Interpreting Risk Index Scores

Index Structure

The Pesticide Risk Mitigation Engine (PRiME) has been designed as a probabilistic risk ranking tool, as opposed to an absolute measure of environmental harm. It is strongest in its ability to compare pesticide options relative to each other and to assess the usefulness of various mitigation measures. For some of the indices (see descriptions below), field results have been used in order to calibrate the risk scores against actual ecological damage.

Risk index scores in PRiME range from zero to one. This number represents the probability of an undesirable effect or, in the case of chronic risk indices, the proportion of the breeding season where reproduction is inhibited. For example, if the undesirable effect for the avian acute index is the killing of birds in and around a crop field, then a score of 0.1 means that there is an estimated 10% chance that the pesticide application will kill birds; a score of 0.5 means a 50% chance, and so on.

Risk is presented on a low/moderate/high scale. We consider risk index scores below a 10% chance of an undesirable effect to be in the low risk category, where no further risk mitigation is needed. Realistically, 10% is within the margin of error for our risk models. Risk index scores between 10% and 50% fall into the moderate risk category where risk mitigation is recommended. Risk index scores above a 50% chance of an undesirable effect fall into the high risk category.

Reading the Risk Outputs

Image one (below) shows a risk summary for two products grouped by risk index.

Image 1 (risk scores grouped by index)

The risk indices are shown on the y-axis, while the risk index scores are shown on the x-axis. The bars indicate the risk scores for each pesticide application in each index. The legend in the left hand corner shows which color represents which application. Also, by hovering your mouse over a bar, you can display which product the bar represents.

Image two (below) shows a risk summary for the same two products but this time grouped by product.

Image 2 (risk scores grouped by product)

The products are shown on the y-axis, and the risk index scores are shown on the x-axis. The bars indicate the risk scores for each index. The legend in the left hand corner shows which color represents which risk index.

Image three (below) shows a cumulative risk summary for several products applied over the course of a growing season.

Image 3 (cumulative risk summary)

The box in the upper left hand corner reports the total number of outputs for all applications and all risk indices. This total is then broken down into five categories: (1) High Risk scores, (2) Moderate Risk scores, (3) Low Risk scores, (4) Passes and (5) Missing scores.

A “Pass” is given when PRiME lacks the data necessary to make the calculation, but the application is nevertheless considered low risk because the active ingredient is presumed to be of low hazard or have a low exposure potential for the given endpoint. In the case of a missing score, PRiME lacks the data necessary to make the calculation.

The bar chart gives a graphical representation of the cumulative risk for each risk index. The length of the bar represents the cumulative probability of an undesirable effect of all applications combined for each risk index. The colors within each bar indicate the number of individual risk scores that fall within each risk category.

For example, the bar for the Aquatic Invertebrate Risk Index (above) shows a cumulative score near “1.” A total of 13 calculations went into this cumulative score, of which eight were in the low risk, or “yellow,” zone, four were in the moderate risk, or “orange,” zone, and one was in the high risk, or “red,” zone.

The Meaning of Risk Index Scores

Avian Acute

  • Undesirable effect: Visible mortality defined as mortality that would be detected in the course of an intensive and competent search of the treated area given the normal complement of species in and around the treated field.
  • Index score: The probability that a given application will give rise to visible bird mortality. This probability has been calibrated against actual field studies carried out with a variety of pesticides.

Avian Chronic

  • Undesirable effect: Birds breeding in and around fields are prevented from breeding because of persistent residues in food sources.
  • Index score: The proportion of the typical breeding season (90 days) with residue levels such that reproduction is compromised in a reasonably sensitive 15g songbird. In other words, the worst possible score of 1 would be an application that potentially inhibits avian reproduction for the entire length of the “normal” breeding season. There is no field information with which this probability can be verified; however, the determination of risk is based on laboratory studies of reproductive effects as well as accepted principles of exposure.

Small Mammals

  • Undesirable effect: Population-level effects on mortality, fertility or recruitment.
  • Index score: The probability that residues will persist long enough at a toxic level to cause changes in the population trajectory of small mammals directly exposed to the spray. This probability has been calibrated against actual field studies carried out with a variety of pesticides.

Earthworms

  • Undesirable effect: Loss of 35% or more of earthworm biomass.
  • Index score: The probability that substantial earthworm mortality (i.e. > 35% of a species of average sensitivity) will result from the pesticide application. This probability has been calibrated against actual field studies carried out with a variety of pesticides.

Fish

  • Undesirable effect: Fish in nearby receiving waters are prevented from breeding due to levels of pesticide residues exceeding the Maximum Acceptable Toxicant Concentration (MATC).
  • Index score: The proportion of the typical fish breeding season (30 days) where residue levels exceed the MATC. In other words, the worst possible score of 1 would be a pesticide application that inhibits reproduction in a reasonably sensitive fish species for the entire 30 day period. There is no field information with which this probability can be verified. However, the determination of risk is based on laboratory studies of reproductive effects as well as modeled estimates of exposure.

Aquatic Invertebrates

  • Undesirable effect: More than approximately 10-20% of invertebrate taxa are significantly impacted by treatment, meaning that their population is often decreased by 10 fold or more.
  • Index score: The probability that an application will give rise to this level of impact. This probability has been calibrated against actual field studies carried out with a variety of pesticides applied at known concentrations as well as modeled estimates of exposure.

Algae

  • Undesirable effect: More than approximately 20-30% of algal species are significantly impacted by treatment, meaning that their growth is impeded, usually by 10 fold or more.
  • Index score: The probability that an application will give rise to this level of impact. This probability has been calibrated against actual field studies carried out with a variety of pesticides applied at known concentrations as well as modeled estimates of exposure.

Inhalation Risk

  • Undesirable effect: Workers/bystanders are exposed to a pesticide at a concentration that would result in dose for a one-year old child that exceeds US EPA's short-term Population Adjusted Dose (PAD). The PAD is the Reference Dose divided by all uncertainty factors (interspecies, intraspecies, FQPA and other).
  • Index score: The probability that someone spending a significant amount of time (e.g. living, working) within 100 feet of the treated field will be exposed to the pesticide at a level exceeding the REL.

Missing Data

In some cases, PRiME will not have enough data required to make a risk calculation for a given index. A risk index could fail to calculate because we are missing the necessary physical and chemical properties (e.g. foliar half-life) for the active ingredient or because we are missing the necessary toxicity values for the end point of concern. If a risk bar is present, or a number is given (e.g. 0.00, or <0.01), then a calculation was made. In cases where no calculation is made, a risk score may be replaced by one of the following warnings or pass codes, indicating presumed low risk.


PRiME Risk Summary Legend
0.00 Risk index score is 0.00
<0.01 Risk index score is less than 0.01
Insufficient data for a risk calculation
LE Assumed low risk due to low exposure potential
LH Assumed low risk due to low hazard potential
LEH Assumed low risk due to low exposure and low hazard potential
W1 Rodenticides are extremely toxic to vertebrates. The key to safe use of rodenticides is to exclude all but the target species from the bait through the use of efficient bait boxes. Dead rodents should be disposed of in a way that will make it impossible for scavengers to find.
W2 Although standard toxicity data are lacking, this material is known to be a strong irritant and toxic by ingestion or inhalation.
W3 The use of any copper product leads to an accumulation of copper in the soil which kills earthworms. This can have negative consequences on disease management.
W4 Oils do not represent a toxicological risk, but the consequences of fouling of bird feathers has not been assessed.
W5 Forms glassy sharp coating on surfaces which may damage earthworms and other invertebrates.
W6 No data are available; however, EPA has waived basic data requirements believing risk to be minimal.
W7 This material is registered as an insecticide. Although bee toxicity data is not available, it should not be considered as necessarily safe to bees if used in a way that will expose bees and wild pollinators.