachitea
Basin.
INADE-APODESA. 1990. Desarrollo Sostenido de la Selva: un
Manual Tecnico para Promotores y Extensionistas. Serie
Documentos Tecnicos No. 25.
INEI (Instituto Nacional de Estadística e Informática). 1996. Cen-
sus. Lima Peru: Banco de Información Distrital.
Smith, R.C. 1983. Las Comunidades Nativas y el Mito del Gran
Vacío Amazónico. Asociación Interetnica de Desarrollo de
la Selva Peruana.
INRENA (Instituto Nacional de Recursos Naturales). 1996.
Estudio de Reconocimiento del Uso del Recurso Hidrico por
los Diferentes Sectores Productivos en el Peru. INR-42-
DGAS, Lima, Peru.
McClain, M.E. in press. “Management of Aquatic Systems:
Tropical Rivers.” Zalewski and Harper, eds. Ecohydrology:
Understanding and Applying Ecosystem Properties for Sus-
tainable Uses of Aquatic Resources. Cambridge: Cambridge
IWRA, Water International, Volume 26, Number 3, September 2001