Research on Glyphosate Estimation Methods

[lindsaypoirier]

It would be helpful to know more about how these estimates of pesticide use were generated.

This page provides some basic information.

Specifically, we should try and extract:

• What are the different organizations involved in producing this data, and what role does each organization play?
• What is a crop reporting district, and what is its relationship to a county?
• How do pesticide use estimates get aggregated from the CRD to the county level?
• What is the difference between the EPest-low and EPest-high methods?
• What does USGS mean by "all states except California"? How are estimates being made in California?
• Why does the reliability of the estimates decrease with scale?

[mservin310]

I have tried these links on different browsers and the sites seem to be down. I'll keep checking periodically but they haven't been working for a couple days now... @lindsaypoirier

[shapironick]

hmm! they seem to work for me.

1. https://water.usgs.gov/nawqa/pnsp/usage/maps/show_map.php?year=2010&map=GLYPHOSATE&hilo=L

2. https://water.usgs.gov/nawqa/pnsp/usage/maps/about.php#epest-hilo

but USGS does need to up update their SSL cert

so strange!

[mservin310]

I was able to access the links using a VPN for the UCLA network. ^ Just plugging this in here for anyone's reference.

[mservin310]

What are the different organizations involved in producing this data, and what role does each organization play?

• USGS - put dataset together from various sources

• California Department of Pesticide Regulation Provides data on California’s pesticide use

• U.S. Department of Agriculture Data from census of agriculture: total count of U.S. farms and ranches (>$1000 of product sold/yr)

• U.S. Department of Agriculture National Agricultural Statistics Service (NASS) annual harvested-crop acreage (used in conjunction with crop reporting district data)

• Multi-Resolution Land Characteristics Consortium Land classifications database used for mapping county-level use estimates to agricultural land within each county

What is a crop reporting district, and what is its relationship to a county?

• Statistical units which typically include eight to ten counties and provide a reasonable intermediate aggregation level between an individual county and a whole state

• Aggregated county levels based on crop ➜ One step above an individual farm

How do pesticide use estimates get aggregated from the CRD to the county level?

• EPest crop-use rates were developed on the basis of pesticide use on harvested acres in multi-county areas (Crop Reporting Districts) and then allocated to county harvested cropland

What is the difference between the EPest-low and EPest-high methods?

• Both use proprietary surveyed rates for Crop Reporting Districts; differ in how situations are treated when a CRD was surveyed & pesticide use was not reported for a crop present in CRD

• EPest-Low: assumes zero use in the CRD for that crop

• EPest-High: treats missing unreported use as missing data missing data values ➜ uses pesticide-by-crop rates from neighboring CRDs or CRDs in same region are used to estimate the EPest-high rate for that CRD

What does USGS mean by "all states except California"? How are estimates being made in California?

• 1990: California first state to require full reporting of agricultural pesticide use in response to demands for more realistic and comprehensive pesticide use data-all agricultural pesticide use must be reported monthly to county agricultural commissioners, who then report the data to Department of Pesticide Regulation (most comprehensive globally)

• All forms of pesticide use must be reported in California (only exceptions are home-gardens and most industrial & institutional uses)

• Data is already in county level estimates; low and high rates are the same for counties in CA

Why does the reliability of the estimates decrease with scale?

• Not suitable for county-level estimates: surveyed pesticide-by- crop use was not available for all CRDs and, therefore, extrapolation methods were used to estimate pesticide use for some counties.

• surveyed pesticide-by-crop use may not reflect all agricultural use on variety of all crops grown within a county

[shapironick]

• [ ] Search for lists that indicate the most toxic pesticides that have been used since the data set was created (check when that was, some pesticides may have been discontinued in the middle of this data).

• [ ] Search for lists of banned pesticides.

Search the data for these pesticides and create a list of pesticides that you think we should focus on. Also create the corollary list of pesticides that are not of high priority.

qualities to keep in mind: carcinogenicity, neurotoxicity, reproductive toxicity, endocrine disruption, bio-accumulation, half-life.

Start structuring the spreadsheet w/ your responses only after you've started the research. One good way to start might be to find every unique pesticide in this data set and make a spreadsheet with them and annotate which expert bodies think which pesticides are hazardous

[shapironick]

You could experiment with this function to get a list of unique pesticides within the dataset: http://www.datasciencemadesimple.com/unique-function-in-r/