Samis v. Agricorp, 2001 ONAFRAAT 55

Agriculture, Food and Rural Affairs Appeal Tribunal 1 Stone Road West Guelph, (Ontario) N1G 4Y2 Tel: (519) 826-3433, Fax: (519) 826-4232 Email: AFRAAT@ontario.ca

Tribunal d’appel de l’agriculture, de l’alimentation et des affaires rurales 1 Stone Road West Guelph (Ontario) N1G 4Y2 Tél.: (519) 826-3433, Téléc.: (519) 826-4232 Email: AFRAAT@ontario.ca

APPEAL: Samis v Agricorp Samis v Agricorp 2001 ONAFRAAT 55

STATUTE: Crop Insurance Act

HEARING: November 23, 2001

DATE OF DECISION: December 12, 2001

2001-55

NEUTRAL CITATION: 2001 ONAFRAAT 55

Samis v Agricorp

IN THE MATTER OF Ontario Regulation 140/96 under the Crop Insurance Act (Ontario) 1996, S.O. 1996, C. 17, Schedule C.

AND IN THE MATTER OF: An Appeal to the Agriculture, Food and Rural Affairs Appeal Tribunal by Wilson Samis, Iron Bridge, Ontario, from the decision of AGRICORP regarding the adjustment of a claim concerning his 2000 hay crop under Regulation 380/97 and the Crop Insurance Plan for Hay and Forage.

Before: Murray Cardiff, Chair; Andrew Osyany, Vice Chair; Denis O’Connor, Vice Chair

Appearances: Wilson Samis, appellant Darlene Bowen, OFA, Members Services Representative, for the appellant Fred Thomson, Regional Claims Manager, AGRICORP Jim Zavitz, Provincial Claims Manager, AGRICORP Husseinali Shivji, Technical Specialist, Forage Program, AGRICORP

DECISION OF THE TRIBUNAL

This appeal was heard in Sudbury, Ontario, on Friday November 23, 2001. Mr. Wilson Samis appealed a decision of AGRICORP concerning its adjustment of a claim concerning his 2000 hay crop under Regulation 380/97 and the Crop Insurance Forage Plan.

The Background

The Forage Plan covers loss against drought only. It is an area-based plan based on a computer model that simulates forage growth. A computer generated yield simulation is used to establish an area average which is used in establishing the insured grower’s yield for each crop year. Individual data collected by the grower is also entered into the simulation. Since 1977, the computer program called “Soil Moisture-Based Simulation of Forage Yield” (SIMFOY) has been used to simulate yields for the purposes of this plan. Individual growers may experience yields that are significantly different from the simulated yields.

In proposing and administering the Forage Plan, AGRICORP derives its authority from Section 5 of the Crop Insurance Act (Ontario), 1996 and Regulation 380/97 Crop Insurance Act (Ontario) 1996, General, and the Crop Insurance Forage Plan 2000.

Section 5 of the Crop Insurance Act states:

(1) AgriCorp shall fix the terms of contracts of insurance, or proposed contracts of insurance,

subject to section 4 and the regulations made under section 12.

(2) AgriCorp has all the powers necessary to perform its duties including the power to,

(a) determine the qualifications and requirements for a person to enter into a contract of insurance;

(b) enter into contracts of insurance;

(d) fix premium rates payable by insured persons;

(e) fix the duration of contracts of insurance;

(f) specify the circumstances in which an insured person may terminate a contract of insurance and the methods that the person may use to terminate the contract;

(g) specify penalties imposed on an insured person who breaches the terms of a contract of insurance;

(h) reinsure with any other insurer the risk or any portion of the risk under its contracts of insurance; and

(i) terminate a contract of insurance on the terms that it considers appropriate.

(3) An applicant for a contract of insurance or an insured person who receives notice from

AgriCorp of the terms of a contract of insurance or amendments to the terms, as the case may be, shall be deemed to have accepted them unless the recipient notifies AgriCorp to the contrary within the time period that AgriCorp specifies.

(4) AgriCorp shall not enter into a contract of insurance with a person to insure an agricultural crop or a type of perennial plant if,

(a) the contract insures less than the entire crop or all the plants of the type of perennial plant, as the case may be, in respect of which the person could enter into a contract of insurance under this Act; or

(b) a contract of insurance is already in effect to insure the crop or the type of perennial plant, as the case may be, in which the person has an interest. 1996, c. 17, Sched. C, s. 5.

The Issue

The issue before the Tribunal is:

Did AGRICORP calculate the simulated forage growth for the appellant in accordance with the coverage purchased and according to the requirements of the plan?

The Evidence

Wilson Samis Testimony

Mr. Wilson Samis told the Tribunal that he was a beef farmer in Bright Township, Algoma District, and had purchased crop insurance for more than 10 years. He said he was supportive of the forage program, but he did not believe his 2000 payout was accurate. He said there were three basic errors that are responsible for the difference between the simulated yield for his farm derived from the SIMFOY model and his actual crop performance.

Mr. Samis said that he had compared rainfall records collected between May and August, inclusively and between 1995 and 2000, inclusively. He said that the records proved there was less rainfall in 2000 than in 1998 but the SIMFOY model indicated that there was higher rainfall in 2000. He said that he could not accept that it is an accurate model. He said he had been using an older ‘wedge style’ rain gauge in 2000, but that it had been replaced by a ‘conical style’ rain gauge on or about June 15, 2001. He said the old gauge was clean and in good repair. Mr. Samis said that a comparison of the 2001 readings between the two gauges showed the old gauge significantly over-estimated the amount of rainfall, except when rainfall was greater than 40 mm, in which case it recorded less rainfall than the new gauge. He said that there were only five rainfalls that exceeded 10 mm in 2000 and that most of the rainfall received was at the low end of the scale where the disparity between the two gauges was highest. He said he believed AGRICORP should have supplied him with the new gauge earlier. He estimated that the old gauge inflated his actual rainfall by 25%.

Mr. Samis said the second error in the SIMFOY model was the calculation of initial soil moisture. He said this data was collected in Laird Township, 65 km from his farm. He said Laird Township is one of the wettest areas in the Algoma District. He said it was wrong of AGRICORP to use the initial soil moisture in this location in the simulation of the yield on his farm. He said the third error made by AGRICORP was that it collected rainfall data at the Sault Ste. Marie airport, which is 100 km from his farm, and in an even wetter area than Laird Township. Mr. Samis presented diagrams depicting precipitation patterns in Ontario in 2000, segregated into separate moisture zones, taken from “Ecology of a Managed Terrestrial Landscape, 2000”. He pointed out that his farm was two moisture zones below Laird Township and three zones less than the Sault Ste. Marie airport. He said that the data showed that his farm was in a moisture deficit more often than Laird Township and the airport in Sault Ste. Marie.

Mr. Samis also told the Tribunal that he had two fields which were treated the same in 2000 and 2001, but that he harvested less than half the normal yield in 2000. He said that he understood that he and AGRICORP have a contract, and that he had abided by his agreement to collect rainfall data and report cutting dates. He said it was the responsibility of AGRICORP to run the model and use due care. He submitted that the actions of AGRICORP fell below a reasonable level of care with respect to the rain gauge and the data collection locations.

In response to questions, Mr. Samis said:

He understood that the new gauge was to be used in 2001. At the time he received the new gauge he did not believe the old gauge was faulty, but he recorded data from both gauges out of interest.
The gauges were mounted on nearby fence posts, approximately 20 metres apart, equidistant from the ground.
His home farm is 164 acres but he leases four other farms and has access to approximately 500 acres, of which 133 acres is in forage. He has over 100 head of cattle and he grazes them on hayfields following the first cut, rather than taking a second cut.
His species mix is approximately 30% trefoil on average and 15% timothy with the balance in orchard grass, white clover, sonja dutch clover, alsike, red clover and native grasses. He has some alfalfa in the forage mix on fields that he does not use for pasture and that are tile drained.
He manages his forage mix to optimize grazing; advice is received from OMAFRA and other consultants. He intended to include tall fescue in his mix as a result of data received in local field trials.
He uses certified seed.
He purchased 21,000 pounds of fertilizer in 2000. He uses soil tests to determine what application is to be used on each field.
Bales of hay are weighed every year on his farm by AGRICORP.
The rainfall in September 1999 was lower than that in September 1998 and September 1997. There was a significant drop in rainfall in 2000 compared to the two previous years.
His farm pond dried up in 2000 for the first time in over 20 years.
He does not record wind patterns.
Heat units are in the 2000 range on his farm.
He uses 4x5 round bales and his forage is always put up as dry hay.

Fred Thomson Testimony

Mr. Fred Thomson told the Tribunal that farm management practices were not an issue in this dispute. He said Mr. Samis is a good farmer and has been an AGRICORP co-operator for seven years.

Mr. Thomson said SIMFOY was developed by Dr. Murray Brown of the University of Guelph with the assistance of other researchers. He explained that it is an area-based plan and that drought is the only insured peril. However, he said that for the forage program, drought is defined as a lack of rainfall, temperature or sunshine. He explained that SIMFOY is a computer driven plant growth model designed to reflect growth conditions within a geographic area, not on a specific farm. He said data entered into the computer simulation included: rainfall, minimum and maximum temperatures, sunshine hours, actual first crop yields, initial soil moisture, a fixed species mix per area and soil type. He said that the model assumes that there is a minimum 30 inches of soil. Mr. Thomson said that because the assumptions used in the model are not the same as conditions on specific farms, the results of the simulation might not mirror the results on specific farms. He also pointed out that other factors such as disease might affect crop yield on individual farms. He also submitted that no one year is the same as another year and payouts cannot be compared based on differences in one element of the model, such as rainfall.

With regard to rain gauges, Mr. Thomson said that AGRICORP has used several types of gauges over time. He said he did not know why it switched from the ‘wedge style’ to the ‘conical style’ gauge, but speculated that it may have been due to the availability or cost of the gauges. He said that regardless of what style gauge is purchased, all are purchased from manufacturers who certify that they are accurate.

Mr. Thomson said he compared the ‘wedge style’ and ‘conical style’ gauges in 1999 and 2001. He said that he had the gauges suspended from the same pole, equidistant from the ground. He said he found no significant differences between the two styles in 1999, a year with little rain. He said he did notice significant differences in his 2001 trial, but the magnitude of the variance was less than Mr. Samis found. Mr. Thomson testified that his experience was that the intensity of rainfall affected the rainfall recorded.

Mr. Thomson pointed out that for the purpose of the forage plan, both styles of gauges are considered to be accurate. Mr. Thomson said that Mr. Zavitz had calculated a 10% variability between the rain gauges, based on Mr. Samis’ data. He said that some of Mr. Samis’ data was questionable, a variation between 1 mm and 2 mm of rainfall could be due to errors in reading the gauges. He said that he had also found that the location of the gauge is often responsible for variable data, but that he did not know if that was the case in this instance.

Mr. Husseinali Shivji added that old gauges tended to be discoloured and gummed up over time. He said wind could also affect the amount of rainfall collected in the gauges.

Mr. Thomson said that AGRICORP decided to standardize gauges in 2001 as the result of recommendations made by two separate panels of the Tribunal. He said that the decision to standardize the gauges was not a reflection of the accuracy of the old gauges.

Mr. Thomson suggested that there was an error in the documentation provided to the Tribunal by Mr. Samis, as he had not recorded a $700 payout received in 1996. He said that as a general rule, AGRICORP customers received $3 for each $1 spent on forage crop insurance over the long term. He said that this was roughly the payout percentage received by Mr. Samis.

With respect to the difference in Mr. Samis’ payouts in 1998 and 2000, Mr. Thomson stressed that it was important to look at other factors besides rainfall. He said that some factors were the same in Mr. Samis’ simulation in both years, but that there were differences. He said that the growing degree-days were significantly higher in 1998, indicating that it was a warmer year. He said that the average of co-operator yields was much higher in 2000 than in 1998 and that this has a significant impact in the SIMFOY formula. He also said that there was a poorer distribution of rainfall in 1998 than in 2000. He suggested that these factors contributed to the difference in payouts to Mr. Samis in 1998 and 2000. Mr. Thomson said Dr. Murray Brown concurred with AGRICORP’s logic for explaining the differences between these two years.

With regard to the soil moisture levels, Mr. Thomson stated that SIMFOY calculates the beginning of growth after three consecutive days where the temperatures reach 5ºC or higher, after April 25th. He said the initial soil moisture is tested at 50-60 locations across the province to try to determine if there is enough moisture to start the process of growth. He explained that six samples are taken at each location at various depths and are tested at the University of Guelph. Mr. Thomson said that if the soil moisture is in the normal range the formula is not adjusted. He said that the developers of SIMFOY felt AGRICORP sampled at enough locations.

Mr. Thomson said that there were only a few days of rainfall data collected in Sault Ste. Marie used in the simulation of Mr. Samis’ forage yield. He explained that since Mr. Samis submitted rainfall data, beginning May 1st, his data was used. He said that data from Sault Ste Marie was used from the start of growth, April 25th or later, to April 30th.

Mr. Thomson told the Tribunal that AGRICORP does not offer a guarantee that SIMFOY model results will match individual yields. He gave a brief description as to how the forage plan is administered. He said that he met with other growers in Mr. Samis’ area in the Spring of 2001 to discuss their concerns and none of them had initiated an appeal.

In response to questions, Mr. Thomson indicated:

AGRICORP purchases weather data from Atmospheric Environment Services (AES), a division of Environment Canada. Sunshine data collected in Sudbury and temperature data collected in Sault Ste. Marie were used in the SIMFOY model for Mr. Samis’ crop.
Originally AGRICORP used only 12 main stations, as the theory was that there is little variation in temperature, but it now has 25 sites in order to improve accuracy.
There were two spans of 8-days each in which there was no rain in June 1998.
SIMFOY does not account for growth starting, prior to April 25th.
Only the soil samples taken in Laird Township were used in the simulation of Mr. Samis’ yield.
Co-operator yield is recorded as a percentage of normal yield. Normal yield is a ten-year average of actual co-operator yield, or an underwritten estimate of yield for the years when actual yield is not available.
Bales are weighed and the size of the field is either measured by AGRICORP or provided by the co-operator. Producers are taken at their word as to the number of bales harvested from the field. AGRICORP looks for a stand that is well established, but not run out. The same field is not necessarily tested each year. Co-operators sign the data collection sheet to verify that the information is correct.
A representative sample of bales are weighed to determine the yield of a known acreage.
Mr. Samis’ average yield was 1.32 tonnes/acre. His yield in 2001 was 3.05 tonnes/acre, 231% of normal. If his yield in the year 2000 was 50% of his average yield, that would be approximately 0.66 tonnes per acre.
Co-operator yield data is averaged with simulated data to obtain an average first cut yield in SIMFOY. Co-operator yields are used to account for variances in farm management practices.
The model assumes a 50% trefoil and 50% timothy mix for Mr. Samis’ area.
Laird Township was within the normal range for available moisture to initiate plant growth in 2000. It is very unusual to have insufficient soil moisture to initiate growth.
Differences in soil moisture can be found on the same farm.
Frost would slow down growth but would not necessarily stop it. Frost is not an insured peril.
In Mr. Samis’ area there are only two co-operators in a 15-mile circle; he believed it was the same two in 1998 and 2000.
There were 15-20 growers in the Bruce Mines area in 2000, including Mr. Samis.
In 2001 there were 1,398 insured growers in the forage plan in Ontario; an additional 300 growers participated in a pilot project for a new forage plan.
Species mix affects the amount of dry matter accumulated.
Grasses may shut down growth after an eight-day dry spell but alfalfa has deeper roots.
SIMFOY is a complex formula and it is difficult to determine the impact of a change in one variable.
A letter of explanation of his payout which AGRICORP sent to Mr. Samis did not specifically mention that co-operator yields are averaged within a 15-mile radius of the insured grower.
The co-operator average for Mr. Samis’ simulation was 77.2% of the normal first cut yield; this was an average of 84.15% from Mr. Samis and 70.82% from the other co-operator.
Mr. Samis estimated that his first cut was 50% of 2001 yield, but his 2001 yield was higher than normal.
The main reason for the difference in the amount of the payouts in 1998 and 2000 claim is the difference in co-operator yields.
A small change in rainfall data may not change the simulated result, as SIMFOY caps simulated first cut yield at 120% of normal. SIMFOY does not adjust its calculations according to the rate of rainfall, only to the amount of rain. That data is not available to AGRICORP.
Forage clients are not required to insure their entire crop.
Growers in counties and districts to the north of Renfrew County are assumed to harvest two cuts of hay.
Mr. Samis’ species mix might perform differently than the mix in SIMFOY.
When collecting soil samples, AGRICORP staff consult with the co-operators and try to test a representative part of the field.
Co-operator yields are taken on a field that is representative of the farming operation.
There are more accurate rain gauges available than those used by AGRICORP but they are much more expensive.
AGRICORP had not considered dropping the low end and high end rainfall data.
Rainfall data collected by co-operators does not affect the yield results of other growers.

Reply Evidence

In reply, Mr. Samis said his ‘wedge style’ gauge was clean, but discoloured. He indicated that he was not comfortable with the co-operator yield data entered for his farm and he had no recollection of signing a data collection form. He said he recalled giving his opinion regarding the percentage of his 2000 crop, which was of a normal crop. He also said that AGRICORP could not assume that other growers in his area were satisfied with its explanation of their payouts just because they did not appeal. He said he knew of several who were not satisfied. Mr. Samis said he bought hay in 1997, 1998 and 2000 and that payouts covered most of the cost in the first two years, but did not come close to covering his cost in 2000.

In response to questions, Mr. Samis felt that he should have been given a copy of any document that he signed. He had no explanation for the difference in his perception of his 2000 yield and the AGRICORP estimate based on its on-farm visit in 2000. He relied on his written record in coming up with his estimate that his 2000 crop was less than half of his normal yield. Unfortunately, Mr. Samis did not bring his written record.

Summations

Mr. Samis told the Tribunal that before the hearing he thought the April rainfall data, the initial soil moisture data and the ‘wedge style’ gauge were the problem. He said that after hearing the evidence, he believed that the lack of training he received as a co-operator and the importance of the co-operator data is also important.

Mr. Samis submitted that it was unreasonable of AGRICORP to use data collected in Sault Ste. Marie and Laird Township in the simulation of his forage yield because of differences in climatic conditions. He said he believes he is the easternmost insured grower in Algoma District. He said that he believed the ‘conical style’ gauge was more accurate than the ‘wedge style’ gauge and that AGRICORP erred in not standardizing the gauges after the first Tribunal recommendation. He also suggested that co-operator training was insufficient. Mr. Samis submitted that AGRICORP fell below a reasonable standard of care in each of these areas.

Mr. Samis asked that the Tribunal order AGRICORP to record his first cut yield at 60% of normal yield, re-run the simulation and pay him the difference between the new payout and his original payout. He also requested a $150 per diem and compensation for travelling to the hearing. He said he would also like to see training for co-operators and a more central weather station used in Algoma District.

Mr. Thomson said AGRICORP relied on purchased data and data collected from growers to make the model work. He said that AGRICORP does not usually challenge co-operators on the data they provide. He said that staff collecting the data from co-operators are trained and the forms are signed by co-operator growers. He said that AGRICORP does not modify the data it receives; it inputs what is submitted.

Mr. Thomson said that he understood the issue of location, particularly in the North. He said if more people participate in the forage program, AGRICORP receives better data. He said it was an Environment Canada decision as to where to put its weather stations and that AGRICORP’s only options are to rely on its data or deny service where there are insufficient growers.

Mr. Thomson said that rain gauges used in its pilot project cost $500 each and were very accurate. He said that when AGRICORP compared the response of SIMFOY with the results of the pilot project in the same areas, it saw the same trends. He said he did not know if the ‘wedge style’ or ‘conical style’ gauge was most accurate, but that he suspected that neither is perfectly accurate.

Mr. Thomson told the Tribunal the forage plan has never been sold as providing coverage for forage yield on an individual farm. He said that Mr. Samis did not purchase as much coverage as he was allowed so he should not expect his payout to cover the entire cost of replacement hay.

He submitted that Dr. Brown had reviewed the data in this case and was of the opinion that the simulation performed correctly given the data that was used.

Mr. Thomson agreed that ideally the simulation should closely correspond to the average co-operator yields, but submitted that this is often not the case. He said this was due to different management practices on individual co-operator farms. As well, he said that variations could occur if rain runs off a field because of rate of fall, as SIMFOY assumes it is absorbed into the soil. He said in Mr. Samis’ case AGRICORP had checked the data and could not override the data submitted.

The Findings

The Tribunal finds that there was no proof presented to it that there was a basic error made on the collection of rainfall data on the Samis farm in 2000. The Tribunal was impressed with Mr. Samis’ efforts to record and analyse rainfall data collected from two styles of gauges in 2001, but this does not prove that the data collected in 2000 was incorrect. The Tribunal is pleased that AGRICORP has chosen to standardize the rain gauges used by insured growers in the forage plan.

The weather pattern maps presented by Mr. Samis clearly show that his farm is not in a similar area as those where the weather data is collected. The Tribunal could not determine what impact, if any, this would have on his simulated yield. However, the Tribunal is concerned about the large variance between both co-operators’ actual yields and the simulated yield for the area.

There was conflicting evidence presented with regard to the percentage of normal yield that Mr. Samis harvested in 2000. AGRICORP’s records indicated the yield was approximately 84% of normal, whereas Mr. Samis estimated it as less than 50% of normal. Neither party provided written records that would substantiate their positions. Normally the Tribunal prefers to see more proof than was provided at this hearing before it orders a change in the adjustment of an insurance claim. The Tribunal finds Mr. Samis to be a credible witness, and the large disparity between co-operators’ yields and the simulated yield in his area supports his contention that an error could have been made in recording his actual yield data in 2000. Therefore, in this instance, the Tribunal is inclined to partially grant the appeal and order an adjustment to the claim.

As a result of this hearing, the Tribunal will amend its notices of hearing to require parties to bring all relevant documentation, including production and weather records to their hearings.

The Tribunal is of the view that when an AGRICORP client or a co-operator is asked to sign a document, a duplicate of the document should be left with the client or co-operator. This applies to the records of measurement of the first cut bales, and the estimate of yield by the client or co-operator.

The Tribunal notes that AGRICORP provided Mr. Samis with a detailed explanation of the manner in which his 2000 simulated yield was calculated. The Tribunal compliments AGRICORP on its much improved communications with appellants.

Decision and Reasons

After careful consideration of the evidence and submissions made, the Tribunal decided to grant the appeal in part for the following reasons:

Mr. Samis provided credible testimony that his actual yield in 2000 was less than that recorded by AGRICORP.
The very large discrepancy between the recorded yields of co-operators and the simulated yield in the area supported Mr. Samis’ position.

ORDER OF THE TRIBUNAL

The Tribunal orders AGRICORP to re-run the simulation of Mr. Samis’ 2000 yield using an actual co-operator yield of 66% of the normal yield for Mr. Samis.
The Tribunal orders AGRICORP to pay Mr. Samis the difference between the claim payable to Mr. Samis based on the new simulation, and the amount paid to him on the basis of the original simulation.
The Tribunal orders that the re-calculation of Mr. Samis’ simulated yield is to have no impact on any other insured grower in his area.

DATED AT Guelph, Ontario this 12th day of December, 2001.