STAFF PAPER SERIES
_, UNIVERSITY OF
Institute of Food and Agricultural Sciences
Food and Resource Economics Department
Gainesville, Florida 32611
POTENTIAL RESPONSE OF NORTH CENTRAL
FLORIDA LIVESTOCK PRODUCERS TO LONG
TERM CLIMATE FORECASTING
Norman Breuer, Victor Cabrera, Peter E. Hildebrand, and James W. Jones
Staff Paper SP 03-4
Potential Response of north central Florida Livestock
Producers to Long-Term Climate Forecasting
Norman Breuer1, Victor Cabrera Peter E. Hildebrand3 and James W. Jones4
'Rosenstiel School of Marine and Atmospheric Science. 4600 Rickenbacker Causeway, Miami,
2Ph.D. Candidate in Interdisciplinary Ecology, College of Natural Resources and Environment,
UF. 2126 Mcarty Hall PO Box 110240, Gainesville, FL 32611-0240
3 Professor, Food and Resources Economics, University of Florida, 2126 MCarty Hall PO Box
110240, Gainesville, FL 32611-0240
4Distinguished Professor, Agricultural and Biological Engineering, University of Florida, 288
Frazier-Rogers Hall PO Box 110570, Gainesville, FL 32611-0570
Key Words: ENSO, climate prediction, ethnographic linear programming, Sondeo, cattle,
Beef cattle production in North Central Florida is based on grazing natural and planted
pastures. In this tropical grass-based system cattle production is often constrained by
drought and cold temperatures. We investigated the potential for using improved climate
forecasting of El Nifio / La Nifia phases to aid ranchers in their management decisions.
Stakeholder participation was included through the use of Farming Systems
methodologies, especially Sondeos. Sondeos are rapid, low cost, effective
multidisciplinary team evaluations of rural situations. Climate models were developed by
a consortium of Florida universities in conjunction with the National Oceanographic and
Atmospheric Administration (NOAA.) This paper describes the Sondeo process of
interaction with stakeholders, and examines several scenarios of differential economic
outcomes and potential for adoption of this technology by the rancher clientele using
linear programming-based modeling. Ethnographic Linear Programming with livestock
producers improved input coefficients for the model. El Nifio Southern Oscillation or
ENSO phase effects on growth of commonly used forage species were translated to
stocking rates and used as LP inputs. The worst-case scenario was two consecutive La
Nifia events and the best was two consecutive El Nifio phases. Beef cattle ranchers could
use this information to adjust their management strategies and improve profit or
ameliorate potential losses.
Agriculture is highly influenced by climatic variability. Oram (1989) has characterized
agriculture as the most weather-dependent of all human activities. A logical extension
of this dependency of agriculture on climate variability is that if climate were known
ahead of time, decisions could be made that would reduce the negative impacts of
expected bad weather or take advantage of expected good weather conditions. Recent
advances by scientists in understanding global ocean and atmospheric processes have
led to new capabilities for forecasting climate several months to a year in advance
(Jones et al., 2000; Hansen et al., 1999a; Hansen et al., 1999; Hansen et al., 1989b).
Most of these advances rely in some way on knowledge of the surface temperatures in
the Tropical Pacific Ocean and the El Nifio Southern Oscillation (ENSO)
ENSO refers to shifts in sea surface temperature (SST) in the eastern equatorial Pacific
and related shifts in barometric pressure gradients and wind patterns in the tropical
Pacific (the southern oscillation). ENSO activity is characterized by warm (El Niflo),
neutral, or cool (La Nifia) phases identified by SST anomalies. The impact of these
phenomena affects interannual variability of weather in many regions (Kiladis and
Diaz, 1989: Ropelewski and Halpert, 1986, 1987, 1996). Most current climate
predictions are based on ENSO. Despite the fact that availability of climate data and
information for agricultural use has increased dramatically in the last 20 years
(Changnon and Kennth, 1999), progress in the systematic use of climate forecast has
been rather slow (Hammer et al., 2001; Goddard et al., 2001). Jones et al., (2000 a,b)
suggested that there are opportunities to improve climate forecast dissemination,
communication and interpretations and to develop or adapt research tools, methods
and data products for translating climate forecasts into information required to support
agricultural decision making.
In Florida, El Niflo winter months tend to be cooler and with higher rainfall and La
Nifia years tend to be warmer and drier than normal in the autumn though the spring,
with the strongest effect in the winter. Increased autumn and winter rainfall is
associated with reduced solar radiation in El Niflo years (Hansen et al., 1999).
Production and economic impact of ENSO on Florida field and vegetable crops is well
documented. Previous studies have demonstrated that a substantial portion of the inter-
annual variability for yields of maize (Handler, 1990; Hansen et al., 1998) and several
winter vegetables (Hansen et al., 1999) and tomatoes (Messina, 2002) in Florida is
associated with ENSO-related weather variability. Hansen et al (2001) described the
impact of ENSO events on tomatoes, bell pepper, sweet corn and snap beans (all
decreasing in El Niflo years), and sugarcane, tangerines and grapefruit (all increasing
in EL Niflo years). To date, however, relatively little attention has been paid to ENSO
effects on livestock production.
Florida has been characterized as being particularly vulnerable, with an excess of over
30% of the normal seasonal total precipitation across much of the state during an El
Nifio winter. During La Nifia years, the opposite effect occurs. Deficits of 10% to 30%
can last from fall through winter and spring. Monthly deviation from normal due to
either ENSO phase conditions exceeds 30% in all of Florida's climate divisions, and
50% in the southern peninsula. Higher than normal winter rainfall in El Niflo years can
affect yields of winter-harvested vegetables adversely. ( Letson et. al. 2001, Jagtap et
Evapotranspiration (ET) is another important measure of climate conditions. In most of
the state, reference ET is higher than normal in La Nifia years and lower than normal in
El Nifio years from November to March. Rainfall deficit may be critical from March to
May. In northern Florida, deficits are likely to be more severe in April, while south
Florida may be more affected by this phenomenon in March during La Nifia years. In El
Nifio years, from January to March, rainfall deficit is generally less. Impacts on rainfall
deficit are less consistent during summer.
Accumulation through time of temperatures above a base temperature of 50F (degree-
days) is used to estimate the time to flowering or maturity of several field crops. El Nifio
and La Nifia influence growing degree-days primarily in the winter (December through
February). For non day length-sensitive winter crops, physiological develop-ment is
likely to be about 5-10% faster than normal in December to February of La Nifia years,
and about 10-15% slower in El Nifio years in South Florida.
Livestock weight gains and milk production are reduced during periods when
temperatures exceed 770 ("heat stress degree-days"). Monthly heat stress degree-days
tend to be slightly lower in the spring (March to May) and higher in the summer during
El Niflo years. La Nifia reduces heat stress degree-days in June and July. In Florida,
however, these differences are small compared to the total average heat stress degree-
days in the spring and summer.
Departures from normal daily maximum or minimum temperatures associated with El
Nifio or La Nifia are significant in Florida, especially during winter months. Florida can
see average temperatures 2F to 3F below normal during El Nifio years. Temperatures
2F to 4F above normal in winter are likely in La Nifia years. La Nifia's effect on
temperature is more pronounced in north Florida. Martsolf (2002) identified hard freezes
through the last century and found that most occurred during Neutral ENSO phases.
In December to April (winter and spring months), average daily maximum temperatures
are higher than normal in La Nifia years, and lower than normal in El Nifio years through
most of the state. La Nifia effects on winter temperatures generally increase toward the
north within the state. In southern Florida average daily minimum temperatures in the
June to August (summer) tend to be lower than normal in La Nifia years. Lower nighttime
temperatures potentially improve yield of some crops.
The valuation of climate impacts on agriculture, especially in Florida is a less researched
field than the oceanographic and atmospheric phenomena that lead to climate change.
Contributions have been made by different working groups directly or indirectly related
to the Florida Consortium (recently re-named Southeast Climate Consortium). The cost
of decreased yields of major Florida crops attributable to climate change has been studied
by Hansen et.el 1998, 1999a, 199b; and Letson et al. 2001. Messina and others (1999)
investigated optimal land allocation in the Argentine humid Pampas during different
ENSO phases. They found that tailoring land allocation to different crops could result in
increased net farm income of between US$5 and US$ 15/ha/year. We attempt to build
upon this incipient yet valuable body of knowledge, by adding another realm of
production, i.e. beef cattle in Florida, as another area studied as to how ENSO phases
affect production systems economically. By using several scenarios of combinations of
ENSO phases, we attempt first to understand how climate change interacts with grass
production, and then what management decisions managers may make to reduce risk and
vulnerability and perhaps enhance profits. A longer term objectives of these types of
study is to develop user-friendly decision support systems for the agricultural production
2. Materials and Methods
2.1. Interacting with the clientele: Sondeos
Hansen (2002) pointed out the usefulness of distinguishing between descriptive and
modeling approaches to evaluating decision responses to a given forecast. He also
suggested that the best elements of both approaches can be combined. The process
described in this paper combines descriptive work, where data are elicited from livestock
producers, and modeling (normative or prescriptive) methods. The Florida Consortium,
made up of researchers from the University of Florida, Florida State, and the University
of Miami, has outlined a framework for assessing the potential use of climate forecasts in
agricultural decision-making. One of the fundamental objectives of the consortium is to
reach out to multiple stakeholders in Florida (Jagtap et al. 2002). Four Sondeos were
conducted from March 1999 through March 2001. The Sondeo (Hildebrand, 1981, 1986,
1999) is a team survey process that was developed to provide information rapidly and
economically about agricultural practices in order to guide strategy in agricultural
development programs. It is structured around a series of conversational interviews
between the team and farmers. It is a multidisciplinary process from data collection
through report writing with teams ideally including people from the social and
agricultural sciences. In a Sondeo, data are shared among the different teams and report
writing is done as a group so that observations are confirmed, debated and analyzed with
members of other teams. The results may be quantified or not but the accuracy of the
findings is strengthened by the cross-checking process (Cabrera et al., 2000 after
Hildebrand 1981, 1986.)
In all, 38 ranchers and 41 extension agents were interviewed. Three of the Sondeos were
conducted in conjunction with a graduate class at the university of Florida (AGG 5813,
Farming Systems Research Extension Methods), and one was conducted with graduate
students at UF hired specifically for the purpose. An enormous wealth of qualitative as
well as quantitative data was gathered using this technique at very low cost.
The first Sondeo, conducted with Florida Cooperative Extension Agents, identified
livestock producers and the forest industry as potential users of improved climate
forecasts. The following three Sondeos were conducted with different types of cattle
producers, with the aim of understanding their production systems, and ways in which
improved climate forecasts could be used to improve management decisions. Methods of
delivering improved climate information were also explored during the Sondeos.
Findings common to all four Sondeos were summarized in a report, which revealed that:
"most ranchers typically plan for the worse case scenario (La Nifia)1.
They prefer to produce enough feed to support their herds throughout the
year. If they can do this, they would not have to alter stocking rates based
on long-term climate forecasts. This production, however, is difficult in
drought years. Ranchers indicated there were some potential benefits to
improved forecasts. These include when and if to plant cool season
grasses, and rate of seeding; rate and timing of fertilizer application;
quantity of hay needed for winter; need to purchase bulk feeds and
nutritional supplements; when or if to ship cattle to another region;
adjusting stocking levels during the winter and spring months, and
anticipating market conditions." (Davis et al., 2001 unpublished, Jagtap et
2.2. Beef Production in North Central Florida: livestock production system
Florida is a major producer of feeder calves. Approximately 39% or four million acres in
Florida are used as pasture and rangeland (Agfacts, 2000.) Together with woodland
(20%) and cropland (35%) cattle production is a mainstay of the Floridian economy.
Cattle products account for approximately 500 million dollars in annual sales (Mislevy
and Quesenberry, 1999.) Returns on investment are typically low in this industry.
Although economies of scale are important for profitability, there remain a large number
of small (10-99 head) operations in North Central Florida. This is often due to a life-style
choice rather than a search for profitability.
Stochastic events associated with prices and climate often lead producers to evaluate their
economic circumstances. Following a harsh winter throughout Florida in 1995-6, which
resulted in a large number of thin cows and potential for reduction of 1996 calving rates
and 1997 pregnancy rates, feed prices soared to record-high levels; calf prices dropped
concurrently. There was a significant liquidation of cowherds, and potential replacement
heifers were placed in feedlots (Wade, 2000; Lemaster, 1999).
Producers attempt to critically analyze their costs and evaluate options to minimize
losses. Historically, however, the cow-calf business in Florida is a break-even
proposition (Reiling and Lemaster, 1998). Because much of the variability in production
and profits in livestock production are associated with climate, we hypothesized that
SAlthough ranchers do not necessarily know that a drier than normal winter is caused by and called by a La
Nifia event in the Pacific Ocean, they are aware that two or three times per decade, a winter drought occurs.
The amount of cattle that can be carried on their land during these bad winters is what usually determines
stocking rate for subsequent years. Thus livestock producers can be said to plan for worse-case scenarios.
ranchers could reduce their risks and increase profits by using currently available
methods for forecasting climate to adjust various decisions.
Cow-calf production in Florida involves keeping breeding females and raising calves
born to them each year. Calves are weaned at 6-8 months and often sold to a feeding
operation until slaughter. Typically Florida stockers go to feedlots in Arizona and the
Texas Panhandle. Weaned calves weigh between 182 and 273 kgs when sold. Calves
can also be over wintered and sold in the spring. Although buying weaned calves and
bringing them up to slaughter weight with feed exists as a system in Florida, by far the
most prevalent beef cattle production system is the cow-calf operation.
The objectives of this research were to determine which livestock management practices
might be changed if future climate conditions are known and to estimate the potential
value of using ENSO-based climate forecasts to improve beef cattle ranch management
decisions in North Florida.
2.2.1. Resources required
Typically, 0.8 ha of grazing land is required for each cow-calf pair. When cattle present
weights beneath certain benchmarks it is a sign to cattle producers that the pastures are
stocked too heavily. Because stocking rate is a principal determinant of economic
outcome, and the amount of land available is limited to the property plus whatever
amount can be leased profitably in a given season, land is a major constraint on
production and profits. A hypothetical 160 ha ranch considered to be representative of
many ranches in the North Central Florida area, was modeled in our study. Because El
Nifio/La Nifia events affect rainfall, temperatures, and evapotransipiration, they affect
pasture growth and thus stocking rates.
Running fences and checking cattle are the only constant labor a cow-calf operation
needs. Winter-feeding and health treatments are occasional labor inputs. Although
quality labor is often difficult to obtain at the right time, overall availability does not
usually limit the cattle production system in Florida.
Facilities include start-up infrastructure (one-time expenses) such as pastures ($110-
330/ha), fencing, corral, a squeeze chute, loading ramp, feed and mineral boxes and
automatic water sources. Yearly costs include fertilizer, depreciation for machinery, land
use, supplemental feeds including protein in the form of liquids or pellets, mineral blocks,
hay or silage, molasses, and manufactured feeds; veterinary attention and medicine,
insurance, taxes, maintenance and repair of buildings, interest on loans and miscellaneous
supplies and repairs. Price per head for the average 136-227 kg. calf varies typically
between $1.65 and $2.20 /kg.
184.108.40.206. Pasture establishment and upkeep
Bahiagrass (Paspalum notatum) is the most widely planted grass in Florida. It is well
adapted, easy to establish, relatively productive, drought tolerant and has the advantage
of being planted by seed. The model assumes pastures are already planted.
Winter annual grasses include ryegrass (Lolium multiflorum), and the small grains wheat,
rye and oats. These are planted in the fall and can provide grazing in late winter and
spring. They are either planted into a prepared seed bed or bahia sod after it stops
growing in late October. Grain rye is the most resistant to dry weather while ryegrass
takes excess water best and oats probably make the best hay. Rye and ryegrass mixtures
are very often used in North-Central Florida for longer winter grazing and adaptation to
unpredictable weather. For this reason, the rye-ryegrass mix was included in the model
as an option for winter.
Feed (all forms combined) makes up the largest cost in cattle operations and has a great
impact on both reproductive performance (% calf crop) and weaning weight. The winter
feeding period in Florida may be as long as 120 to 140 days and may account for more of
the actual feed costs than grazing for the remainder of the year (Stanley, 1995.)
In winter, the ranchers' objective is to minimize feed costs while providing adequate
nutrition. They must also decide whether to feed their stockers through the winter or sell
them at weaning. The choice of feed and feeding system depends on local conditions. In
general however, options are:
Planting cool season grazing crops (ryegrass, wheat, oats, rye)
Maintaining cattle on bahiagrass at a lower stocking rate in winter
Preparing or buying hay during the previous summer
Purchasing full feed concentrate. Good quality calves usually gain around 0.37
kg per day on these feeds. This rate of weight gain is considered constant in the
LP model in this study.
Market prices are often depressed by the large number of calves sold in late summer and
early fall, and some cattlemen retain ownership of their calves with the objective of
improving net sales income by added calf weight and by the probability of better prices
by selling at times when fewer cattle are marketed. Substantial economical calf gain is
needed to make retained ownership profitable (Pefia, 2000; Braswell, 1992). While
feeder calves are the most important output from cow-calf operations, sales of cull cows
typically represent 15-20% of total revenues to a cow-calf producer (Spreen and
Simpson, 1992.) Our model sells calves before winter at $320 and stockers after the
winter at $520. Prices can be easily altered according to changing circumstances through
user-friendly macros in using Microsoft Visual Basic and Excel.
"The job of the ranch or farm manager is to develop and coordinate programs which have
the best chance to result in a profit from a given set of resources. This job is best
accomplished by advanced planning for the future with the plans based on reliable
records from previous years of experience (Hentges 1977)." Reliable climate forecasts
run through trustworthy models may aid in decision making. Sondeos described above
also suggests the need to address the probabilistic nature of climate forecast and deliver
them to ranchers in a manner that is clear, timely and user-friendly.
3. The Model
3.1. LP model of livestock ranch
Historically, LPs have been used to study the role of cattle in farming systems. Boggess
et al., (1979) reported findings in which it was necessary to force a cow-calf herd into the
final optimal solution. Melton et al., (1980) took a critical look at the above and many
other models and concluded that the whole-farm LP models were inadequately
representing the reality of cow-calf production. More recently, several studies on cattle
ranching and sustainability have been published(Griffith and Zepeda 1994, Bouman and
Nieuwenhuyse 1999, Kaya et al. 2000).
In the Farming Systems program at the University of Florida researchers apply novel
interaction and participation techniques to elicit first-hand data from producers. This has
resulted in both Ethnographic Linear Programming and the initial applications of
Participatory Linear Programming (Hildebrand 2001, Bastidas 2001, Kaya, 2000,
Breuer, et al. forthcoming). While both these analytical tools were designed primarily for
the analysis of small farm livelihood systems (Hildebrand 1986, Kaya 2000) many
elements belonging to them have been used successfully to overcome difficulties
associated with modeling cow-calf systems from secondary data. Sondeos, in-depth
interviews and participatory model calibration were all used for this study.
For the current study, a two-year plus third summer, dynamic linear program was
developed for the conditions of north central Florida cattle ranch operations. A
hypothetical 160 ha cow-calf production unit was simulated. The model included calves,
stockers, heifers, and cows; their pasture requirements, and their connections with climate
conditions. The model was constructed on an Excel spreadsheet, which was connected
with user-friendly Visual Basic forms that allow easy and fast interactions.
Coefficients from secondary data were improved by interacting with local ranchers in
February 2002 using a modified Participatory Linear Programming process.
There are 20 Activities (Xi) for each year in the model. They include livestock, pasture or
hay production as well as selling and buying activities, hiring labor and transfer animals
from one account to another. The year is divided in appropriate livestock seasons:
Summer and Winter.
Cows on bahia X,
Transfer calves X2
Heifers on bahia X3
Sell calves end Summer X4
Sell cullcows end Summer X,
Sell cows end Summer X6
Hired labor X,
Make hay for Winter X,
Buy hay for Winter X9
Heifers on ryegrass X10
Heifers on bahia XlI
Cows on bahia, hay and molasses X12
Cows on bahia and citruspulp X13
Cows on oats and ryegrass X14
Sell stockers end Winter X,
Replacement calves end Winter X,
Buy cows end Winter X,
Buy hay for Winter X18
Buy citruspulp X19
Hired labor X20
In addition, there are yearly fixed activities to retire money from the farm for family use
X21 for year one and X22 for year two.
Credits are also activities included in the model for the Winter seasons: X23 and X24.
Finally, there are nine cash transfer activities that move the money from one season to the
next: X25to X33.
Each year there are 17 constraints, nine in Summer and eight in Winter. With the
exception of the fixed land acreage constraint (160 ha) the others are tracking constraints
that keep the money flow throughout the model. All of them, with the exception of the
last money transfer, are less than or equal to zero constraints. The last constraint is
greater or equal to zero and is the maximization objective function.
The mathematical model can be summarized, for each year, as:
Activities: X, i= 1 to 33
Constraints B j= 1 tol7
Objective function: Max H = XC,
Model Constraints: ajX, BJ and X, > 0
3.1.3. Client Participation in model calibration
The LP model was tested with real data through farm interviews in the winter of 2002,
and proved to be trustworthy. A wide range of ranch types, sizes and resource
endowments, were represented. Livestock systems were relatively homogeneous
regarding production practices. Extension Agents from three north central Florida
counties (Marion, Alachua, and Levy) provided names of several beef cattle producers in
their districts. Interviews were arranged by phone and the research team visited the
producer in situ with the model on a laptop computer. Farm-specific data were re-
entered into the model through input tables. After running the nine basic scenarios,
outputs were discussed with ranchers. After each visit, the model was modified and
prepared for the next ranch visit. These interviews were continued until the model
satisfied the team, and the clients. This methodology, Participatory Linear Programming,
could be promoted in future work of this type.
3.2. Crop model component
Regionally adapted and tested crop simulation models prove useful to study variation in
yields in response to climate variability (Boote et.al., 1996; Boote et.al., 1998). The
biophysical nature of these models allows us to translate climate forecasts into
agricultural outcomes (Phillips et al., 1998). Combined with extended weather series
these models can be used to evaluate decision capacity provided that the decision
variables can be handled by the models (Jones et al., 2000). Models exist for many crops,
such as maize, soybean, peanuts, rice, and wheat (Jones et al., 1998). However, we did
not have access to comprehensive models for the pasture species considered in this study
(winter rye grass (Lollium multiflorum), rye (Secale cereale), and bahia grass (Paspalum
notatum)). Thus, a well-established relationship between dry matter production of plants
and crop water use was used to estimate pasture yield reductions under conditions of
limited rainfall. This relationship was first described by de Wit (1958) by relating crop
yield (dry weight) to the ratio of crop transpiration to evaporation of water from a free
water surface. The theoretical basis for this relationship is that under water stress
conditions, stomata on plant leaves close thereby reducing water vapor loss from leaves
and proportionally reducing photosynthesis (C02 assimilation, thus dry matter
accumulation). Since then, various authors have produced modifications of this
relationship for use in research on irrigation and on estimating yields under water deficits
(e.g., Hanks, 1958; Doorenbos and Kassam, 1986). The most widely used form of this
relationship is given by Doorenbos and Kassam (1986) as,
(1 )= Kr (1 -) 
where Ya is actual yield under water deficit conditions, Ym is maximum yield under non
water-limiting conditions, ETa is actual evapotranspiration, ETm is maximum or potential
evapotranspiration of the crop when water is not limiting, and Ky is an empirical yield
response factor. When we are interested in dry matter production instead of grain yield,
the Ky value is approximately 1.0. For example, Doorenbos and Kassam (1986)
presented values of Ky for alfalfa ranging from 0.7 to 1.1. When the ground is shaded by
plants, as is the case most of the time in pasture production, the soil evaporation
component of evapotranspiration is small and stomatal closure would proportionally
affect ET and dry matter growth.
To use this equation to predict effects of limited water on pasture production, we needed
to compute ETa and ETm, have a value of Ky for the pastures, and know how much yield
would be under non water-limiting conditions (Ym). We used the DSSAT v3.5 (Jones et
al., 1998) wheat model to compute ETa and ETm for winter pastures (rye and oats) and
the bahia grass model to compute these values for summer pasture. Historical daily
weather data were input into the DSSAT models to compute ETa and ETm and the ratio of
seasonlong (ETa/ETm) for 43 years of historical weather data from Gainesville, Florida
(1958-2000) for use in equation . We did not know how much biomass is typically
produced in each of the seasons and by pasture type, so we did not use Ym in dry weight
units. Instead, we had information on carrying capacity of pastures for each season. Thus,
we used equation 1, assuming a Ky of 1.0 for each type of pasture, to compute Ya in
terms of carrying capacity of the land (in units of cattle/ha). The production coefficients
(ETa/ETm ratio) were then linked to the linear programming model though input tables.
For summer months, bahia grass was simulated whereas a mixture of oats and rye grass
was simulated for winter pastures. Average pasture production for each ENSO phase and
season were computed and used in decision models for studying options for adjusting
management condition on each phase. These average pasture production results were
transformed into livestock carrying capacities for each season (summer vs. winter) and
each ENSO phase. Animal unit and animal unit equivalents (Scarnecchia, 1985) were
used to relate the different classes of cattle to carrying capacity. Relating the ENSO
phase to stocking rate was accomplished by using crop growth curves for bahiagrass in
summer and a mix of rye-ryegrass in winter as follows.
Considering a neutral year for Florida as baseline for bahia grass growth. Production in
an El Nifio year was considered to be Z of neutral production. In an La Nifia year,
bahia grass was calculated to produce X of neutral. Similarly, growth of rye in a neutral
year in Florida was considered to be the baseline. Rye in an El Nifio winter was
calculated to produce Y, and rye in a La Nifia winter was calculated at 0.X of neutral
production. These ratios were translated into carrying capacity in the model. Indexes are
available in Table 1.
When crop production for the grasses is translated into carrying capacity, La Nifia winters
are significantly lower than in neutral or El Niflo years. These results imply that cow-calf
management could benefit by taking into account ENSO phase predictions.
In order to analyze the effects of different combinations of possible ENSO phases,
different scenarios were tested in the 2-year plus one summer model for a hypothetical
160-ha ranch. Overall economic results varied from the worst-case scenario (two
consecutive La Nifia events) to the best-case scenario (two consecutive El Nifio events,
Figure 1). Differences observed were due to the variation in carrying capacity caused by
available pasture as affected by climate conditions, Table 1. Variations were generated
from historical temperature data.
4.1. ENSO effects on pasture production
Table 1. Bahia and Ryegrass responses to different ENSO phases
Indexes Bahia grass summer Rye winter
Nifio 1.050981 1.091682
Neutral 1 1
Nifia 1.042820 0.859967
Bahiagrass can carry approximately 10% more cattle in either El Niflo or La Nifia years
than it can in neutral years (this is why ranchers ordinarily plan for neutral summers; it is
the worse case scenario.) The rye-ryegrass mix on the other hand, carries about the same
number of head in a neutral as in an El Nifio year. In La Nifia winters, carrying capacity
is greatly reduced. By planning ahead according to an expected La Nifia winter
prediction, ranchers could buy hay in the summer, which ordinarily costs about half as
much as what they would pay in winter.
2 Available: www.fawn.ifas.ufl.edu
In the model, the decision to graze a mixture of cool season grasses is greatly influenced
by the prediction of the ENSO phase for the winter period. When a La Nifia is predicted,
the model planted nearly 75% of the 160 ha to rye-ryegrass, whereas for El Nifio a little
less than 30% is planted (Figure 1). This fact has important connotations for economic
outcome because more stockers can be carried over the winter with good weight gains
and receive higher prices when sold in spring. The great variability in herd size is
accounted for mostly in the number of calves carried through winter and heifers during
the summer in the different probabilistic scenarios, Figure 2.
NEUTRAL NIAA NIFO
Figure 1. Proportion of bahiagrass and rye-ryegrass grazed during winter
Figure 3 shows the combined value of gross margin (GM) plus the herd value at the end
of two years for the nine possible ENSO combinations. The worst-case scenario is when
two consecutive La nifia years occur while the best-case scenario happens when two
consecutive el nifio years occur. Results reflect dynamic solutions in which results for the
first year depend on the previous year, and second year results depend on outputs from
4.2. ENSO and economic output, including probability
Changes in economic output are mostly due to changes in the different categories of
cattle carried over in different ENSO phases. In La Nifia situations the model drastically
reduces the amount of calves it carries over the winter. It sells many calves before the
winter at a lower price. In Neutral and El Nifio years the amount of calves carried over is
approximately the same; however in El Nifio substantially more heifers are kept.
Figure 2. Changes in herd size in different scenarios
End Profit 0.4
200000 0 .
185000 I 0
NIN4A NIIIA NINIA NEUTRAL NIFO NEUTRALNEUTRAL NIRO NIIAO
NIIIA NEUTRAL NIFO NINA NINIA NEUTRAL NIFAO NEUTRAL NIRO
Figure 3. Two-year economic output values for nine scenarios tested and probability of
m CALVES WINTER
* HEIFER SUMMER
4.3. ENSO and herd size changes
Figure 4 shows how the 160-ha ranch would change herd size in response to different
ENSO combinations. Herd size changes were calculated from available climate data for
1949 trough 1999. Average herd size calculated over this period for the 160 ha ranch was
248 head (sd=19.86).
NNo .- E..--- ..- -. .- -.--- .. --U---- -- UU --- U ----. ---. ---. --... ...-....
Average number of cows in winter = 150 -WINTER AVERAGE
N 1 & A ... B -- .. ....-. ........- .... .. .......-........-... .............................* ........_ .........-................ .................................
Figure 4. Modeled herd size over 40 years of known climate effect.
4.4. ENSO, ryegrass planting and associated probabilities
Ryegrass planting is a critical activity within the cow-calf operation because this is one of
the best forage alternatives for the winter months. Ryegrass establishment and
production are associated with the ENSO phases. The LP model was solved for the three
different scenarios (El Niflo, La Nifia, neutral). The model suggested, for each scenario,
different herd size and different area of ryegrass planting. With that initial information,
the following possible situations could be tested from the point of view of profitability:
Rancher follows recommendation
Rancher does not follow the recommendation based on ENSO phase
If the rancher follows the recommendation and plants that amount of ryegrass, there are
two possible situations:
Ryegrass is established (correct climate prediction) or
Ryegrass is not established (incorrect climate prediction).
When the rancher follows the recommendation and the prediction is correct, the incurred
costs will be from planting on time and there are no unexpected costs. When the rancher
follows the recommendation and the ryegrass is not established because the prediction
was not accurate, the rancher not only loses the money of planting the ryegrass, but also
he or she needs to buy expensive hay in the winter to maintain the herd. In the case that
the rancher does not follow the recommendations and does not plant any ryegrass,
meaning that he or she buys cheap hay in the summer (preparing for the winter) there are
no unexpected costs. Table 2 and Figure 5 show the difference in end value of herd plus
Gross margin under different ENSO scenario combinations.
Overall, end value is double or more when recommendations are followed and are correct
versus the case where the rancher does not follow recommendations. The worst-case
scenario is when the rancher follows recommendations and establishment fails due to
incorrect climate predictions. These results represent a "perfect case" and do not take
into account probabilities of occurrence or relative strength of a particular El Nifio or La
Table 2: Ryegrass production and risk associated with ENSO predictions
Gross Margin +
cows Ihelfers calves Herd Value
ESTABLISHED correctprediction follow recommendation 155.90 27.43 105.70 11990.55
PLANT RYEGRASS summer buyhay 0.00
307.82 FAILURE wrong predicton follow recommendation 155.90 27.43 105.70 -11367.14
winter buy hay 338.46
NOT PLANT RYEGRASS not follow recommendation 155.90 27.43 70.57 5544.86
summer buyy hay 33.46
W ESTABLISHED correctprediction followrecommendation 155.90 27.43 189.09 22433.97
I PLANT RYEGRASS summerbuy hay 173.49
N 117.21 FAILURE wrong predicon followrecommendation 155.90 27.43 159.00 4275.27
T NIwinterbuyhay 142.68
E NOT PLANT RYEGRASS not follow recommendation 145.57 27.43 122.02 11409.00
R summer buy ay 316.17
ESTABLISHED correct predcion olollw recommendation 155.90 27.43 159.09 28315.31
PLANT RYEGRASS summerbuyhay 63.89
216.80 FAILURE wrong prediction follow recommendation 155.90 27.43 159.09 -3348.10
NEUTRAL winter buy hay 236.01
NOT PLANT RYEGRASS not follow recommendation 129.38 27.43 133.47 8452.40
summer buy hay 299.90
GROSS MARGIN AND VALUE OF HERD, US$
ryegrass management in winter
-o -0 -o -o U)= *0
-a -g -0 a* a*
U I c C 0
Figure 5: Ryegrass production and risk associated with ENSO predictions
Sondeos are a low cost, effective and rapid method for conducting stakeholder analysis.
Through their use, livestock producers were identified as an important target group for
the Florida Consortium's efforts to reach out to clientele to reduce risk and increase
profits. Findings suggest that livestock producers could alter their management
decisions based on ENSO forecasts. Some practical options for tailoring management
were obtained through modeling the livestock production system. Management options
include altering area of rye-ryegrass planted, changing date of hay purchase and adjusting
size of herd to be carried over the winter. ENSO-based climate forecasts could improve
beef cattle ranch management decisions in North Florida by allowing more plentiful
stocking during good rainfall winters, savings accruing from buying hay ahead of time
before a dry winter, and planting rye-ryegrass mixtures only in years when good
production and thus stocking possibilities exist.
6. Limitations of the Study and Further Research
The use of participatory methodologies, linear program models, and crop modeling have
allowed for rapid, inexpensive and more accurate insights into how useful new prediction
technologies available for ENSO phenomena could be for livestock producers. Further
scenarios could be run as need arises in new areas or for different systems. A user
friendly, decision support tool may emerge from further work on the LP used in this
study. Research should be extended geographically to include other areas of the state of
Florida and the southeastern United States. Simultaneous effects of ENSO phenomena in
more distant areas of the country are a topic for further investigation. Among these, corn
and soybean prices, as well as climate conditions in feedlot states may be important.
Above all, ENSO predictions must be delivered in user-friendly ways to be useful to
ranchers. Additional analysis is needed to understand the broader consequences of using
climate forecasts in livestock management, in particular to understand the wider aspects
of value and risks associated with this approach. Future studies should take this into
account, because ranchers are highly concerned about uncertainty and risk, not just
We gratefully acknowledge the National Oceanographic and Atmospheric Administration
Office of Global Programs for supporting the Florida Consortium. We thank the Florida
Cooperative Extension Service, and participating livestock producers.
Bastidas, E. 2001. Assessing potential response to changes in the livelihood system
of diverse, limited-resource farm households in Carchi, Ecuador: modeling
livelihood strategies using participatory methods and linear programming. Ph.D.
Thesis. University of Florida, Gainesville.
Bogges, W., McGrann J., Boehlje, M., Heady, E.O. 1979. Farm level
impacts of alternative soil loss control policies. Journal of Soil and Water
Conservation. 34 (Jul-Aug,1979):177-183
Boote K.J., Jones,J.W., Pickering,N.B. (1996). Potential uses and limitations of crop
models. Agronomy Journal 88, 704-716.
Boote K.J., Jones, J.W., Peart, R.,M., Curry (editors) 1998. Agricultural systems
modeling and simulation. Marcel Hoogenboom Simulation of crop growth:
CROPGRO model. Dekker,Inc., New York .pp 651-692
Bouman, B.A.M., Nieuwenhuyse, A., 1999. Exploring options for sustainable beef cattle
ranching in the humid tropics. Agricultural Systems 59 (2), 145-161.
Braswell, R.M. Jr. 1992. Factors that influence and determine calf prices. Beef Cattle
Short Course. Http:/www.ifas.ufl.edu/-animal/short92/braswell.htm
Breuer, N., 2000. The Role of Medicinal Plants in Rural Paraguayan Livelihoods.
International Journal of Agricultural Sustainability, in review.
Cabrera, V. E. 1999. Farm problems, solutions, and extension programs for small
Farmers in Cafiete, Lima, Peru. M.S. Thesis. University of Florida, Gainesville.
Changnon, S.A., and Kunkel, K.E.,(1999):Rapidly Expanding Uses of climate data
and information in Agriculture and Water resources : Causes and characteristics
of New application. Bulletin of the American Meteorological Society.80:821-830
Davis, K., Drew, M.Payson, K., Warick, J. 2001. "Livestock Producers'
Attitudes Toward Potential Use of Improved Long-Term Climate Forecasting in
North Centra Florida." AGG 5813, FRED, IFAS, Gainesville. Unpublished.
De Wit, C. T., 1958. Transpiration and crop yields. Verslag van Landbouwk. Onderzoek.
No. 64.4, 88 pp. Doorenbos, J., and A. H. Kassam. 1986. Yield response to
water. FAO Irrigation and Drainage Paper 33. Food and Agriculture Organization
of the United Nations, 00100-Rome, Italy. 193 pp.
Downtown ,M.W., Miller, K.A.,1993. The freeze risk to Florida citrus. Part II:
Temperature variability and circulation patterns. J.Climate 6:364-372
Florida Agricultural Facts. 1999. http://wwww.fl-ag.com/agfacts/farmland.htm
Goddard L., Mason S.J., Zebiak S.E., Ropelewski C.F., Basher R., Cane M.A. (2001)
Current approaches to seasonal-to-Interannual Climate Predictions Int J Climatol,
Griffith, K., Zepeda, L., 1994. Farm level trade-offs of intensifying tropical milk
production. Ecological Economics 9 (2), 121-133.
Hammer G.L., Hansen J.W., Phillips, J.G., Mjelde, J.W., Hill H.S.J., Love A, Potgieter
(2001). Advances in application of climate prediction in Agriculture Agric sys
Handler, P. 1990. USA corn yields, the El Nifio and agricultural drought: 1867-
1988. Int.J.climate 10:404-411
Hanks, R. J. 1974. Model for predicting plant yield as influenced by water use. Agron. J.
Hansen, J.W., 2002. Realizing the potential benefits of climate prediction to agriculture:
issues, approaches, challenges. Agricultural Systems 74 (2002) 309-330.
Hansen, J.W., Jones, J.W., Irmak, A., Royce, F.S., 2001. ENSO impacts on crop
Production in the southeast US. Impact of climate variability on agriculture.
qAmerican Society of Agronomy Special Publications 63 (2001), pp. 55-76.
Hansen, J.W., Irmak, A., Jones J.W., 1999. El Niflo Southern Oscilation influences
On Florida crop yields. Soil and Crop Science Society of Florida Proceedings,
Hansen, J.W., J.W. Jones, C.F. Kiker, A.H. Hodges. 1999. El Nifio Southern Oscilation
Impacts on winter vegetable production in Florida. Journal of Climate. 12: 92
Hansen. J.W. A.W. Hodges, and J.W. Jones. 1989. ENSO influences on agriculture in
The southeastern United States. Journal of Climate 11:404-411.
Hentges, J.F., Jr. 1977. Management of the Beef Cattle Ranch or Farm. In. Beef Cattle
In Florida. Bulletin No. 28. Florida Department of Agriculture and Consumer
Hildebrand, P.E., Caudle, A., Cabrera, V., Downs, M., Langholtz, M., Mugisha, A.,
Sandals, R., Shriar, A., Veach, K. 1999 Potential Use of Long-range Climate
Forecasts by Agricultural Extension Advisors in Florida: A Sondeo Report
(Staff Paper SP 99-9). Food and Resources Economics Department, University
of Florida, Gainesville
Hildebrand, P.E. 1986. Perspectives in Farming Systems Research and Extension.
P. E. Hildebrand, Editor. Lynne Rienner Publishers, Boulder.
Hildebrand, P.E. Combining Disciplines in Rapid Appraisal: The Sondeo Approach.
Agricultural Administration 8. 1981: 423-432.
Jagtap, S.S., Jones, J.W., Hildebrand, P., Letson, D., O'Brien, J.J., Podesta, G., Zierden,
D., Zazueta, F. 2002. Responding to stakeholder demands for climate
Information: from research to applications in Florida. Agricultural Systems 74 (3)
Jones, J.W., Hansen, J,W., Royce, F.S. Messina, C.D. (2000). Potential benefits of
Climate forecasting to Agriculture, Ecosystems and Environment, 82:169-184
Jones, J. W., G. Y. Tsuji, G. Hoogenboom, L. A. Hunt, P. K. Thornton, P. Wilkens, D. T.
Imamura, W. T. Bowen, Singh, U., 1998. Decision support system for
agrotechnology transfer: DSSAT v3. In Tsuji, G. Y., G. Hoogenboom, and P. K.
Thornton (eds.), Understanding Options for Agricultural Production. Kluwer
Academic Press, Boston. pp. 157-177.
Kaya, B., Hildebrand P.E. and Nair, P.K. 2000. Modeling changes in farming systems
with adoption of improved fallows in Southern Mali. Agricultural Systems 66.
Kiladis,G.N., and W.A.Wallis. 1989.Global climatic anomalies associated with extremes
in southern oscillation. J.Clim. 2:1069-1090
Lemaster, W. J. 1999. What is the cost of Production in a forage based system?
Proceedings of the 16th annual Florida Cattlemen's Institute and Allied Trade
Letson, D., Hansen, J., Hildebrand, P.E., Jones, J.W., O'Brien, J.J., Podesta, G. P.,
Royce, F.S., Zierden, D.F. 2001. Florida's Agriculture and Climatic Variability:
Reducing Vulnerability. The Florida Geographer, vol. 32, pp. 41-44.
Mislevy, P, Quesenberry, K.H. 1999. Development and Short Description of Grass
Cultivars Released by the University of Florida (1892 1995). Soil and Crop Sci.
Soc. Florida Proc. 58:12-19 (1999.)
Martslof, J.D. 2001. Relationship between Florida Freeze Incidence and Sea Surface
Temperature in the Tropical Pacific. Proc. Fla. State Hort. Soc. 114.
Oram, P.A.,1989. Sensitivity of agricultural production to climatic change, an update.
In:Climate and Food security. IRRI,Manila,The Philippines,pp.25-4
Pate, F.M., Crockett, J.R., Phillips, J.D. Effect of calf weaning age and cow
supplementation on cow productivity. J. Anim. Sci. 61:343-348
Pefia, J.G. 2000. With Calf Prices at Near Record High, Retained Ownership
Enterprises will Require Careful Planning and Monitoring. Ag-Eco News.
Extension Economist Management. Vol. 16, Issue 32, November 1, 2000
Peterson, G.A. 1955. Selection of maximum profit combinations of livestock enterprise
and crop rotations. Journal of Farm Economics. 37 (Feb,1955): 546-54
Phillips, J.G., Cane, M.A. Rozenzweig, C.(1998) ENSO, seasonal rainfall patterns and
simulated maize yield variability in Zimbabwe. Agricultural and Forest
Reiling, B., Lemaster, J.W., 1997. Producing the Calf- How much does it really cost?
Florida Beef Cattle Short Course, University of Florida, Gainesville.
Ropelewski,C.F., Halpert, M.S..1986. North American precipitation and temperature
patterns associated with El nino-southern -oscillation (ENSO). Mon.weather Rev.
114:2352-2362 Ropelewski,C.F., and M.S.Halpert.1987. North American
precipitation and temperature patterns associated with El niflo/southern oscillation
(ENSO). Mon.weather Rev. 115:1606-1626
Ropelewski, C.F., Halpert, M.S..1996. Quantifying southern oscillation relationships
Scarnechia, D.L. The animal unit and the animal unit equivalent concepts in Range
Science. Journal of Range Management 38(4), July 1985.
Stanley, R.L., Jr. 1995. Winter Grazing Options for Florida.
Wade, M. 2000. Outlook 2001. Florida Food and Resources Economics.
IFAS-Cooperative Extension Service, No. 146. November-December, 2000.