|
|
![Contact Us -- webmaster[at]sidney.ars.usda.gov](../../images/l_contb4.gif){kind=small}
Appendix A: How Hopper Works and Why
By Larry Zaleski
Why You Should Know How Hopper Works.—You should know how Hopper works to help accomplish your treatment responsibilities skillfully and accurately.
Whether you’re a rancher or government official, professional and financial considerations demand that you work skillfully and accurately. Applying pesticides when not needed may threaten the environment and waste money. Conversely, failure to apply pesticides when conditions warrant may jeopardize native rangeland and potentially threaten the local ranching economy.
Hopper helps you decide objectively whether to treat or not. But you must use Hopper correctly for good results. And to use Hopper correctly, you must know how the program works.
What You Should Know.—You should be familiar with the following:
•
How Hopper
can save time, improve accuracy, and save money
•
What
the economic research shows
•
How
Hopper’s components work together
•
What
the expert system (Consult) does
•
What
the forage model (RangeMod) does
•
What
the grasshopper model (HopMod) does
•
What
the economics model (RanchMod) does
•
Your
role
As you become familiar with Hopper, you will become more knowledgeable about treatment technology, rangeland ecology, and ranching economics.
How Hopper Can Save Time, Improve Accuracy, and Save Money.—Hopper saves time, improves accuracy, and saves money by
• Automating
expensive and time-consuming tasks, and
• Using
ecological and economic information previously unavailable to decisionmakers.
Automating Expensive and Time-Consuming Tasks.— Hopper automates many tasks that require time, money, VI.2–18.and personnel to accomplish. You still collect information about local conditions, but with Hopper, your treatment decisions are greatly improved with little additional effort. To understand the value of automation, you should know
• What Hopper
does automatically,
• How
Hopper automates tasks, and
• How
automation improves treatment decisionmaking.
What Hopper Does Automatically.—Hopper automatically
• Estimates
the average instar of a grasshopper population (for integration
with field data);
• Estimates
the effects of precipitation on forage production;
• Estimates
forage production and, then, forage loss to grasshoppers;
• Chooses
treatments based on local conditions; and
• Determines
if treatment is cost effective.
Without automation and computer simulation, many of these tasks are impractical or more likely to be completed with errors.
How Hopper Automates Tasks.—Hopper automates tasks by integrating an expert system with simulation and economic models (Berry et al. 1991, 1992). Hopper’s expert system is rule-based. Rule-based expert systems are computer programs consisting of rules. These rules are the same as those used by human experts, but the expert system uses the computer’s ability to apply logic, instead. For example, an expert system program for reacting to a traffic light might look like this:
IF THE LIGHT IS RED: Stop and wait for the light to turn green.
IF THE LIGHT IS GREEN: Go on.
IF THE LIGHT IS YELLOW: Slow down, and...
— If the light turns red, stop, wait for it to turn green, then go on.
— If the light doesn’t turn red, go on.
The computer runs through the program until it encounters an “if statement” that matches the current condition. Then the program follows the programmed procedure. Hopper’s expert system works similarly, but it’s designed to select treatments. Hopper asks questions, matches your answers to its rule base, then lists treatments accordingly.
Models, on the other hand, are mathematical formulas that imitate events in the real world. Models allow you to make predictions and estimates about events in the real world. Previously, such models were too time-consuming and complicated for everyday use. Only scientists could use them. But the personal computer has changed that.
How Computer Automation Improves Decisionmaking.—Computer automation improves decisionmaking in two ways. First, automation is comprehensive. That is, Hopper requires that you answer questions needed to make accurate decisions, each time. Critical factors, including those you might not ordinarily consider, are routinely considered. Without this prompting, you might ignore some factors to save time or because you don’t know how to evaluate them.
Second, automation is consistent. It’s consistent because users answer critical questions each time and because Hopper evaluates data the same way each time—something that people seldom do. Consequently, two people independently entering the same data into Hopper achieve the same results each time. Thus, Hopper transforms treatment decisionmaking into a more objective and scientific process.
Simulation, completeness, and consistency result in improved accuracy at roughly the same cost.
Using Ecological and Economic Parameters Previously Unavailable to Decisionmakers.—Hopper achieves improved accuracy because it uses parameters and variables that were previously impractical or unavailable (Davis et al. 1992). Even though these parameters were important, they were often not used because they were too costly and time consuming to obtain or because they could not be analyzed fast enough to help. As a result, treatment decisions were based on partial information.
Recently, however, researchers have shown that many of these unused but critical variables can be simulated mathematically. Other variables have been determined by the Grasshopper Integrated Pest Management Project and cooperators.
Before Hopper, Treatment Decisions Were Based on Less Extensive Information.—Hopper estimates critical variables previously unavailable to decisionmakers. Biologists and economists knew these variables were important, but only well-funded research projects could collect and analyze the data. And the results of their analysis usually came too late to help.
But the economic basis for control of grasshoppers on rangeland depends on several variables, not just grasshopper density (Davis et al. 1992). These critical variables include
• Rangeland
productivity,
• Soil
moisture,
• Livestock
prices,
• Accessibility
and cost of alternative forage,
• Effectiveness
and timing of treatments, and
• Grasshopper
numbers and composition.
These variables, however, are difficult and expensive to measure. Many could not be analyzed quickly. And few scientists, ranchers, or government officers could measure and interpret all of the variables. Consequently, no one could integrate the critical variables into a practical decision support system.
Critical Variables Can Be Estimated Mathematically.—Recently, researchers demonstrated that many critical variables could be estimated mathematically (Berry and Hanson 1991, Berry et al. 1995, Dennis et al. 1986, Kemp and Onsager 1986). Therefore, for some variables, mathematical simulation provides an alternative to sampling and measurement.
When combined with a personal computer, mathematical simulations provide quick, reliable estimates of difficult-to-measure variables. For the first time, critical variables are routinely available to decisionmakers. What’s more, estimated variables can be combined with economic calculations to determine if treatment is cost effective.
What the Economic Research Shows.—The economic research reveals three key facts (Davis et al. 1992):
1. Decisionmakers
should use an economic threshold as their basis for applying treatment.
2. Economic
justification for grasshopper control programs depends on several
variables, not just grasshopper population density.
3. Economic
justification for grasshopper control programs varies from place
to place and year to year.
Decisionmakers Should Consider Economic Threshold in Their Decision About Applying Treatment.—Economics is a primary justification for treating grasshopper infestations. So ranchers should treat grasshoppers not to reduce their numbers but to improve the profitability of the ranch. Reducing grasshopper numbers is only a tactic for managing the rangeland resource.
From a ranching perspective, even rangeland management— a continuous effort which some use as a justification for grasshopper control—is simply an economic endeavor aimed at preserving rangeland productivity. Preserving productivity preserves profit. To illustrate the profit motive: one way to manage the land and prevent range damage during a grasshopper outbreak is to remove cattle. But this option is unprofitable, so ranchers tend to avoid cattle removal when possible. Generally, ranchers seek more profitable alternatives.
Environmental factors are important, too, and may prevent treatment. But in most cases, the basis for your decision to treat or not is economic.
To apply an economic threshold to treatment decisions confidently, you need to understand the concept of the economic threshold and the concept that treatment is an investment.
The Economic Threshold.—The economic threshold is the population density of a pest at which the cost of management intervention equals the resulting benefit from controlling it. The economic threshold varies with the benefits and the cost of treatment (Davis et al. 1992).
When Does Treatment Become Profitable?—The economic threshold is reached when the benefit–cost ratio equals 1 or more.
Hopper determines the economic threshold by calculating the benefits of treatment, then dividing the benefits by the cost. This measure is called the benefit–cost ratio (BC):
|
BC = |
Benefits |
|
Cost |
When the benefits
equal the cost, the ratio is equal to 1 and the economic threshold
is achieved. For example:
Benefits of treatment = $5,000
Cost
of treatment = $5,000
|
BC = |
Benefits |
= |
$5,000 |
= 1 |
|
Cost |
= |
$5,000 |
BC’s greater than 1 are profitable, but BC’s less than 1 are unprofitable. The economic threshold (BC = 1) is the break-even point.
The cost of grasshopper control includes wages and the cost of chemicals, baits, and equipment. The benefit of grasshopper control, on the other hand, is equal to the value of the forage saved by treating grasshoppers.
Treatment Is an Investment.—Treatment is an investment in the agricultural economy. You apply treatment to attain or improve profitability.
Typically, you expect a return on your investments. For example, if you invest $100 in a savings account, you expect to collect interest, which is a return. If the account pays 5 percent simple interest, then after a year you would have $105. The BC of your account would be $105 ¸ $100 = 1.05. Because the BC is greater than 1, the account is profitable.
You would never knowingly invest in a savings account that loses money (an account whose BC is less than 1). Investing when the BC is less than 1 is unprofitable, and thus, economically unjustified.
Treatment, too, should show a return. Treating when the BC is less than 1 is unprofitable, and thus, economically unjustified.
Variables Affecting Economic Justification of Grasshopper Control Programs.—At least seven variables determine the economic justification for grasshopper control programs on rangeland:
• Rangeland
productivity and composition,
• Precipitation
and soil moisture,
• Accessibility
and cost of alternative forage,
• Effectiveness
of treatment,
• Cost
of treatment,
• Timing
of treatment, and
• Grasshopper
population density, life stage, and species composition.
Put simply, these variables determine the value of the forage grasshoppers eat (the damage grasshoppers cause) and how much damage can be prevented. The interaction between critical variables is complex.
For example, if rangeland produces too much or too little forage, you cannot economically justify treatment. If excess forage is produced, there is enough to feed both grasshoppers and livestock, so you cannot justify treatment. On the other hand, if too little forage is produced, there is no forage to protect, so again, you may not be able to justify treatment purely based on forage value.
Consequently, the effects of the variables below assume that there is forage to protect, but not too much or too little. Otherwise, some of the following information would contradict. In practice, Hopper accounts for the effects of forage production automatically.
Rangeland Productivity and Composition.—On highly productive rangeland, you can economically justify treatment at lower grasshopper population densities than you can on less productive rangeland (Davis et al. 1992). This is true because treatment saves more forage per acre on highly productive rangeland.
The more forage you save per acre, the lower the cost per unit of forage saved and the greater your benefit for a given per-acre treatment cost. Consequently, on productive rangeland, you can treat fewer acres and still get the same per-acre benefit. The fewer acres you treat, the lower the cost.
In addition, some forage species are more valuable than others. Generally, the more valuable the forage, the easier it is to justify treatment.
Precipitation and Soil Moisture.—During dry years, you can economically justify treatment at lower grasshopper population densities than in years of normal or high precipitation.
Precipitation is the most important factor affecting rangeland productivity (Berry et al. 1991). Obviously, if it doesn’t rain or snow, forage won’t grow. When forage is scarce, its value increases because you must supplement it by buying hay or leasing additional land. Remember that, although the value of the forage may increase in dry years, the amount that will be protected by controlling grasshoppers is reduced. Hopper considers both of these factors.
In contrast, during normal and wet years, when forage is plentiful, there is often enough forage to feed both livestock and grasshoppers—even at high grasshopper population densities.
Hopper evaluates the effect of precipitation by calculating soil moisture.
Accessibility and Cost of Alternative Sources of Forage.—When alternative sources of forage are expensive or inaccessible, you can justify treatment at lower grasshopper population densities than when prices are low and forage accessible. This is true because when alternative sources of forage are expensive, you pay more to supplement or replace your existing forage. Therefore, your existing forage is worth more, and you can justify paying more to protect it.
Effectiveness of Treatments.—Other things being equal, when treatment is highly effective, you can justify treatment at lower grasshopper population densities than when treatment is ineffective. The more effective treatments are, the greater their value, and the higher the benefit–cost ratio.
Cost of Treatment.—When treatment is inexpensive, you can justify treatment at lower grasshopper densities than when treatment is expensive. Several factors influence the cost of treatment, including the price of pesticides, biological control agents, equipment, and personnel. In addition, the cost of treatment varies with demand. In years with lots of spraying, sprayers demand higher fees. Clearly, you need higher grasshopper densities to justify treatment at $4 per acre than you do at $2.25 per acre.
Timing of Treatment.—Timing influences the effectiveness and value of treatment. If you treat too early or too late, you reduce effectiveness. If you treat too early, many grasshopper eggs are still unhatched and will be unaffected. And if you treat too late, the forage is already eaten and next year’s eggs are laid. In either case, the benefits are reduced.
Grasshopper Population Density and Composition.—Clearly, you can more readily justify treatment at higher grasshopper population densities than you can at lower grasshopper population densities. The higher their population density, the more forage grasshoppers eat. If the grasshopper density reaches the economic threshold, then grasshoppers literally eat up your profits.
In addition, species composition is important. Some grasshopper species do more harm than others. You can justify treating more-harmful species at lower densities than less-harmful species.
But as you’ve seen, several factors, in addition to grasshopper population density and composition, determine the economic threshold.
The Economic Justification for Grasshopper Control Varies From Place to Place and Year to Year.—Because the variables affecting the cost effectiveness of treatment vary from place to place and year to year, the economic justification for grasshopper control varies, too.
Conditions vary from place to place. For example, one pasture is more productive than the next, or one county has normal precipitation, while another is dry. Consequently, you may treat grasshoppers profitably at 1 location when densities reach 18 per square yard but not at another location until they reach 25 per square yard.
VI.2–22.Similarly, conditions vary from year to year. Over time, a ranch may experience fluctuating precipitation, livestock prices, and lease costs. In 1 year grasshoppers may be worth treating at 30 per square yard; the following year, grasshoppers may be worth treating at 20 per square yard.
Normal variation of ranching conditions demands a flexible response to grasshopper treatment. Hopper provides flexibility by accounting for differences in conditions that vary with location and time.
How Hopper’s Programs Work Together.—Hopper uses three kinds of software technology to assist you in making treatment decisions (fig. VI.2–27):
1. An expert
system—to select treatments,
2. Simulation
models—to estimate difficult-to-measure variables, and
3. An
economic model (ranch model)—to determine if treatment is cost effective.
Figure VI.2–27—Overview
of Hopper user interface and internal modules.
These technologies work together to provide decision support. Below is an overview of each class of technology. As each technology is introduced, you’ll learn how it works with the others.
The Expert System.—The expert system (Consult) helps you choose grasshopper treatments as accurately as an expert. It does this by asking questions about the site, giving some of this information to simulation models to estimate grasshopper life stage, evaluating the data against an internal set of rules, and then providing you with a list of suggested treatments appropriate for the situation (Berry et al. 1991).
The Simulation Models.—The simulation models (HopMod and RangeMod) calculate values for critical variables that would otherwise require additional sampling and analysis.
Hopper uses simulation models to estimate the effects of precipitation, forage production, treatment mortality, grasshopper species composition and life stage (Berry et al. 1995). Information from the simulation models is used by the expert system and economics model.
Simulation models allow Hopper to respond to factors that change over time, like grasshopper life stage and forage production (Berry et al. 1991).
The Economics Model.—The economics model (RanchMod) is a linear programming model that does two things. First it determines if treatment is cost effective. Second, it determines which of the treatments listed by Consult is most cost effective. The economics model uses information from the expert system and simulation models to determine a benefit–cost ratio.
Hopper’s models work together to provide reliable decision support. As a result, you can be more confident in your treatment decisions.
What the Expert System Does.—Hopper’s expert system (Consult) provides you with a list of treatments appropriate for the conditions you specify. Consult uses internal rules to decide which treatment(s) to list (Berry et al. 1991). In addition, only treatments approved by the Environmental Protection Agency and the Environmental Impact Statement for the Cooperative Rangeland Grasshopper Program are considered.
Where Consult Gets Its Information.—Consult uses information from three sources. First, Consult asks you the following:
• Location?
• Species
composition?
• Grasshopper
census date?
• Treatment
date?
• Presence
or absence of managed bees?
• Should
treatments harmful to beneficial insects be eliminated from consideration?
• Do
conditions prohibit the use of toxic chemicals?
• Vegetation
thickness?
• Current
weather conditions?
• Percent
of the hopper population already hatched (if known)?
Second, Consult uses Hopper’s own weather model to enter weather data for the site.
Third, Consult uses the grasshopper model (HopMod) to calculate the average life stage at the time treatment will be applied and number of grasshopper eggs that will be deposited during the current year. This allows Consult to decide if it’s too early or too late to treat the infestation economically.
What Consult Does With the Information.—Consult evaluates the information supplied against an internal set of rules. These rules allow Consult to choose treatments appropriate for local conditions.
Consult selects from five treatments approved by the Environmental Protection Agency and the final Environmental Impact Statement for use against grasshoppers on rangeland:
• Acephate
spray,
• Carbaryl
spray,
• Malathion
spray,
• Carbaryl
bait, or
• Nosema
locustae bait (a pathogen of grasshoppers and Mormon crickets).
Depending on the conditions you specify, Consult may recommend none, one, or all of the treatments for economic evaluation. Carbaryl bait, for example, might be recommended when the presence of commercial bees or endangered species prohibit spraying in the area. Nosema locustae may be recommended for use near bodies of water, where chemicals are prohibited.
Consult considers species composition and development in making treatment recommendations because:
• Some species
don’t take baits, so you can’t use baits.
• Some
species won’t eat the predominant local forage, so you don’t
have to control them.
• Some
species develop faster or slower then the bulk of the population,
so you should adjust treatment timing.
By accounting for these factors, Consult can alter its treatment list and, ultimately, the decision whether and when to treat.
What the Forage Model (RangeMod) Does.—RangeMod simulates growth of cool- and warm-season grasses and forbs on rangeland during a single growing season (Berry and Hanson 1991). Important features of the model include the following:
• Forage production
depends on soil moisture and projected peak standing crop.
• Temperature
starts and ends plant growth.
• Forage
production occurs logistically (forming an S-shaped curve).
Forage Production Depends on Precipitation and Peak Standing Crop.—RangeMod determines forage production based on daily precipitation and an estimate of peak standing crop. The model uses either known precipitation averages from nearby cities or precipitation data that you supply. Forage consumption by wildlife is not estimated or considered by Hopper.
|
|
|
Figure VI.2–28—The effect of drying with occasional precipitation on soil moisture content. This pattern is typical for northern latitudes in the West. |
Precipitation directly affects soil moisture, which RangeMod calculates as a function of dry days (consecutive days without precipitation). The model causes soil to dry exponentially (quickly when wet but more slowly as moisture decreases–fig. VI.2–28) down to a minimum of 3 percent by weight. For comparison, the permanent wilting point for plants is reached when soil moisture is 10 percent (Berry and Hanson 1991).
Temperature Starts and Ends Plant Growth.—RangeMod uses a threshold temperature to begin growth in the spring, and to end growth in the fall. The model starts calculating growth when the temperature (the average of the daily high and low) exceeds 32 °F for 5 consecutive days. Growth occurs if daily minimum temperature is above the threshold for the plant type— 44.6 °F for forbs and cool-season grasses, and 50 °F for warm-season grasses (Berry and Hanson 1991).
In RangeMod, temperature is not a factor in forage production except for its role in starting and ending growth (Berry and Hanson 1991).
Forage Production Occurs Logistically.—When graphed, forage production forms a logistic (S-shaped) curve (fig. VI.2–29). The logistic curve simulates forage production in
|
|
|
Figure VI.2–29—The logistic growth of forage appears S-shaped. |
pounds per acre over time. RangeMod simulates forage production for forbs, cool-season grasses, warm-season grasses, and total production, producing a logistic curve for each.
The exact
shape of the logistic curve varies with precipitation and forage
consumption by grasshoppers. Hopper simulates grasshopper forage
consumption in the grasshopper model, HopMod.
What the Grasshopper Model (HopMod) Does.—HopMod determines forage loss caused by grasshoppers and determines the loss that you can prevent by applying treatment (Berry et al. 1991).
HopMod simulates grasshopper development through time. Predicting development is important because the amount of forage eaten by grasshoppers per day varies greatly for each life stage. Early instars eat less than later instars. And because the proportion of each instar in the population changes daily, forage consumption changes daily, too.
HopMod’s simulation of grasshopper development, in conjunction with the forage and economics models, allows you to decide whether or not to treat at a given time in the grasshopper’s growing season.
To understand HopMod, you must understand the following:
• What the
grasshopper phenology (growth and development) model does,
• How
HopMod determines population size,
• How
HopMod calculates forage consumption,
• How
HopMod determines oviposition, and
• How
accurate HopMod is.
|
|
|
Figure VI.2–30—General progression of a grasshopper population during the spring and summer. HopMod begins when the population has peaked and egg hatch has finished. |
What the Grasshopper Phenology Model Does.—Phenology is the study of the relationship between climate and recurring biological events, such as grasshopper lifestage. The grasshopper phenology model estimates the proportion of the grasshopper population in each life stage on any given day as a function of time and temperature (fig. VI.2–30).
A proportion is a percentage divided by 100. For example, the proportion “0.8” is derived as follows: 0.8 = 80 percent ¸ 100. Most people use proportions frequently for various routine calculations.
How Development Is Calculated.—The model determines grasshopper development based on time and temperature, called development time (Kemp and Onsager 1986). Grasshopper development is controlled primarily by temperature, so development time is measured in degree-days.
Degree-Days Are Accumulated Heat.—A degree-day is a measure of accumulated heat. Degree-days accumulate in HopMod when the air temperature is between 40 °F (4.4 °C) and 100 °F (37.8 °C) (Berry et al. 1995).
For example, when the daily minimum and maximum temperatures are between 40 °F and 100 °F, HopMod calculates degree-days like this: If the air maximum temperature is 70 °F and the minimum is 40 °F, then there are 70–40 = 30 degree-days of development.
HopMod averages degree-days over a day–night cycle. The program adds degree-days when the temperature is within the thresholds. HopMod uses a modified sine-wave formula to adjust and accumulate degree-days as the value changes during the day–night cycle. (A sine-wave formula creates a curve similar to the wave pattern you’d see on an oscilloscope. The wave fluctuates above and below a line. In this case, above the line represents daylight; below the line represents night.) In this way, HopMod calculates the average instar, which is displayed in Consult.
When necessary, you can change Hopper’s estimate of the average instar. For example, if you measure an average instar that is different than HopMod’s estimate, you can replace Hopper’s estimate with your measurement, and HopMod will adjust.
Development Is Based on Accumulated Increments of Development Time.—HopMod assumes that the development rate of a grasshopper depends on accumulated increments of development time (Kemp and Onsager 1986). The process is defined as the amount of development time that a grasshopper has accumulated by a given actual time.
HopMod uses Hopper’s weather database to calculate degree-days. Then HopMod calculates grasshopper development for each calendar day of the growing season. The result is a list of proportions for each life stage for each day. For example, on a given day, you might see the following: instar 1 = 0.1 (10 percent), instar 2 = 0.3 (30 percent), and so on. The proportions must add up to 1.00 (representing 100 percent of the grasshopper population) for the day.
How HopMod Determines Population Size.—HopMod gets the grasshopper population size from you. For example, you count 20 grasshoppers per square yard and type in that number. HopMod, however, adjusts over time for natural grasshopper mortality itself.
HopMod calculates average natural grasshopper mortality using a density-dependent model. The larger the grasshopper population, the faster grasshoppers die.
HopMod, however, does not have an egg-hatch model. Consequently, HopMod cannot add newly hatched grasshoppers to the population. The program assumes all eggs have hatched by the census date.
How HopMod Calculates Forage Consumption.—HopMod calculates forage consumption in five steps:
1. HopMod determines the proportion of grasshoppers in each instar (life stage), each day. For example, instar 1 = 0.1, instar 2 = 0.3, instar 3 = 0.4, instar 4 = 0.15, instar 5 = 0.05. Remember, the total must add up to 1.00, meaning 100 percent of the population. The proportions in each instar change each day but always add up to 1.
2. HopMod determines the number of grasshoppers in each instar by multiplying the proportion in each instar by the population density of first grass feeders, then mixed feeders (usually, grass feeders won’t eat forbs, so forbs are protected from grass feeders without treatment). You supply the data on population density and composition.
For example, if the grasshopper population density is 20 per square yard and is 80 percent grass feeders, then—assuming the proportion of instar 2 = 0.4 —the number of grass-feeding grasshoppers in instar 2 is: 20 ´ 0.4 ´ 0.8 = 6.4 per square yard.
3. HopMod determines how much forage each instar consumes by multiplying the feeding rate of grasshoppers in each instar (supplied by Hopper and based on scientific measurement) by the number of grasshoppers in the instar.
4. HopMod determines total forage consumption by adding the consumption of each instar for each day of the growing season. This value is passed to RangeMod and subtracted for each forage type from the amount of forage for the day. If conditions are favorable, forage continues to grow, and forage loss is usually less than the total consumption by grasshoppers.
5. Finally, HopMod repeats the process (steps 1–4) after applying simulated treatments. For example, if there are 20 grasshoppers per square yard, and the treatment is 92 percent effective (only 8 percent survive), then after treatment the population is 20 ´ 0.08 = 1.6 grasshoppers per square yard.
HopMod calculates forage consumption by grasshoppers on both treated and untreated rangeland to determine the difference in consumption. This difference is the benefit to the ranch.
HopMod repeats the process for each treatment selected by Consult. Available forage is used in the economics model (RanchMod) to determine the benefit–cost ratio for each treatment.
How Oviposition Is Determined.—HopMod assumes that grasshoppers lay eggs at a constant rate. The rate is different for grass feeders and mixed feeders. For grass feeders, the rate is 0.6550 eggs/adult female/day; and for mixed grass feeders, the rate is 0.4564 eggs/adult female/ day (Berry et al. 1995).
How Accurate Is HopMod.—HopMod has been field validated (Berry et al. 1995). HopMod correctly simulates the general patterns of rangeland grasshopper population dynamics within a given year (Berry et al. 1991).
Comparison of Field Data and the Grasshopper Model.—Figure VI.2–31 shows a comparison between field data and the phenology model’s plots. As you can see, the calculated values closely match the field values. In addition, the estimates of forage consumption by the different grasshopper instars are based on scientific measurement. Therefore, you can expect HopMod to produce reasonable approximations of grasshopper forage consumption.
Steps You Can Take To Improve Accuracy.—You can improve accuracy in two ways:
1. Conduct
the grasshopper census as close to the treatment date as possible.
2. Enter
actual measurements of the average instar instead of accepting calculated
values.
Remember, HopMod does not have an egg-hatch model. Consequently, HopMod cannot add newly hatched grasshoppers to the population. As a result, the greater the time between field measurement and treatment the greater the error in estimating average instar and density. So for best results, use current data.
Also, observed measurements are the best estimate of reality. Therefore, whenever possible, enter observed measurements instead of relying on Hopper’s initial life-stage estimates.
What the Economics Model (RanchMod) Does.—RanchMod determines the value of the forage. With this information, and with information from the other Hopper models, RanchMod can determine if a treatment is cost effective. In addition, RanchMod compares the cost effectiveness of each treatment listed by Consult, so you can decide which treatment is most cost effective. The model reports cost-effectiveness as a benefit–cost ratio.
To understand RanchMod, you must know the following:
• How RanchMod
determines the benefit–cost ratio,
• What
information you may supply, and
• How
reliable RanchMod is.
How RanchMod Determines the Benefit–Cost Ratio.—Using the forage and grasshopper models, RanchMod estimates the value of forage consumed by grasshoppers when treatment is applied and when treatment is not applied. The difference is the damage avoided by treatment, called the benefit. RanchMod assumes that the forage saved (less the forage set aside by the proper use factor) is available to livestock. The proper use factor is the proportion of the forage that will not be consumed by livestock, to prevent overgrazing.
The model divides the value of the forage saved (benefit) by the cost of treatment to determine the benefit–cost ratio.
RanchMod combines information from the forage and grasshopper models within its economic model to determine the value of forage. The value of forage directly affects the benefit–cost ratio.
What Information You May Supply.—The economics model asks you for information on the arrangement, and operation of the local ranch(es). This information includes the following:
• Lease costs,
• Cost
and availability of hay,
• Livestock
prices, and
• Herd
information–size and composition.
Hopper provides default values for most of these variables. Default values are averages. When you don’t know the actual value, you can use the default value to get a reasonable approximation.
Do not, however, use default values for grasshopper population size and species composition. These values are so variable that your results will be useless. So, for grasshopper density and composition, always use field data. Supply the best information you can for other values as well.
Figure VI.2–31—Validation
runs showing average life stage (S, field data; solid line, model)
and density (D, field data; dashed line, model) for GHIPM sites
in North Dakota.
Remember, Hopper is only as accurate as the information you supply. The closer this information matches reality, the more reliable Hopper’s recommendation is. Use default values when you must, but supply the best information you can.
How Reliable RanchMod Is.—RanchMod is both reliable and justifiable. RanchMod uses factors previously unavailable to decisionmakers. These factors allow you to account for variation in the ranching environment and to justify your treatment decisions based on economic criteria.
RanchMod’s accuracy depends on the accuracy of the data. The closer the data are to reality, the more reliable the benefit–cost ratio. During average years and on the average ranch, the default values will produce good results. But the more conditions stray from average, the more critical that you enter factual data instead of allowing the program to use default values. With accurate data, expect reliable results.
Remember, RanchMod’s results are not exact. Rather, RanchMod gives you a “ballpark figure,” an estimate. RanchMod’s estimate, however, is more accurate and more reliable than any you get by other means.
Your Role.—Your role (the role of ranchers, ranching committees, and government officials) in making treatment decisions with Hopper is twofold:
1. To provide
accurate data to Hopper.
2. To
make the final decision.
Providing Accurate Data to Hopper.—Hopper’s recommendation relies on the data you enter. Therefore, to ensure reliability, you must enter the best data available. Collecting this data, however, requires skill, professionalism, and discipline.
Give Hopper the best data you can—it’s worth the effort.
Making the Final Decision.—You must make the decision to treat or not. Hopper supplies you with benefit– cost ratios and other useful information. You must decide whether to treat based on the benefit–cost ratio, and other factors not accounted for by Hopper, that you judge important. Hopper is a decision support tool, not APHIS policy.
Remember, under normal circumstances, treating when the benefit–cost ratio is less than 1 is economically unjustifiable. Failure to treat when the benefit–cost ratio is greater than 1 threatens the ranching economy.
Hopper provides support for your treatment decisions based on scientific and economic research. If you use Hopper’s benefit–cost ratio to make your decision, you can claim Hopper’s support. But if you use another criteria, you cannot.
Back to Hopper 4.0 User’s Guide
References Cited
Berry, J. S.; Hanson, J. D. 1991. A simple, microcomputer model of rangeland forage growth for management decision support. Journal of Production Agriculture 4: 491–499.
Berry, James S.; Kemp, William P.; Onsager, Jerome A. 1992. Hopper decision support system: rangeland grasshopper management for the 1990’s. In: Proceedings, 4th international conference: computers in agricultural extension programs; 28–31 January 1992; St. Joseph, MI. St. Joseph, MI: American Society of Agricultural Engineers: 610–615.
Berry, James S.; Kemp, William P.; Onsager, Jerome A. 1991. Integration of simulation models and expert system for management of rangeland grasshoppers. AI Applications in Natural Resource Management 5: 1–14.
Berry, James S.; Kemp, William P.; Onsager, Jerome A. 1995. Within-year population dynamics and forage destruction model for rangeland grasshoppers (Orthoptera: Acrididae). Environmental Entomology 24: 212–225.
Davis, Robert M.; Skold, Melvin D.; Berry, James S.; Kemp, William P. 1992. The economic threshold for grasshopper control on public rangelands. Western Journal of Agricultural and Resource Economics 17: 56–65.
Dennis, Brian; Kemp, William P.; Beckwith, Roy C. 1986. Stochastic model of insect phenology: estimation and testing. Environmental Entomology 15: 540–546.
Kemp, William P.; Onsager, Jerome A. 1986. Rangeland grasshoppers (Orthoptera: Acrididae): modeling phenology of natural populations of six species. Environmental Entomology 15: 924–930.
