Final Report submitted to NOAA’s Human Dimensions of Global Change Research (HDGCR) Program
Project Title: Climate
Variability and Household Welfare in the
Farmer adaptation and use
of weather forecasts in decision-making
Principal and Co-Principal Investigators:
Corinne Valdivia, Research Associate Professor
Department of Agricultural Economics,
200 Mumford Hall
(573) 882 4020
(573) 882 3958 (fax)
ValdiviaC@missouri.edu
Jere L. Gilles, Associate Professor
Department of Rural Sociology
University of Missouri-Columbia
106 Sociology Building
(573) 882 3791
GillesJ@missouri.edu
Roberto Quiroz, Head
Department of Production Systems and Natural Resource Management
R.Quiroz@cgiar.org
Christian
Jetté, Program Officer
United
Nations Development Program
La Paz, Bolivia
Christian.jette@UNDP.org
NOAA Award No. NA96GP0239
Period: August 1999 - July 2003
A. Project Abstract
Climatic variability, characterized by periodic droughts, El Niño
events, floods, and weather events such as frosts result in significant losses to
rural households of the Andes. The project answered three questions: 1) What
have farmers developed as successful strategies to cope with climatic variation
in the Andean region; 2) How do farmers currently use information from
forecasts and local sources to make production and consumption decisions; and
3) What mechanisms and institutions facilitate or constrain the utilization of climate
forecast information.
Three research activities were designed. The first, research on rural livelihoods strategies to study the factors that contribute or constrain coping with climate variability. Second was to understand the role of institutions and networks in coping with climate variability, and accessing forecast information. Finally, we studied current sources of forecast information known and/or used by farmers, the characteristics of the information, of the producers of the information, and the crops affected by climate variability. Our units of analysis were rural households, rural communities, institutions and biological models. At the household level we studied diversification strategies. At the community level we studied social networks in the diffusion of local knowledge forecasts. At the institutional level we studied the role of Radio, and Early Warning and Food Security in use of probabilistic forecast. At the biophysical level we worked on modeling climate variability in local potato varieties to measure impact of local conditions.
We conducted research in three agropastoral communities in the
Altiplano of Bolivia and
The study of
networks in
B. Objectives of the Research Project
The
research project addressed three questions: 1) What have farmers developed as successful
strategies to cope with climatic variation in the Andean region; 2) How do
farmers currently use information from forecasts and local sources to make
production and consumption decisions; and 3) What mechanisms and institutions
facilitate or constrain the utilization of information about climatic
risk. To identify how climate and other factors
interact and impact the coping strategies of rural households in the Andean
region of
C. Approach
Our framework to understand how households, individual members and communities cope with climate variability our framework is informed by the sustainable livelihoods literature. This framework provides entry points for economics, sociology and the biophysical sciences involved in our research, to address the question of how households cope and adapt to stressors and shocks. Specifically we apply household economics, political economy, network analysis, and rural sociology to understand what factors contribute to coping and adapting to climate variability. We developed a methodology to identify household economic strategies and portfolios. Focus groups, household formal surveys and in-depth interviews were techniques to collect data from these three agropastoral communities. Through cluster analysis we identify the relationship between access and control of assets, life cycle stage, and composition of strategy.
To evaluate changes through time in terms of income and
diversification of the economic portfolio, cluster analysis is applied to
household data from 1993, 1995 and 1999 in
We used a case study approach – multiple embedded design- to understand dynamic processes in coping and adapting to climate. The study of potato production, household diversification and coping strategies in San José Llanga aimed at understanding how people cope in the short run. A study of the factors affecting market integration (buying and selling products) identified the socioeconomic and marketing characteristics of transaction costs, using both case study methodology and regression analysis.
Methods to study local knowledge and use of forecasts included
focus groups, household surveys, and ethnographic methods. Three case studies of community networks to access
information about local knowledge indicators of climate were completed. The
types of networks, nodes of information, and the type of information flow were studied.
Biophysical models were developed to identify the potential performance of local varieties under climate scenarios. The formal household survey included questions on production inputs, soil types, feed resources, yields, and total outputs. Our collaborators experimented with farmers in their fields to obtain data to calibrate biological models. Potato varieties were modeled, as well as performance of animal species. Agroecological zoning maps were constructed to understand the land use potential. These combined with biological maps will be used to understand possible outcomes of various climate events, at a local scale.
The working hypotheses of this research are:
The
II. Interactions with decisions-makers
Our project included collaboration with local institutions as
partners or as members of the Expert Panel.
The Expert Panel role was to act as advisor to the research activities
being conducted each year. The decision
makers that we collaborated with are two local nongovernmental organizations
that work with end users in development activities: PROINPA (
We had the opportunity to share our research approach and
findings with several institutions. We presented
at the IRI Communication of Climate Forecast Information Workshop (June 6,7 and
8 2001) in Palisades NY, organized by Dr. Jennifer Philips.
Instituto Geofísico
del Peru to be part of the advisors in the Expert Panel for the project. Dr.
We collaborated with
the Sistema Nacional de Seguimiento de la Seguridad y Alimentaria y Alerta
Temprana SINSAAT (National System of Food Security and Early Warning), a
Bolivian institution. Javier Choquevilca coordinator of this project was a
member of the expert/advisory panel. This organization was a case study to
understand how to reach end users.
Irene Trebejo and Esequiel Villegas from SENAHMI Peru were invited
to participate in the final workshop of the project. Trebejo was also invited to participate in El
Niño workshop in
C. Coordination
with other projects of the NOAA Climate and Societal Interactions Integrated
Sciences and Assessments
Our coordination has mostly
been with the Human Dimension program, especially focusing on agriculture and
developing country settings. Valdivia
and Gilles participated in the Principal Investigators Meetings of 1999 and
2002. Valdivia, Gilles, and Barreda
participated in Open Meeting of the Human Dimensions of Global Environmental
Change Research Community in Montreal 2003. Valdivia and Gilles presented at
the Conference on Climate Prediction Agriculture and Development, organized by
IRI (
III. Accomplishments
A.
Brief Discussion of Research Tasks Accomplished
The first activity of the project was to constitute an expert panel of
researchers and potential users of forecasts in Peru and Bolivia to review the
proposal and discuss development of field instruments to collect data to
understand the production systems, the effect of climate especially of El Niño
97-98, and to develop the plan for field work in Bolivia the first year of the
project. A team of researchers (Valdivia,
Jette, Espejo, Gilles, Carrillo) developed, tested and applied the household survey.
Carrillo from PROINPA monitored crops, including varieties provided by
PROINPA. This was added following advice
from reviewers on the role of availability of risk reducing technologies. The first household survey was developed,
tested, and applied to 45 households in
Household Portfolios and
Livelihood Strategies
Focus groups were conducted at three sites at the beginning
of the research. This informed the
development of household surveys to collect information on household
production, income and assets, as well as production during El Niño of 1997-1998, and sources of
information for decisions. Data was
collected of 45 households in San José Llanga, 52 households in Santa María,
and 58 households in Anccaca. A data base for
Study of Local
Knowledge Networks and Institutions
While awareness of El Niño existed among some households,
most of the information on climate for decisions was based on local
knowledge. Two approaches were used to
understand access and use of forecast information. Three case studies were developed, the
community networks, Radio San Gabriel mentioned in the survey by community
members, and SINSAAT the early warning and food security project in the
Ministry of Agriculture. The
methodology to identify networks was based on household surveys, ethnographic
methods, focus groups, and informal interviews. An ethnographic approach using
non-structured interviews and detailed observations of community life was used
in San Jose Llanga to determine the structure of networks. A matrix of all members of the community was
created and we determined who people looked to whom for weather information. Persons who had many people referring to them
were identified as nodes in the networks.
The ethnographic approach is very resource and time intensive because it
requires that the researcher live for long periods in the community. In
Biological
Models for Decisions
The focus of this component of
the project is to develop the capacity for modeling technological options under
climatic variations in the high-plateau. The tasks included: 1) definition of
agro ecological zones in the watersheds where the target communities are
located, 2) the adaptation of crop and livestock models to the soil, climate,
and management conditions of the area, 3) ex-ante assessing the expected
outcomes of production decisions using biophysical models (scenarios), 4) the
development of tools for integrating climatic, soil, and management conditions
at different spatial scales, 5) training stakeholders in the use of tools and
methods developed.
PROINPA implemented a
series of field experiments to select potato varieties tolerant and resistant
to drought and frost. The findings were systematized together with soil and
climate data and used to re-parameterize the DSSAT-SUBSTOR model. With this
adjustment the model is now suitable to simulate the expected response of
several potato varieties to climate variability.
B. Provide two or three overheads of key research results
in bullet form. (CD)
C. Elaboration
of key findings
Livelihood
Strategies
Access to resources, stage in the life cycle, livestock assets, quality of resources, and off farm employment are significant variables in developing distinct economic portfolios.
• The composition
of activities determines the type of diversification. If the diversification includes less
covariant activities the household has the capacity to smooth income and cope
with stress or shock. In
• In
2000 in
Institutions: Markets and Networks
• Rural-urban
integration plays a significant role in both Peruvian communities, as it
resulted in off farm income/employment. In
• Social
networks as a form of intangible asset play an important role in accessing and
claiming labor, land, remittances and information. It played a significant role in smoothing
consumption and income among the elderly through remittances, both in
Coping and adapting
•
Income differences among households in rural
communities are significant. Households
with assets in dairy production are not worried about droughts or frosts. If
crops are lost they have access to credit to purchase seed and food. Their source of stress and shocks are market
policies and political unrest(Materer, 2001; Valdivia et al 2002).
•
Potato producers are concerned about the season.
If losses are experienced, there are impacts on consumption and schooling, as
they have no access to credit. Markets
are also a concern as good crop years often face low prices. Opportunistic activities like chuño
production are used to store and sell when prices rise.
•
Households smooth consumption with sheep and chuño,
the latter can be stored in good years for years of stress or shock. Lack of chuño is an indicator of a bad year,
which can actually be a cumulative effect of many years of climate stress.
•
Who are interested in local knowledge forecasts
and new information? The less diversified, more vulnerable crop producers, and the
wealthier farmers investing in both dairy and potato production.
Local Networks/San Jose Llanga Bolivia
• Andean producers used forecasts to make
production systems but they did not use scientific forecasts.
• Farmers used local networks organized around
local experts (nodes) to get forecast information.
• Climate information networks were organized
on a neighborhood level around a local expert.
The expert of one neighborhood was the center of forecast information
for the entire community. Neighborhoods
networks were primarily connected through their nodes.
• Most farmers did not discuss forecasts with
local experts but simply observed their behavior and followed their lead.
• Local experts were older and were among the
most productive potato producers. They
did not use scientifically generated information in their forecasting and they
did not have contact with local extension services.
• Farmers in San Jose Llanga believe that
forecasts are only valid for the location in which they are generated.
• Knowledge of traditional forecasting
indicators is widespread, but the ability to interpret multiple and sometimes
contradictory indicators is not.
Findings Extra-local sources/SINSAAT and Radio San Gabriel
SINSAAT is the
office of the Ministry of Agriculture charged with producing
forecasts–especially those that are related to food security. Radio
• SINSAAT is not currently producing information
to be used by producers. Their audiences are government agencies and
non-governmental groups interested in agriculture and food security. It is using European Union assistance to
develop a system for distribution forecasts down to extension workers at the municipio
level.
• Present methods of delivering forecasts via
Spanish language bulletins are not adapted to the needs of producers. They arrive after production decisions are
made and are not in the appropriate language.
• The people on whom producers rely on for
local forecasts do not have contacts with municipio level
agricultural officers, so that more timely delivery of forecasts would still
not reach producers.
• Radio
Findings in Puno/Ancacca and
• Peruvian producers like their Bolivian
counterparts did not use scientific forecasts.
• The structure of the networks and the
characteristics of the nodes observed in
• Knowledge of forecast indicators was
widespread in the communities with non-experts citing more traditional
indicators than the nodes.
• Forecasting seems to be incompatible with
off-farm employment. All nodes depended
on crop production for their livelihoods.
•
One
community with some communal cropland had open discussions of weather forecasts
surrounding decisions about the use of common land. This information flow was in addition to
flows through the networks.
• Agroecological zoning
(AEZ) is a method that uses biophysical attributes of the land to cluster
land-use types into more homogeneous areas. This exercise facilitates planning
for the sustainable use of natural resources, particularly in areas with high
climatic variability. The application of AEZ is limited by the lack of
geospatial data, particularly in mountainous areas. Remote sensing and
process-based models for both climate interpolation and crop and livestock
production were used in the Ilave-Huenque,
•
Crop and livestock simulation models were
developed or adapted for the agro ecological conditions of the high plateau.
Field experiments were conducted to calibrate the mathematical equations to
describe the biophysical processes under cold temperatures and drought stress.
Figure 2 (Annex 1) shows the confidence interval of biomass production for
three potato varieties throughout the growing period. The results show that the
model tends to overestimate the aerial biomass of the plant in the early
development stage but as the plant matures, the simulation lies in between the
95 % confidence interval of the experimental data. The simulation at harvest
time is the critical one, so the models can adequately simulate the expected
yields under stressed conditions and can now be used to assess the expected
yields under climatic changes.
•
Once the simulation models are validated they
can be used to explain the impact of climate variability and ex-ante asses
future risks. Rainfall, and temperature patterns and the presence or absence of
ENSO events are not enough to explain the behavior of crops (Figure 3 Annex 1)
under climatic stress. For instance, the
seasons 82-83 (or 3 in the sequence), 86-87 (7), 91-92 (12), 92-93 (13), 94-95
(15), and 97-98 (18) were El Niño years. It is evident that the behavior of the
climate in the highlands was different for all of those years. A similar trend
is seen for La Niña (88-89 and 95-96). As can be seen in the graphs, there is
not a direct relation between the presence/absence of an ENSO event and potato
yield. A closer look to rainfall, average minimum temperature and absolute
minimum temperature indicates that even these average figures do not correlate
well with potato yield. The distribution, intensity and duration of rain or
frost events are the determinants of crop behavior. For instance, water logging
in critical periods explained the very low yields in 84-84 and 95-96, whereas
higher yields were attained in 81-82 with similar average values of rainfall
and temperature. The simulation models thus constitute a valuable tool to
integrate the different factors affecting crop yields and are useful to develop
alternative technologies to cope with the impact of climate variability, once
the forecasting skill improves.
•
Figure 4 (Annex 1) shows an example of the use
of models in livestock production. The models were built and validated with
farm data and then used to assess the impact of technological options to
increase household income. It is interesting to note that the introduction of
alfalfa is an alternative that not only decreases the negative impact of
climatic variability on the economic activities of the farmers but also
increases income through milk production. An additional benefit from strategic
supplementation to dairy cattle comes from the environmental side. The
environmental cost of producing milk, in terms of methane production, increases
with low quality roughages and less efficient animals, particularly during the
Andean winter. This cost is reduced substantially with adequate
supplementation.
•
The tools for up- and downscaling are being
developed. A preliminary version is already operational. Training in the
development and use of systems analysis tools is a continuous activity of the
project. Not only institutions collaborating have been using the models
developed for the
• Analysis of tercile probabilities was undertaken for
the Aroma region where San José Llanga is located. The rainfall and temperature were the
measured in terms of crops yields. Under
this new consideration, there is a drastic change in the way we measure the
effects of different phases of ENSO in the agriculture. For example, under El
Niño year, a statistical approach gave a probability of 60% of a
What this means
1. Some farmers in the
Andean region have developed diversification strategies, successful in coping
with climate and other factors, while others are becoming increasingly
vulnerable. Those that are vulnerable, in farming seek information about
climate forecasts from local sources. Those that have crops in their portfolio
seek information, though the degree of actions they may take with the
information may be constrained. While
most of the studies show that farmers in developing countries have other
stressors and can’t benefit from forecasts, our study in Andean communities
found that in a given community some will benefit from more information while
others won’t. Previous work
indicated that because of their level of vulnerability farmers are not willing
to risk change. Other stressors play a
greater role than climate. Only farmers
that are specialized and wealthy can bet on forecasts.
Our research
indicates there is a diversity of livelihood strategies even in one community. Farmers
in the Andean communities have been coping and adapting to climate variability
and other stressors. Our research shows that the interactions between the structures
(social market and political), climate, and policies may result in increased
vulnerability through time for households constrained by their assets,
increasing vulnerability. While some
farmers benefited from the government and foreign assistance agencies in the
establishment of dairy activities in the Altiplano, many did not have the land
and irrigation resources to take advantage of this opportunity. These
households have relied increasingly on potato marketing, and therefore do take
into account climate, but locally generated.
They also have less choices and insurance mechanisms, but still
flexibility to manage resources to adjust planting and varieties. Households
experiment and have been incorporating new varieties in their portfolio of
crops. New technologies on the one hand
require new management knowledge, as well as understanding of the interactions
with climate. This creates an opportunity for collaborating in the production
of information for decisions. On the
other hand, increased dependence on capital intensive technologies like
tractors, which are rented, has limited the flexibility in planting decisions.
New technologies
with supporting policies and secure markets have provided opportunities for
some rural households to incorporate activities less sensitive to climate
variability. These households are more diversified in non covariant activities,
and still pursue local knowledge forecasts for potato planting. These households can assume the risk of loss
when climate stress or shock takes place.
Secure dairy markets provide an insurance (credit) against crop
losses. Political unrest and
privatization are the shocks of concern to this group. Our study in
2. Farmers don’t incorporate scientific forecasts in their decisions because of a preference for locally based forecasts. Previous work in more developed countries has suggested that a barrier to forecast use has been lack of understanding of probabilistic forecasts. Considerable effort has been placed on educating users on how to interpret these.
Understanding probabilities does not seem to be the main
barrier in the Altiplano. The local,
traditional forecasters that producers rely upon, use techniques that include
an intuitive approach to probabilities.
Local forecasters are potentially the conduit by which forecasts could
reach producers. If local experts
believed in the accuracy of scientific forecasts, this information would be
incorporated into their predictions. In the case of Altiplano producers, the
widespread belief that forecasts are only valid for the location in which they
are generated – i.e. a forecast produced by an office in
3. The current research suggests that we should modify some of our beliefs about indigenous knowledge. It has long been noted by anthropologists and folklorists that traditional knowledge disappears with modernization and the incorporation of indigenous peoples into global and national economic systems. Traditional knowledge becomes confined to the elderly and is lost when they die.
The situation found in the 3 communities examined in this study suggests that these ideas need to be modified. Although local climate experts are elderly, the knowledge of traditional forecast indicators is not limited to this cohort of individuals. Knowledge of traditional indicators remains widely distributed across age and economic groups. The ability to manage multiple, sometimes contradictory, indicators is being lost however. Almost all of the local experts were nearly full time farmers who spent nearly all of their time in their villages working on their farms. People with off-farm or non-farm activities did not spend sufficient time in the field to develop accurate forecasts and people who were primarily livestock producers were not as interested in forecast information. In short one of the techniques that farmers use to deal with risk, income diversification, is undermining their abilities to forecast climate risk.
Because traditional knowledge is still widespread and because it uses some indicators that have a scientific basis, the best way to improve and to communicate forecast information is probably through a partnership with local experts. We propose that local scale information tools act as a means to communicating and discussing forecasts, providing a space to asses the perceptions of risks, and the actual risks that any new technology entails, and that farmers evaluate.
D. List of
Publications and presentations arising from this project;
Publications
Valdivia, C. and R. Quiroz. 2003. “Coping
and Adapting to Increased Climate Variability in the
Valdivia, C. and J. Gilles. 2001. “Gender and Resource Management: Households and Groups, Strategies and Transitions.” Agriculture and Human Values. 18 (1): 5-9.
Valdivia,
C., C. Jetté, R. Quiroz, J. Gilles, and S. Materer. 2000. “Peasant Household Strategies in the
Valdivia, C., J. L. Gilles, and S. Materer. 2000. “Climate Variability, a Producer
Typology
and the Use of Forecasts: Experience From Andean Semiarid Small Holder Producers.” Proceedings of the
International Forum on Climate Prediction Agriculture and Development.
International Research Institute for Climate Prediction.
Related Publications (benefited by the project funded
by NOAA in
Valdivia, C.
2001. “Household socioeconomic diversity and coping response to a drought year
at San José Llanga.” Chapter 6 in Coppock, D. L. and C. Valdivia (eds). Sustaining
Agropastoralism on the Bolivian Altiplano: The Case of
Markowitz, L.
and C. Valdivia. 2001. “Patterns of
technology adoption at San José Llanga: Lessons in agricultural change.” Chapter 7 in Coppock, L. D. and C. Valdivia
(eds). Sustaining Agropastoralism on the Bolivian Altiplano: The Case of
Coppock, D. L.,
C. Valdivia, J. Yazman, C. Jetté, J. de Queiroz, L. Markowitz and I. M. Ortega.
2001. “Conclusions and Recommendations.” Chapter 8 in Coppock, D. L. and C.
Valdivia (eds). Sustaining Agropastoralism on the Bolivian Altiplano: The
Case of
Agricultural Economics Working Papers
Valdivia, C.,
C. Jetté, R. Quiroz, J. Gilles and S. Materer. 2001. "Peasant Household
Strategies in the
Materer, S., C.
Valdivia and J. Gilles. 2001. "Indigenous Knowledge Systems:
Characteristics and Importance to Climatic Uncertainty." AEWP-2001-03.
Department of Agricultural Economics,
Materer, S. and
C. Valdivia. 2000. "Analysis of a Climatically Variable Production season." AEWP-2000-10. Dept. of Agricultural
Economics,
Materer, S. and
C. Valdivia. 2000. "Household Production Strategies in a Climatic Variable
Zone." AEWP-2000-9. Dept. of Ag. Economics,
Abstracts
Valdivia, C.
and J. L. Gilles. 2003. “Coping and adapting to climate variability in the
Barreda, C., C. Valdivia, and R. Quiroz. 2003. “Diversification of economic portfolios to
deal with climate variability: Case study of livelihood strategies in two
Andean communities.” Open Meeting of the Human Dimensions of Global
Environmental Change Research Community. 16-18 October. Montréal, Canada.
Valdivia, C.
and R. Quiroz. 2003. “Coping and Adapting to Increased Climate Variability in
the
Espejo,
Rigoberto, Jere Gilles and Hector Machicado, 2003 “El Valor de los estudios de redes para la
difusion de informacion climatica,” Presented at the Final Workshop of the
Research Project on Family Viability and Well-being in the
Gilles, Jere,
and Rigoberto Espejo, Conocimientos
cientificos y locales. 2003 Presented at
the Final Workshop of the Research Project on Family Viability and Well-being
in the
Valdivia, C.
2002. “Local Level Factors Affecting Forecast Use and Application.” Session
with
Gilles, J.L. 2002. Panel member, “Eliciting User
Needs and Methods from Different Disciplines.” Human Dimensions of Global
Change Principal Investigators Meeting, October 23 – 25,
Valdivia, C., R. Quiroz, P. Zorogastua and G.
Baigorria, 2002. “Climate Variability, Andean Livelihood Strategies,
Development and Adaptation in the Andean Region.” Invited presentation Climate
and Development From Seasons to Centuries: How Our Understanding of and
Responses to Seasonal Climate Variability Can Build Insight into Human
Adaptation to Long-Term Climate Change. S 123. Spring Meeting American
Geophysical Union Vol 83, No 19, 7 May.
Valdivia, C.
2001. “Andean Livelihood Strategies and the Livestock Portfolio.” American
Anthropological Association 100 Annual Meeting. Invited Session,
Contentments and Contentions: Living with Livestock. November 28- December 2.
Valdivia C, J. Gilles, R Espejo and R. Carrillo. 2001.
“Current Users and Diffusion Nodes of Local Climate Forecasts in the Andes of
Bolivia: Lessons on potential users, timing and content of climate forecast
communications.” IRI Communication of Climate Forecast Information Workshop
Proceedings. June 6,7 and 8.
Gilles, J., S.
Materer, and C. Valdivia. 2001. "Re-evaluating Climate Forecasting:
Lessons from Indigenous Systems." 64th
Rural Sociological Society Meetings, August 15-19.
Valdivia,
C., J. L. Gilles, and S. Materer. 2000.
“Climate Variability, A Producer Typology and the Use of Forecasts: Experience
From Andean Semiarid Small Holder
Producers.” Proceedings of the International Forum on Climate
Prediction Agriculture and Development. International Research Institute
for Climate Prediction.
Gilles, J., F. Galindo, C. Valdivia, and S. Materer. 2000. "Knowing the Future: Climate Forecasting, Farmers and the Food System." 63rd Rural Sociological Society Meetings, August 14-18. Washington D.C. Abstract and paper available.