Whole-Farm Nutrient and Agchemical Input Budgeting for Sustainable Farming: Analysis and Demonstration

Sub-objective A. Baseline data 1992

Farm fields and soil test information

For each farm, the field-by-field soil test results were used to compute the weighted farm averages of phosphorus (P) and potassium (K). The field-by-field by results and the weighted farm average were compared to the soil test level considered to be "optimum" by Wisconsin soil test recommendations for alfalfa, the most P and K demanding crop in the rotation on these farms. At "optimum" soil test levels, recommended phosphate and potash applications are the amount the crop is expected to remove. Optimum soil test levels vary by soil series, and for the comparison, the highest optimum of any soil on the farm was used. The intent of this comparison was to determine whether existing soil P and K levels were adequate for the crops grown and to look at the distribution of soil test levels across the farm.

Thull Farm:
Figure 1 is a map of the cropland the Thull family owns, along with a table showing the rotations specified for each field on the Farm Conservation Plan. There are 52 fields, averaging 10 acres. Most of the soils are silt loams. About 48% of the cropland acreage has sufficient slope to be designated "highly erodible land" in the Conservation Plan.

Figure 2 shows the soil test levels of P and K for the 52 fields, along with the farm's weighted average for these nutrients and the level considered optimum for alfalfa. In general, the highest testing fields were those closest to a manure storage area. The weighted average of P on the Thull farm is 39 ppm, which is in the excessively high category for alfalfa (>30 ppm) according to UW soil test recommendations. If this P were spread evenly across all the fields the farm, the Thulls could plant for several years before having to add P to replace that removed by the crops. K levels are different. The weighted farm average for K is 119 ppm, which is in the optimum category (91-120 ppm). Farm-wide, K must be added each year to make up for crop removal. For both nutrients, soil test levels vary widely over the farm (10 ppm to 155 ppm P, 65 ppm to 220 ppm K) - both K and P have soil test levels in the very low (P:<10 ppm, K: <70 ppm ) to the excessively high (P: >30 ppm, K: >170 ppm) range.

Montgomery Farm:
This farm has 88 fields, as shown in Figure 3. Average field size is 3.3 acres. The large number of small fields on this farm is typical of farms in the unglaciated part of western Wisconsin, which is characterized by steep slopes and narrow valleys. On 170 of the cropland acres on this farm, the slopes are greater than 9%; 70 acres have 6-9% slopes, and 50 acres have slopes of 6% or less. The soils are silt loam. With the exception of a few fields, the allowed rotations are 1 or 2 years corn, 1 year oats, and 4 years hay.

The weighted average of P on the farm is 23 ppm, which is at the upper end of the optimum range (Figure 4). The very highest testing fields are close to the barn, but other fields are also high. The weighted average of K is at 115 ppm, in the middle of the optimum range. Here again soil test levels for both nutrients vary from very low to excessively high (9 ppm to 102 ppm P, 75 ppm to 302 ppm K).

Guttmann Farm:
This farm has only 13 fields with an average field size of 7.2 acres. As Figure 5 shows, the weighted average P is 14 ppm, which is below optimum according to Wisconsin soil test recommendations for corn as well as alfalfa. Soil test levels of K are in the optimum category, with a weighted average of 100 ppm. At this level the Guttmanns can replace what the crop removes without affecting the soil fertility status. The range of soil test levels do not vary as widely on the Guttmann farm as on the Thull and Montgomery farms (9-16 ppm P, 76 - 118 ppm K).

1992: On-Farm Fertilizer Resources

Thull Farm:
Manure -
The Thulls have two manure storage facilities; one for the dairy operation, which is designed for 9 months storage, and one for the dry cows and steers, with 6 months storage. About 1.2 million gallons of liquid manure and 2,100 tons of semi-solid dairy manure are collected annually from their livestock, with an estimated first-year fertilizer content (i.e. the amount of nutrient available for crop growth during the first growing season after spreading) of 15,900 pounds of nitrogen, 15,900 pounds of phosphate, and 32,700 pounds potash.

An essential part of evaluating manure as a fertilizer resource for a farm is determining spreader load size. The liquid spreader here holds 3,400 gallons, but was assumed to have delivered 3,000 gallons because it can never be completely filled due to foaming. The box spreader for the semi-solid manure was weighed full and empty using portable pad scales and was found to hold 6.5 tons per load on average.

Legume residues -
In 1992, the Thulls had 182 acres of first year corn following alfalfa. With an average alfalfa nitrogen credit of 120 pounds, this represents about 22,000 pounds of nitrogen available from legume residues.

Residual nitrate -
Soil tests for residual nitrate were conducted before planting on this farm in 1992. However, no carryover nitrate was found. Samples were taken before planting in 1993 and 1994, and the results were incorporated in the 1993 and 1994 Farm Nutrient Management Plans.

Montgomery Farm:
Manure -
According to the Midwest Plan Service (1993) guidelines for estimating manure production, about 1,800 tons should be produced and 1,600 tons collected annually for spreading from the 75 dairy cows and replacement youngstock. A portion of the manure is not collected because the milk cows are out on unmanaged pasture 25% of the time and the bred heifers (1 to 2 year old replacements) are out 50% of the time. Milking herd manure is collected daily from the barn for spreading. Spreader loads were weighed to find that the average daily haul load from the barn is about 4 tons. Manure from the barnyard (5.5 tons per load) and heifer lot and shed (6.5 tons per load) is hauled out less frequently, when these areas are cleaned. The collected manure is estimated to have a fertilizer value of 4,800 pounds nitrogen, 4,800 pounds phosphate, and 12,800 pounds potash per acre in the first year after spreading.

Manure was applied to corn fields at a rate of about 20 tons/acre.

Legume residues -
There were 16 fields (52.2 acres) in first year corn following alfalfa. Alfalfa nitrogen credits were 120 pounds per acre or 6300 pounds farm-wide.

Residual nitrate -
Pre-plant residual nitrate tests on 8 fields (31.4 acres) where corn was grown the previous year resulted in an average credit of 70 pounds (range 38 to 128 pounds) of N on these fields. The pre-plant nitrate test was taken before planting in 1993 and 1994 and the results used in the 1993 and 1994 crop plans.

Guttmann Farm:
Manure -
The 35 cows and 20 head youngstock and dry cows produce an estimated 630
tons of manure, of which an estimated 380 tons is collected. The collected manure contains about 1,500 lb of fertilizer equivalent nitrogen, 1,500 lb phosphate, and 3,000 lb potash.

There was no manure storage on this farm, but manure was often stacked for several days or longer before it was hauled to a field. Manure was applied to corn fields at a rate of 8-10 tons/acre and was often incorporated with a few days after application; some manure was applied to alfalfa fields that were to be plowed under and planted in corn in 1993. A typical spreader load was 4 tons.

Legume residues -
In 1992, there were 16 acres in first year corn following alfalfa, with 1,920 pounds of nitrogen credited.

Residual nitrate -
The nitrate test was used on appropriate corn fields in 1992, but there was not enough residual nitrate to credit. It was used again before the 1993 growing season. Again, the amounts of residual nitrate found by the test were so low that they were not incorporated into the 1993 nutrient management plan.

1992: Purchased Fertilizer

Thull Farm:
Starter fertilizer rates for row crops were 250 pounds of 9-23-30 per acre (68,800 pounds applied farm-wide). Potash was also applied as bulk potash (0-0-60) (9,600 pounds farm-wide.) Nitrogen was applied to corn as urea (46-0-0) (23,100 pounds farm-wide) or ammonium sulfate (21-0-0) (10,680 pounds farm-wide). Some fields received less than the recommended amount of nitrogen. The Thulls did not topdress fertilizer on their alfalfa and have not generally done so in the past.

Montgomery Farm:
From 1983 to 1989, the fertilizer recommendations of the farm's agricultural supply co-op were generally followed. On second year corn, 120 pounds of nitrogen per acre was applied as anhydrous ammonia in addition to 20 tons per acre of manure. Following the on-farm demonstrations of nitrogen management on one field in 1990 and 1991, the Montgomerys began to refine their management practices farm wide. In 1991, they were impressed that pre-plant nitrate tests on four corn fields showed residual nitrate levels ranging from 48 to 184 pounds per acre. By 1992, they had reduced their nitrogen fertilizer applications and no longer used anhydrous ammonia.

Starter fertilizer (9-23-30) was applied at a rate of 250 pounds per acre. Alfalfa was top-dressed in the spring. A planned fall application was skipped due to lack of funds. Altogether, the Montgomerys purchased 20,000 pounds of starter (9-23-30), 4,600 pounds of diammonium phosphate (18-46-0), 3410 pounds of TSP (0-44-0), and 25,300 pounds of bulk potash (0-0-60).

Guttmann Farm:
The Guttmanns applied 60 pounds of actual nitrogen per acre to all of their corn with the exception of first year corn following alfalfa. They applied starter fertilizer (20-20-18) to corn at a rate of 200 pounds per acre. In 1992, they purchased 7,500 pounds of 28% nitrogen, 10,200 pounds of starter (20-20-18), 525 pounds of diammonium phosphate (18-46-0), 1,740 pounds of TSP (0-44-0), and 9,835 pounds of potash (0-44-0).

1992: Pest Management

Weather played a big part in weed control (and yields) in 1992. The growing season began with a cool, dry spring. Some areas in Wisconsin went for nearly eight weeks in May and June without measurable rainfall. There were hard frosts on May 28 and June 20. Yields were reduced statewide due to uneven germination, frost-kill, and poor weed control from herbicide non-performance and slow plant growth.

Thull Farm:
The baseline year was a terrible year for weed control on the Thulls’ corn fields. Problem weed species were velvetleaf, common lambsquarter, and foxtails. There were some problems with leaf hoppers and leaf hopper burn on alfalfa. The older stands had dandelions and quackgrass invading, but the new stands were weed-free. Soybeans did not have insect or disease damage and were weed-free (following an aplication of Pursuit at 4 ounces per acre).

Field records indicated that the Thulls were not applying pre-emergence herbicides soon enough after planting. Late applications meant the weeds had already emerged before the application. The pre-emergence herbicides they were using had no plant activity; consequently the herbicides were ineffective.

They did not use a soil applied insecticide for corn insect control. Instead, they used turpentine-treated corn seed (20cc turpentine per corn bag 24 hours before planting) While admitting that turpentine does not kill corn rootworm, they believed that rootworms did not like the smell of turpentine and left the plant alone. They used to use Thimet, but they decided it was not working on their farm.

Montgomery Farm:
The Montgomery Farm also had very poor weed control due to weather conditions that interfered with herbicide action. Multiple applications of herbicides were made to some fields. Herbicide application costs (including custom spraying) averaged about $50 per acre for corn and soybeans. Not much cultivation and no rotary hoeing were done due to lack of manpower, although the Montgomerys had recently purchased a rotary hoe.

During the 1992 growing season, the Montgomerys conducted a reduced herbicide application demonstration on one of their fields, and found that in that comparison they could successfully grow corn with less herbicide.

Guttmann Farm:
Weed control has been a problem at the Guttmanns. This year, they started to use lower (low end of the label) rates of herbicide in combination with timely cultivation to cut production costs, and the result was satisfactory weed control. Eastern black nightshade and quackgrass were a problem on some fields.

The Guttmanns only use a soil-applied insecticide in second and third year corn, and started corn rootworm beetle scouting to make decisions easier.

1992: Labor and human resources

Thull Farm:
The farm is run by William (Bill) Thull and his two sons, Ralph and Mike. Mike is in charge of the crops and Ralph is in charge of the dairy and livestock operation. They have four part-time employees. They are very capital intensive, but are labor-poor. Crop production practices rely heavily on pesticides, purchased fertilizers and big equipment. Due to the large scale of their farming operation and the need for additional labor, timely field operations have been a big problem.

Montgomery Farm:
Lee Montgomery operated this farm with part-time help from his dad Chuck and his wife Tammy, and seasonal hired help. The Montgomery family kept detailed, day-to-day labor logs from October, 1991 to September, 1993. From October, 1991 to September, 1992, 6,567 hours of labor were used to run the farm, an average of 18 hours per day. Figure 6 shows the average number of hours worked per day by month and by person. Figure 7 shows the hours worked by month and by labor category.

Such long hours for running a mid-size dairy farm are not unique. For comparison, the Mundth family, with 57 milking cows and 120 cropland acres, spent an average of 20 hours per day running their dairy farm in the period from October 1992 to September, 1994.

As you can see from figure 7, milking and feeding were the two most time consuming activities of the farm, taking 35% and 34% of the hours spent, respectively. The feeding category includes time spent going from one part of the farm to another to feed animals housed in different areas. In contrast, the categories of labor that might be most affected by revising nutrient and pest management strategies for crops -- manure handling and field operations (tillage, planting, rotary hoeing and cultivating) -- take up only 5% and 2% of the total hours.

Guttmann Farm:
The primary labor force on the farm is John Guttmann, his wife Annette and one of their sons, John. They receive help from John's father, Bill, who is a retired carpenter. John also works off the farm, hauling milk for the Cedarburg Dairy four times a week. They have no hired farm hands. Unlike the other two farms, they do not have a labor shortage that causes field operations to be delayed.

Sub-objective B. Analysis of patterns and strategies

1992: Nutrient flow patterns

Nitrogen

Due to a poor quantitative understanding of the dual role that alfalfa plays in nitrogen cycling, acting as a scavenger when nitrogen is available and a fixer when it is not, along with the tendency of unused nitrogen to leach from the system as nitrate, we felt that it was not possible to produce a whole farm nitrogen budget that accurately reflects reality. The large number of legumes, particularly alfalfa, in the rotations on these dairy farms makes any sort of farm-gate nitrogen budget for the farm meaningless. Instead, we quantified residual, organic (legumes and manures), and purchased nitrogen on the farm in 1992 and compared it to corn crop nitrogen needs (Table 1). Corn nitrogen requirement for the fields on these farms is, depending on soil type, 120 or 160 pounds per acre. The pre-plant soil profile nitrate test was used to identify residual soil nitrate levels in second year corn fields. As corn is the non-leguminous crop with by far the greatest nitrogen needs on these farms, other crops were not included in this comparison.

Thull Farm:
As Table 1 shows, the Thulls should have been able to meet the majority of their corn nitrogen needs with manure and alfalfa credits, and they applied enough commercial nitrogen to meet most of the rest of the need. However, they were not accounting for sufficient credits for the manure and alfalfa, and applied fertilizer nitrogen at an even rate over all of their corn fields. On some fields they applied excess commercial nitrogen when there was already enough nitrogen available from alfalfa residues and/or manure. Other fields were short of nitrogen.

Montgomery Farm:
Available nitrogen came very close to meeting the total recommended nitrogen requirement. Most fields had adequate nitrogen although very little commercial nitrogen was purchased. As was stated previously, the Montgomerys began to refine their nitrogen management practices in 1990 and have already greatly reduced their nitrogen fertilizer purchases.

Guttmann Farm:
Purchased and on-farm nitrogen only accounted for about 70% of the corn nitrogen requirement. Many second year corn fields were short of nitrogen.

Phosphate and Potash

An analysis of the amount of phosphate and potash entering and leaving the farms is shown in Table 2 and Figure 8 below. This analysis did not take into account losses from run-off or leaching.

When looking at the nutrient flows on these farms, the first thing you notice is the large amount of P and K entering the farmsteads versus leaving. The major part of the outputs of phosphate and potash are in milk, with a smaller portion as animals sold off the farm. Major inputs are in the form of feeds and fertilizers.

With the soil types on these farms, it takes an addition or removal of 18 pounds of phosphate per acre to cause a 1 ppm rise or decline in P, and 7 pounds of potash to cause a 1 ppm rise in K. (Combs et al., 1996). If the surpluses shown in this analysis are not lost through run-off or other unaccounted for factors, soil test levels for P and K will rise.

Thull Farm:
Inputs over outputs of 7,000 pounds of phosphate per year will increase average soil test values for P less than 1 ppm per year if spread over the farm's fields. The surplus of 22,000 pounds of potash will increase average soil test K by 6 ppm per year.

Montgomery Farm:
An increase of more than 10,000 pounds phosphate on the farm annually will result in an increase of approximately 2 ppm P per year if spread evenly over the farm. An increase of almost 22,000 pounds potash annually will increase soil test K values almost 11 ppm per year.

Guttmann Farm:
The increase in phosphate amounts to about 2,800 pounds per year. This additional amount will bring soil test values up by about 2 ppm per year if evenly distributed across the farm. Approximately 8,000 more pounds of potash are entering than leaving, which will increase soil test levels by about 12 ppm per year.

This nutrient flow analysis indicates that these three farms took in much more P and K than they are exported in 1992, but without knowing what soil test levels are on a farm, it is not possible to know if this is "good" or "bad." For the Thulls, who have high soil test levels of P, the surplus indicates that more efficient use might be made of on-farm source of P so that fertilizer imports can be reduced. On the Guttmann Farm, however, with an average soil test P below optimum, the importation of P is agronomically beneficial.

Sub-objective C. Developing and implementing revised strategies in 1993 and 1994

Thull Farm:

Nutrient management plans 1993

The nutrient management plan for 1993 is Appendix 1. The goals are to redistribute nutrients on the Thull farm through applying manure to meet the P and K needs of crops as much as possible in light of crop rotations, land restrictions caused by water quality concerns, and slope. In choosing a strategy, Proost found that applying manure to meet corn nitrogen needs, which would cause applications in excess of requirements of phosphate and potash, would save only $1,800 in annual fertilizer costs. Applying just the amount of manure to a field to meet phosphate and potash needs and applying additional nitrogen fertilizer where needed meant that manure could be spread on more fields and resulted in an estimated annual fertilizer savings of $3,800.

Manure spreading plans were based on the nutrient requirement of the rotation on the field. Corn fields were divided into four management units: 1) fields with high phosphate and potash needs, 2) high phosphate and low potash requirements, 3) high potash and low phosphate, 4) low or no phosphate and potash requirements. Each field within these units was treated similarly (but not exactly the same - different manure and nitrogen application rates were still required). The amount of manure to apply was indicated on the plan in loads per field, rather than tons or gallons per acre, as the Thull family found this an easier way to think about manure application rates.

The plan took into account nitrogen credits from legumes, manure, and residual nitrate in the soil. In 1993, three fields (26.5 acres) were converted from alfalfa with average credits of 120 pounds per acre for a total of 3,180 pounds. One field went from soybeans (15.8 acres) with 40 pounds per acre for 632 pounds of credit. Two fields had a residual nitrate credit from the pre-plant residual nitrate test.

Starter fertilizer rates for corn were specified as 150 pounds per acre 9-23-30 (as compared to their standard practice of 250 pounds per acre).

Following soil test recommendations, the plan called for substantial phosphate and potash applications to alfalfa. Farm-wide, the plan called for an increase in total phosphate and potash purchases, rather than a decrease, because alfalfa yields were suffering from nutrient deficiencies. As was indicated earlier, alfalfa fields were not routinely fertilized on this farm in the past.

1994: The nutrient management plan for 1994 is Appendix 2. As 1993 was a relatively poor growing season, residual nitrate levels were high in many corn fields in 1994. In the spring of 1994, the Thulls used the pre-plant nitrate test in their second year corn fields and found that on most fields they did not need additional nitrogen.

In this year's plan, in response to requests from the Thulls, the nutrient application recommendations gave the actual amount of a standard fertilizer mix to apply, rather than pounds of a particular nutrient, per acre (e.g. "360 lb/a of 0-0-60" rather than
"216 lb K/a").

Outcomes:

The Thulls indicated they liked these things about their nutrient management plans:

• Each field had its own plan, allowing them to identify the fields with the highest need for nutrients.
• The plans called for lower starter fertilizer rates over all.
• The plans identified the crop, where it was in the rotation, what was allowed by the conservation plan on each field.
• All corn was not treated the same; in the plan it was split into management units.
• The plans gave them manure management (spreading) plans consistent with their agreement for watershed cost-sharing.

Were the plans followed? Planned versus actual nutrient applications

Tables showing the differences between planned phosphate and potash applications and actual applications at Thull Farms in 1993 and 1994 are in Appendix 3.

1993: According to 1993 field records, there was a difference between the nutrient management plan and what the Thulls actually applied for phosphate on 36 out of 52 fields. (Not surprisingly, all of the 16 fields where the plan matched the recorded application were fields where 0 phosphate was called for). On 19 fields, the amount applied was less than the plan - amounts ranged from 2 to 90 pounds less per acre. Applications in excess of the plan on 17 fields ranged from 11 to 83 pounds.

Differences between the plan and potash applications occurred on all but 10 of the fields. (In half of those 10 fields, the plan called for no potash to be applied.) Less than the planned amount was applied on 29 fields, with under-applications ranging from 10 to 384 pounds per acre. More potash than planned was applied to 13 fields, with over-applications ranging from 15 to 214 pounds per acre.

Most of the larger discrepancies in the amounts applied versus planned were on fields where manure was planned but not applied or vice versa. Proost identified the major cause of the differences this year with the newness of the plan. The Thulls had to make a change in the way they thought about their fields. In the past, they grouped their fields by location, applying the same fertilizer and manure rates to fields in the same area with the same crop. The plan called for grouping fields by nutrient needs; meaning that sometimes adjacent fields would get different application rates.

1994: In 1994, there was a difference between phosphate called for in the nutrient management plan and what the Thulls applied on 30 fields. (Again not surprisingly, of the 22 fields with no differences in application rates, 20 were fields where no phosphate was planned.) Over-application ranging from 10 to 90 pounds per acre was made on eight fields, while under-applications ranging from 15 to 85 pounds were made on 22 fields.

As for potash, the plan and reality were the same for only seven of the fields, and all but two of these fields were fields where no potash was planned. Only six fields received more than the plan called for, and these were fields that got more manure than planned. The majority of the fields (39) got less than planned by 15 to 384 pounds.

In 1994, high rainfall caused the Thulls to make an emergency mid-season distribution from their pit to avoid overflow. Some manure was spread on alfalfa fields that would otherwise have been put on corn fields after harvest. The wet weather also meant that some of the fields were too wet for spreading and the manure was diverted to other fields.

The major reason for the discrepancy between planned and actual potash applications in 1994 was financial. The Thulls purchased no potash (0-0-60) in 1994 while they purchased almost 20,000 pounds in 1993.

Fertilizer purchases

As shown in Table 3, the total amount of starter fertilizer went down over the three year period as the Thulls cut their minimum starter rate from 200 to 150 pounds per acre in 1993 and from 150 to 100 pounds in 1994. There is a jump in nitrogen applications from 1992 (when some fields’ corn had less than the recommended amount of available nitrogen) to 1993. In 1993, the Thulls followed UW Extension nitrogen recommendations for corn, but they did not fully credit alfalfa nitrogen or use the results of the pre-plant nitrate test. In 1994, they fully credited all available nitrogen - alfalfa, manure, and residual (from the pre-plant test) and their nitrogen use dropped to less than half of the 1992 amount. According to Proost, the Thulls credited nitrogen "incredibly well" by the end of 1994.

Pest management plans

1993: The weed management plan did not include reduced herbicide rates on any of the Thull fields because they do not have the labor to do the cultivation needed for success with reduced rates. It simply is not practical for this operation at this time. Instead, the plan focused on low cost herbicides and improved application timing. Planned application rates were normal for the soil type and organic matter.

Scouting reports in 1992 showed that many of the corn fields that were planted to corn in 1993 had corn rootworm counts well over the economic threshold level. These levels justify an application of a soil-applied insecticide. The Thulls agreed to use Lorsban on one field that had a high corn rootworm beetle count. This field had alternating strips (6 rows) of turpentine-treated seed and Lorsban applied at planting. The field was scouted for corn rootworms throughout the season. Root ratings showed significantly lower root damage where Lorsban was applied versus turpentine (2.1 vs. 3.3).

The Thulls planted corn-after-corn on very few fields, even where their conservation plan allowed them to. During the course of the project, they changed some of their rotations; switching from 3 years of corn (followed by 3-4 years alfalfa) to 2 years of corn, 1 of soybeans. This year of soybeans breaks the life-cycle of some major corn pests.

1994: The basic plan for weed management did not change in 1994.Corn rootworm beetle counts in 1993 on most fields did not warrant applying insecticides for 1994 corn.

Outcomes of the pest management plans:
The Thulls continued to rely on low cost herbicides and rotations to control weeds.

Following the on-field comparison of turpentine-treated seed version insecticide for corn rootworm control, they quit using the turpentine treatment, but did not start using insecticide. They preferred to control insects through crop rotation.

Proost has determined that the Thull family expended so much energy making major changes in nutrient management that it was not practical for them to make more than minor changes in other areas, such as weed control, at the same time.

Montgomery Farm:
Nutrient management

1993: The nutrient management plan for 1993, along with a record of the actual applications made to each field, is Appendix 4. Alfalfa was divided into four management groups in 1993 according to soil fertility status and age, with fertilizer application rates standardized within each group. The plan called for a major increase in importation of phosphate and potash fertilizers to topdress alfalfa. The 1992 alfalfa yield of 2.5 tons was low for the area. The goal was to increase average yields to 4 tons per acre. Due to the nutrient needs of alfalfa the plan did not call for an over-all decrease in potash and phosphate fertilizer purchases. In 1992, 8,300 pounds of phosphate were applied – 4,600 pounds as starter fertilizer and 3,700 pounds for alfalfa. The 1993 plan called for 10,700 pounds of phosphate – 3,060 for starter and 7,640 for alfalfa, and for similar increases in overall potash use.

The team decided manure could not be used to fertilize the alfalfa because:

• With the farm's lack of manure storage, only a small amount of manure (4-5) tons is available each day, an insufficient amount to apply to a whole field, even at low rates, during the narrow (2-3 day) windows when manure could be applied after cutting.
• The farm's box-end spreader applies manure in comparatively large clumps, which might damage alfalfa crowns. A spreader capable of fine, even manure applications costs about $14,000.

Feeling that this would be the best use of the manure in his system, Lee Montgomery chose to apply manure at 20-25 tons per acre to all of his second year corn fields (25.4 acres), with the remaining manure applied at lower rates (5 tons/acre) to first year corn following alfalfa (55.1 acres) and grassy hay fields that will be going into corn in 1994 (Management Group III). In this way, second year corn fields would receive all their P and K requirements and roughly half of their nitrogen from the manure. He felt that the pre-plant nitrate test would again probably show residual nitrates as in the past, making up the additional nitrogen. (In the spring, the pre-plant nitrate test did show an average residual of 40 lb/a.) In addition, although second year credits are often overlooked in nutrient management planning in Wisconsin, the plan took into consideration that there is a 50 pound per acre nitrogen credit the second year after alfalfa plowdown and that manure applications have a 1 pound nitrogen per ton credit the second year after application. (USDA-Soil Conservation Service, 1991). The plan called for commercial nitrogen to be applied to any fields requiring nitrogen after the residual nitrate test results were received.

Alfalfa supplies more nitrogen to the following crop if it is allowed some regrowth after the last cutting before it is killed (Bundy, et al, 1992). Looking to the next year, the plan called for cutting old alfalfa fields first in round of haymaking, allowing for four cuts as well as substantial regrowth before the fall application of Round-up.

1994: The 1994 nutrient management plan, along with a record of actual applications, is Appendix 5. Alfalfa was categorized into three management units. Most of the farm's manure was to be spread on the 12 fields of second year corn (51.2 acres) at 20-25 tons per acre with the remainder to again go on first year corn fields. Again, it was expected that with credits from the pre-plant nitrate test in the spring, most corn fields would have sufficient available nitrogen. Any that were short would receive commercial nitrogen as required.

1993 and 1994 Nutrient management outcomes:

Alfalfa fertilizer inputs and yields increased as planned. In 1993, the amount applied to alfalfa fields was within 15 pounds of the planned applications of phosphate and 50 pounds of potash, with the exception of fields in Management Group III, which received less 30-15-50 instead of 0-65-220, but also got manure in the summer. In 1994, all alfalfa fields received the planned applications. Alfalfa yields exceeded the 4 ton per acre goal in both years; average yields were more than 5 tons acre in both years.

Most corn fields had sufficient nitrogen. With residual nitrate, manure, and legume credits, all but a few fields had sufficient nitrogen for corn without commercial fertilizer.

Manure application rates were less than planned.
Actual manure application rates were generally less than planned rates. In both years, the amount collected was about 1,200 tons rather than the 1,600 tons estimated. Two possibilities are likely to account for this discrepancy. One is that the Montgomery cows may produce less manure than the Midwest Plan Service estimations. Another is that when calculating manure application, volume numbers are routinely rounded down to ensure that the fertilizer value is not over-stated when determining nutrient credits (e.g. a 4.4 ton spreader load weight is rounded down to 4 tons).

Actual fertilizer costs did not decrease. As is shown in Table 4 below, the substantial cuts that were made in starter fertilizer applications were offset by increases in potash purchases in both years and phosphate (as TSP) in 1994.

Commercial nitrogen costs were low in all three years; as planned, little nitrogen was required for corn. The Montgomery's plans anticipated that substantial residual nitrates would be found on second year cornfields, and residual nitrate credits are an integral part of the Montgomery nitrogen management strategy. However, in 1993, the cost of the test to identify those credits, the pre-plant nitrate test, was more than the amount saved by crediting the residual nitrate it found. The 13 fields (31.2 acres) with second year corn had residual nitrate credits ranging from 14 to 77 pounds per acre, or about 1,200 pounds in all. The savings for 1,200 pounds of nitrogen at $ 0.18 is $216. Having these time-consuming tests done by the crop consultant cost $595. In 1994, the residual nitrate credits on the 12 fields (40.6 acres) going into second year corn averaged 110 pounds per acre nitrogen (range from 19 to 181 pounds). At $ 0.18 acre for nitrogen, the saving for 4,600 pounds of nitrogen was $832. The cost of the test was $540.

Pest management

1993: One of Lee Montgomery's goals was to reduce input costs for field corn. Crop scouting reports were to be used to identify weeds and weed pressure, and post emergence herbicides used as needed. The plan calls for as many acres of corn to be grown without using chemical inputs as is feasible without suffering yield losses due to weeds. Bryan Black, the crop scout and consultant, was responsible for making herbicide choices and rate decisions based on weekly field visits for weed identification and ratings. When determining treatments, fields were categorized by rotation, tillage, and weed pressure. Corn was to be rotary hoed 2 times and cultivated 1 time. Lee Montgomery planned to hire help to do rotary hoeing and cultivating when it conflicted with his schedule for the first cutting of alfalfa hay. The estimated cost of this hired labor was about equal to estimated herbicide costs.

Alfalfa was to be fall-killed with Round-up at 1 qt/a (one half of normal label rate).

The crop scout was also responsible for pest identification and determining pest pressure on weekly field visits. Insecticides were to be used at 0.75 rate where needed.

1994: The 1994 pest management plan was essentially the same as in 1993. The crop scout was again responsible for making herbicide choice and rate decisions based on weekly field visits.

1993 and 1994 Pest management outcomes:

Pest management strategies appear to be a success. The weed and insect control strategies appeared to be successful. By using rotary hoeing and cultivation, they were able to grow corn without herbicides on 28 acres in 1993 and 30 acres in 1994.

The crop consultant played a crucial role. Lee Montgomery has indicated that one of the most beneficial aspects of this project was having the crop consultant to do field scouting and make recommendations. Without the timely field-by-field recommendations of the crop scout, he would not have been able to apply herbicide on an as-needed basis. In addition, because of his experience and knowledge of pest patterns on other farms in the area and over the years, he was able to have a wider perspective than the farmer on pest problems on the Montgomery Farm. Cost of the consultant to do whole acre management including soil testing, crop history and planning session, fertility and pesticide recommendations, variety selection information, bi-weekly crop monitoring throughout the growing season, hot field scouting, and color computer field maps was $6.90 per acre in 1993 and 1994, or $1,946 to do 282 acres.

Hours worked did not decline; quality of life remained a concern. Quality of life is a big concern to the Montgomery family. Lee Montgomery is discontented with the amount of time spent in farming, "Quality of life doesn't exist. I don't even have a life." The family budgeted $24,000 per year from the farm income for family living, while in 1991-1992 family members averaged 15.5 hours per day 7 days per week for a wage of $4.25 per hour.

In the period from October 1992 to September 1993, they worked even more - 16.4 hours per day. This increase was partially due to an increase in time spent harvesting (as the 1992 harvest was delayed well past October also, increased hay yields took longer to harvest) and milking (the barn was remodeled this year, and the resulting disruption made milking take longer). As a result of hiring someone to do rotary hoeing and cultivation, the amount of time spent by others working on the farm increased by almost 400 hours.

Revising the farm's nutrient and pest management strategy to be able to use the farm's resources more efficiently did not adequately address the problems of long hours and low pay for the Montgomery family. After 1995, Lee sold his dairy herd and took an off-farm job.

Guttmann Farm:
1993 Nutrient Management Strategies

The Guttmanns felt that if their farm was to be sold to developers, it made little sense to raise soil test values for P and K up to recommended levels for agricultural production. Instead, their plan only included enough additions to meet the need for crop removal. It is Appendix 6.

1992 to 1994 Whole Farm Changes

Average soil test values for K have changed from 1991 to 1994. As only one set of routine soil tests was made for each farm in these years, it is not prudent to make too much of changes on any particular field between 1991 and 1994. However, we have looked at the changes in the farm weighted averages.

Thull Farm:
Contrary to predictions made from looking at the farm nutrient flow in 1992, at the Thulls, the farm weighted average P remained practically the same and the farm weighted average K went down 29 ppm. (Figures 9 and 10.) Decreases in fertilizer purchases in 1994 caused the farm-gate nutrient flow for phosphate and potash to be close to a balance between imports and exports.

Montgomery Farm:
Farm-wide weighted averages of P remained about the same, and average K went up 14 ppm (Figures 11 and 12).

Yields have increased

Overall yields on both the Montgomery and Thull Farms were much better in 1994 than they were in the beginning of the project. However the effect of the new management strategies in these gains can not be distinguished from the weather. In 1992, weather conditions led to poor yields statewide. The next year, 1993, was somewhat better, and 1994 was a very much better year for most of the state.

Obstacles to making whole farm management changes identified

Thull Farm:
Richard Proost concludes from long-term interviews that these are obstacles to change at the Thull farm:

• Peer groups and local experience
  When a demonstrated practice has been used in an area, other farmers know about it. If it failed everyone knows. This makes it extremely difficult to try and get farmers to look at the practice again, regardless of whether it was done incorrectly. In addition, if the practice was done by a farmer who has a local reputation for doing weird things, that practice will also be considered "weird." It is guilt by association.
• Inadequate knowledge base (i.e. they do not know about new practices)
  Depending on the practice, the farmers may be unaware of it. If it is a "new" practice, farmers must first seek general information about it, and then determine if it will fit into their operation. This can take two or more years.
• Time and labor requirements
  Farmers know how long it takes them to complete their current crop production practices with their current time and labor commitments, and also the relationship to other farm enterprises. They do not know anything about how the new practices fit. They also do not know if they have the time and labor to accomplish the necessary operations of the new practice. If they do not have the on farm labor, do they need to hire it? If they hire the labor is the practice still profitable? And if it is, can they even find the labor? In addition, if the new practice does not have a readily observable advantage, it probably will not go too far. It takes a lot of time and energy to decide if a practice will fit. Many farmers do not have this luxury.
• Appearance of increased risk
  If a new practice looks riskier than a current one, the odds that a farmer will look into the new one are very low. If the new practice contains new ideas or operations that the farmer isn't familiar with, it looks riskier. However, if the new practice contains familiar operations, it may not look as risky. Both of these are independent of the actual risk as measured from research data.
• Government programs
  This is self-explanatory. Because of the inflexible nature of many farm programs, the adoption of new practices may be "ruled" out.
• Specific physical aspects of the farm
  This has to do with distance to fields, topography of fields, etc. They all dictate what can be done realistically, and what can be done only on paper.
• Philosophical differences in decision-making within the family
  This involves the complex interactions within the farm family. It often results from differing farm goals, or no definable goals at all. When each family member has different goals for the farm, the decision made most often is no decision at all - with a lot of conflict. This also occurs when family members feel that they are doing the majority of the work, while the others are off doing other non-farm things. This can be the most troublesome area to contend with. Meddling into personal family affairs is not something that an agricultural specialist can or should do.

Montgomery Farm:
From her conversations with the Montgomery family, Karen Talarczyk concludes that these were obstacles to change and difficulties in using the whole farm approach on the Montgomery Farm:

• Lack of control over variables such as weather and crop prices.
• Time management - prioritizing on paper differs from an actual day.
• Labor constraints.
• The long hours and low pay that dominate an agricultural career.
• Unexpected challenges when working with livestock.
• Depending on hired help.
• Complex decisions required when taking into account needs of livestock and farm family as well as crop.
• Management practices that they view as impractical will not be adopted.

Sub-objective D. Educational programs

While this project worked directly with three farm families, we tried to continually alert other farmers to what was happening on these farms through multiple methods, including:

• signs on the farms
• articles in watershed and Extension newsletters
• articles in the agricultural and local press
• annual field days involving both Extension and County Conservation personnel
• field day invitations extended through a variety of methods to area farmers (direct mail, personal invitation, announcements in Extension newsletters)

All of the field days included a "freebie" - an inducement to come, in addition to the field day itself (e.g. food, free testing of water samples).

All of the project farms were located in one of Wisconsin's Priority Watersheds. Priority Watershed programs are run by local County Land Conservation Staff. Educational programs were tied in with the Priority Watershed programs' events and field days. The central message of the educational efforts stemming from this project was that it is possible to reduce inputs through nutrient management planning while maintaining yields, thus protecting profits and water quality at the same time.

Both surface and groundwater quality information was included in addition to agronomic information at field days. Groundwater profile cross section diagrams were produced for each farm to help explain groundwater. (Attached as Appendix 7.) Reducing excess nitrogen applications for the purpose of avoiding nitrate contamination of groundwater was a particularly pertinent part of the message at field days for the Montgomery farm -- of the 22 farm drinking water supply wells sampled in the Whiteside creek sub watershed where the Montgomery Far m is located, 58% had nitrates above the drinking water health standard of 10 ppm.

Field days and events:

Thull Farm - 8/18/92, 9/15/93- Balancing water quality with crop production

Montgomery Farm - 9/9/92 - Field visit as stop on the Lower Pecatonica Watershed Educational Tour; 9/2/93 Watershed tour theme was “Working together for better water quality for our farm families.” Lee Montgomery spoke on "Whole farm experience reducing ag chemicals and fertilizers"; 8/28/94 - Lower East Branch Pecatonica Watershed Picnic, 8/30/94 - Watershed tour stop Lee Montgomery and Karen Talarczyk spoke on "Farming practices to improve water quality."

Guttmann Farm -- 8/27/92

Nutrient management workshops for farmers in project areas:

Washington, Sheyboygan, Dodge, Jefferson, Ozaukee Counties - winter/92
Green and Lafayette County - 3/93
Lafayette County - 3/9/94

Subobjective E. Evaluating the effectiveness of educational outreach in promoting adoption of demonstrated farming practices on other dairy farms in the region.

The project's initial plan was to survey the operators of farms matching the demonstration farms in farm enterprise mix, farm organization, soils, topography and climate using a "Farm Practices Inventory" survey (developed by Nowak and his students) that identifies current agrichemical management practices, and perceived or real obstacles to adopting practices that reduce reliance on purchased agrichemical inputs. Matching farms within a given area is not easy. We were unable to identify suitable match farms for the Thull and Guttmann Farms.

Five farmers with farms that somewhat matched the Montgomery Farm (though herd sizes varied from 90 to 140 cows) were interviewed in spring 1992. While these farms were not immediate neighbors, they were located within a five-mile radius and within the Priority watershed area. They were re-interviewed in fall 1996 with a special survey to identify any changes in behavior relevant to the project, where they get their information from, and major shifts in their operation. Admittedly this is a small sample, but the responses do not bode well for making much change in the arguably short time frame of four years, without direct one-on-one assistance.

Although four had increased their milking herd size significantly, none had made big changes in their nutrient management. Although none of them were making major over-applications of nutrients to begin with and all used soil tests to some degree and two were adequately crediting manure in the first place. One of those who had already been crediting manure and legumes had received a nutrient management plan in 1994 as a requirement for cost-sharing a new barnyard through the priority watershed. He found the plan "very complicated," but very similar to the way they were already managing their manure. One said his brother had tried crediting nutrients from manure and had a bad experience. Another said he did not have the time to learn how to credit adequately. In general, all of these farms had had little to no contact with county-based Extension personnel; three had had some positive contact with the Land Conservation District staff.

Some of the answers to this very short survey did confirm some of this project's findings regarding obstacles to whole farm adoption of nutrient management planning. Farmers indicated a need to have all fields managed the same or said that they can not make complicated changes due to lack of help or lack of time. Those with manure storage had an easier time of manure management than those without.

Obviously, this project was successful in getting the farmers we worked with directly to make some major changes. This has been the experience of Nutrient and Pest Management (NPM) Program with on-farm work since 1990 - intensive one-on-one assistance and demonstration is likely to lead to adoption of recommended practices. In a 1995 survey of NPM cooperating farmers (45 out of 60 responded), we found that NPM cooperators are much less likely to apply excessive amount of nitrogen fertilizer than the general farm population questioned in similar surveys and 81% are within 40 pounds per acre of the nitrogen application guidelines (NPM, 1995).

Although one-on-one assistance is leading to changes, there is little concrete evidence that these demonstrations are influencing neighboring farms, at least in the short term. However, 56% of the NPM cooperators thought that their participation in a demonstration had influenced other farmers to consider using the demonstrated practice. In addition, 44% said other area farmers have started using the practices they demonstrated (51% did not know).

Sub-objective F. Investigating whole farm soil nutrient flow patterns on a farm using intensive rotational grazing

A fourth farm was added to the project in 1992. It is owned and operated by Larry and Bridget Mundth in Sauk County in south central Wisconsin. They have 57 milking cows and 120 tillable acres. In 1992, they began a system of rotationally grazing their cows on 35 acres of grass and legume pastures for most of its spring-to-fall forage needs. The pasture is divided into 21 paddocks for intensive rotational grazing. Typically, the cows are moved from one 1.7 acre pasture to the next daily. The heifers follow, grazing the paddock that the cows grazed the day before. In addition to providing relatively high yields of good quality, cattle-harvested forage, the pasture provides an efficient means for managing and recycling manure nutrients.

In 1993 and 1994, the amount of nutrients supplied to these pasture areas by manure during grazing (estimated by grazing time, assuming cows produce 4.17 pounds and heifers produce 2.67 pounds of manure per hour) was compared to the nutrients removed in the forage (estimated by dry matter removal, using forage quality samples). Grazing cows deposited about 12 tons manure per acre during the grazing season, with a first year fertilizer value of 54 pounds of phosphate and 126 pounds of potash. They consumed about 3.23 tons of dry matter per acre over the season, removing an estimated 59 pounds of phosphate and 269 pounds of potash.

Kevin Shelley and Karl Hakanson (1997) are continuing to analyze the implications of intensive rotational grazing for whole farm nutrient management. Their report, which is now in preparation, will be filed as an addendum to this report.