Summary of County Survey Questions Pertaining to Acres Planted

Background

In September 1999, a survey was mailed to all County Agricultural Commissioners (Appendix A). The entire survey was designed to better understand how the counties record data and interpret certain data fields in the Pesticide Use Report (PUR). Specific questions regarding acres planted were incorporated to better understand the accuracy of acres planted information in the PUR and to search for potential factors relating to the discrepancy between acreage from the PUR and other sources seen in certain commodities. By gathering this information and refining our data collection process, we will ultimately be able to improve the accuracy of data in the PUR. This document summarizes the questions regarding acres planted. The remainder of the survey questions will be summarized at a later date.

The California Department of Pesticide Regulation’s (DPR’s) PUR is probably the largest and most complete database on pesticide use in the world today. A system to collect information on pesticide use in California has been in operation in some form for over 50 years, with the current use reporting system started in 1990. The PUR contains information on nearly all production, agricultural, and some nonagricultural pesticide use in California.

The data are collected by the counties and can include information such as: the pesticide product used, the date it was applied, the acres treated, as well as the number of acres planted for a given commodity.

DPR expanded pesticide use collection in 1990, primarily to more accurately assess dietary risk as well as exposure and potential risk to workers. In certain risk assessments, the percent acres treated with a given pesticide or group of pesticides is an important measure. The percent of acres treated with a pesticide is based on the number of acres planted for a given commodity or commodities. Therefore, accuracy of the acres planted information in the PUR is important if our risk assessments are to be meaningful.

In addition, there is interest in discrepancies between the acres planted information in the PUR and other sources of information on planted or harvested acres. One source of such data is the county crop report, which summarizes harvested acres in a county for each commodity. For certain commodities, such as almonds, there is reasonable agreement (within 10%) between acreage from different sources. For other commodities, such as strawberries, the PUR acreage is often as much as twice that from other sources.

Study Methods

A survey was constructed to clarify how the counties interpret certain data-field definitions and how these data are updated or revised during the period of data collection (Appendix A). The survey was designed in consultation with staff from DPR, the California Department of Food and Agriculture, the Sacramento County Agricultural Commissioner’s Office, and the U.S. Department of Agriculture. The surveys were distributed to all counties in California where pesticide use information is collected: a total of 55 single or two county combinations. The entire survey is attached in Appendix A and consists of six groups of information: general, operator identification number (or permit number), site location, acres planted, acres treated, and additional field questions.

This summary will focus on responses to the questions concerning acres planted and a related question from the site location group of information. These questions include:

1. Acres planted (AP) 1. Are the estimated acres planted entered into the Pesticide Permit/Operator ID Programs and automatically transferred to the PUR?
2. AP 2. Do operators generally update the estimated acres planted for a crop on their permits during the year? If yes, under what circumstances?
3. AP 3. If an operator updates the acres planted during the year, are these numbers (acres planted) corrected retroactively for that crop cycle (planting through harvest) in the PUR, or only from the date of the update forward, as automatically provided by the software?
4. AP 4. At any time during the calendar year do planted acres differ from actual acres in the field? If yes, under what circumstances?
5. Nuance (Nu) 11. If a crop is planted in the fall of 1997, harvested in the spring of 1998, and the same site location ID is used for the field each of these years, do the planted acres in 1998 refer to the over wintered planting still in the ground in the spring, or to the subsequent fall planting later in 1998? Standard options for responding to this question include: current planting, fall planting, and not applicable.
6. AP 5. How does your county estimate planted/harvested acres of a particular crop for the County Crop and Livestock Report? Please describe this process, and list other sources of information.

Responses to each question (organized alphabetically by county) are included in Appendix B. Responses are also summarized in tables one through three and figures one through three.

Results and Discussion

The surveys were distributed to all 55 county agricultural commissioner offices. All 55 counties responded to this survey (Table 1). Response rates for individual questions ranged from 95% to 100% and indicate a high level of commitment to collecting and maintaining pesticide use records.

Response by Question

AP 1. Are the estimated acres planted entered into the Pesticide Permit/Operator ID Programs and automatically transferred to the PUR?

In certain counties (especially San Benito, San Luis Obispo, Santa Barbara, and Monterey), acres planted reported in the PUR for a single site, often vary (up and down) with acres treated, even for permanent crops, rather than remain constant during a single year. Therefore, this question, AP 2 and AP 3 (below) were included in the survey to determine if a potential reason for this pattern lies in how acres planted are actually entered and edited in the PUR database.

Of the 55 county responses, 76% indicated acres planted are entered into the permit program and automatically transferred to the PUR (Table 1, Appendix B). This procedure ensures the original information submitted by the grower when applying for a permit/operator ID is actually transferred, without error, to the PUR.

In contrast, 20% of the counties (11 counties total) indicated acres planted are not entered into the program and/or the transfer of this information is not automatic. These counties include Del Norte, Humboldt, Inyo/Mono, Lassen, Modoc, San Bernardino, San Luis Obispo, Tuolumne, Ventura, Yolo, and Yuba. Del Norte does not use a computer program to issue permits and Lassen does not use a DPR program to enter data into the PUR. The remaining nine counties, however, use the DPR computer programs to enter permit and PUR dat (data not shown) so it is likely that an automatic transfer could occur. From responses to the survey, it is unknown if acres planted are simply not entered into the permit program or somehow the transfer to the PUR does not take place. This question could be used as a follow-up question for these counties.

Two of the 55 counties (Marin and Santa Cruz) had a response categorized as other. We interpret the Marin County response to mean the data are automatically transferred since they use the computer program designed by DPR and have not developed any additional guidelines. In Santa Cruz County, acres planted are not always transferred automatically, indicating the county may modify data in the PUR or opt not to enter acres planted at the time a permit is issued.

AP 2. Do operators generally update the estimated acres planted for a crop on their permits during the year? If yes, under what circumstances?

This question was asked to determine if operators (growers) tend to revise their original acres planted estimates. Since many factors lead to changes in an estimate of acres planted (Table 1), updating this information is important for the accuracy of these data in the PUR.

Of the 55 county responses, 51% indicated operators generally update acres planted information while 45% said they do not (Table 1). Counties in the Sacramento Valley and northern San Joaquin Valleys generally update, while southern San Joaquin Valley and most coastal counties do not (Figure 1). Two counties had a response categorized as other. San Francisco indicated they had a fixed amount of acreage, implying that updates were not common. Santa Barbara indicated vegetable farmers update acreage on a Notice-of-Intent or PUR while growers with permanent acres will notify the county.

There were a number of circumstances reported that lead to updates in acres planted initially estimated on a permit. General categories of circumstances that lead to updates include weather, water, seed availability, grower changes, discrepancies between permit and PUR acreage, and errors in acreage estimates (Table 1).

AP 3. If an operator updates the acres planted during the year, are these numbers (acres planted) corrected retroactively for that crop cycle (planting through harvest) in the PUR, or only from the date of the update forward, as automatically provided by the software?

This question was asked to determine how the counties update acres planted in the PUR. Of the 52 counties that replied to this question, 29% update acres planted retroactively, indicating acres planted should remain the same throughout a single year in the PUR. In contrast, 63% update acreage from the date of update forward (Table 1 and Figure 2). In these counties, the acreage reported closer to the end of a given year might more accurately reflect the acres planted for that year. Four counties (8%) had a response categorized as other (Amador, Kern, San Bernardino, and Shasta).

The response to AP 3 is in apparent conflict with AP 2, where approximately half the counties did not generally update. We interpret replies to question AP 2 to mean that 50% do not generally update their acreage. Replies to AP 3 indicate that if an update is made, most counties (65%) simply update from the date they were notified of a change forward, a procedure automatically provided by the DPR software. This indicates the latest application date in the year might be the most accurate reflection of acres planted and potential harvested acres since a majority of counties do not update the PUR retroactively.

AP 4. At any time during the calendar year do planted acres differ from actual acres in the field? If yes, under what circumstances?

We attempted to address the accuracy of acres planted in the PUR in questions AP 1 through AP 3. In question AP 4, we tried to determine other possible reasons why acres planted in the PUR might not be accurate. Of 54 counties that replied to this question, 59% indicated there is a difference between planted acres and actual acres while 37% said there is not (Table 1). Most of the coastal and Sacramento Valley counties, and several San Joaquin Valley counties indicated there is a difference (Figure 3). Reasons cited for this discrepancy include changes in grower decisions on what and how much to plant, buffer zones, weather, crop failure, fire, fallow ground in winter or between plantings, etc. Responses to AP 3 and AP 4 indicate that at any given time during the year, acres planted in the PUR may not reflect the acreage for a given crop in the field.

Nu 11. If a crop is planted in the fall of 1997, harvested in spring of 1998, and the same site location ID is used for the field each of these years, do the planted acres in 1998 refer to the over wintered planting still in the ground in the spring or to the subsequent fall planting later in 1998? Standard options for responding to this question include: current planting, fall planting, and not applicable.

In the PUR, acres planted are nearly twice that reported by other sources (such as the County Crop Reports and the California Agricultural Statistics Service) for certain over-wintering crops. It is believed the current reporting system may not be specific enough to distinguish acres planted in different calendar years at the time of permit issue. In all counties except Imperial, growers (operators) apply for their permits at the beginning of the calendar year. Hence, it is unknown if growers applying for a permit in January are actually submitting acreage planted for the prior fall planting or for the planting to be made later in the calendar year. Therefore, this question was designed to better understand how growers/counties report acreage for crops spanning two calendar years.

Sixty-four percent (33 of 54 counties) indicated the acres reported in 1998 refer to those acres planted in 1997 and harvested in 1998 (i.e., current planting, see Table 1). A sizable percentage (30%) indicated this question is not applicable, perhaps because over-wintered crops are not common in their county. Imperial County is a notable exception in that their permit year begins July 1. In addition, Kings County uses a different site location ID to distinguish crops grown under these circumstances.

AP 5. How does your county estimate planted/harvested acres of a particular crop for the County Crop and Livestock Report? Please describe this process and list other sources of information.

This question was included to gain additional information about the sources of information used to estimate harvested acres in the County Crop Reports. There were nine general methods by which counties estimated harvested acres for the County Crop Report. A grower survey, interview, or contact with a grower group was used by 73% of the 55 counties that responded to this question (Table 2). Use of the permit/PUR programs was the second most common method (58% of 55 counties), followed by state/federal data (33%), industry information (31%), processors/packers (15%), county agricultural commissioner staff information (11%), U.C. Cooperative Extension 11%), and county assessor maps/information (9%). Most counties (78%) use more than one method to estimate harvested acres (Table 3). Only 12 counties use one method, and the method of choice for eight of those counties is the grower survey/interview. Therefore, none of the counties used the PUR as the sole source of information to estimate harvested acres.

Summary

In general, county responses to these questions were clear. However, responses to certain questions indicate some degree of confusion about the meaning of the question and/or what functions the DPR computer programs perform. Clarification with the counties in follow-up interviews would help to clarify the practices used by individual counties.

The questions described above were incorporated into the survey to examine the accuracy of acres planted reported in the PUR and to identify potential reasons for the discrepancy with harvested acres in the County Crop Reports. Three-fourths of the counties use the DPR program to automatically transfer acres planted from the permit database to the PUR, ensuring consistency with the original database. In addition, about 50% of the counties indicate operators generally update their permitted acreage as changes occur during the year. About two thirds of the counties make updates in the PUR only after the date changes were made to the permit program. These responses indicate the variation in county procedures used to collect data for the PUR and emphasize the need to understand those procedures when using acres planted information in the database.

About two-thirds of the counties indicate that permits issued in January for over-wintering commodities refer to the currently planted fields, not the fields to be planted later in the year of the permit. However, it is still not clear if a different site location ID is used to indicate the subsequent planting that might happen at the same location once the crop is harvested. If this is the case, it would actually double the acres planted that are reported at a single location. Since this question was developed to address specific commodities and counties, it is recommended that follow-up questions be posed to those counties about the commodities in question.

According to our survey, a number of methods are used to estimate harvested acreage in the county crop reports. In addition, most counties do not use the PUR alone to estimate harvested acres for their crop reports. A primary reason for this may be that the PUR records acres planted not acres harvested. Also, the PUR will only reflect the true acres planted if updates are routinely made, if pesticide applications are reported for all fields, and if acreage for fields without pesticide application are available. Errors in the PUR, changes in grower planting, weather variables, ambiguity in PUR data collection methods, and inaccuracies in other sources of information used to estimate harvested acres are all possible reasons for discrepancies seen between the different sources of information. When comparing acreages from different sources, it is important to review this information to get an idea of the specific methods used in a particular county. Additional information may have to be solicited from the county to gain a more complete understanding of any discrepancies seen.

We hope to continue our review of the county responses to this survey and develop further information in the future. If you have any questions, please contact Dr. Lisa Ross, Senior Environmental Research Scientist, at (916) 324-4116 or lross@cdpr.ca.gov.

Sincerely,