Imidacloprid and Groundwater   Edward Frank
  Jul 18, 2007 09:10 PDT 

There has been some concerns expressed on the use of imidacloprid based upon discovery of the chemical in monitoring wells in Long Island, NY.  Below are excerpts from a document entitled  ”HYDROGEOLOGIC FRAMEWORK OF LONG ISLAND”   This outlines the overall geology of Long Island.  The surface materials of the island consist primarily of glacial drift and outwash including materials primarily of  sand and gravel sizes.  The text from Table 1 states:  “Mainly sand and gravel of moderate to high permeability; also includes clayey deposits of low permeability.”   The clays are generally incorporated as lenses within the larger grained deposits and tend to not serve as a barrier to water migration. 


 The permeability of this material is very high with a porosity of 30 to 35% primarily in interconnected pore spaces.   The horizontal hydraulic conductivity (K) of this surface deposit was measured at 230 ft/day and a vertical hydraulic conductivity of 23 feet per day was measured in the sediments.  Hydraulic conductivity, symbolically represented as K, is a property of soil or rock, that describes the ease with which water can move through pore spaces or fractures.These are very high values.  In typical soil materials the hydraulic conductivity is measured in inches or fractions of inches per day. 



K (ft/day)














Relative Permeability








Unconsolidated Sand & Gravel

Well Sorted Gravel

Well Sorted Sand or Sand & Gravel

Very Fine Sand, Silt, Loess, Loam


Unconsolidated Clay & Organic



Layered Clay

Fat / Unweathered Clay

Consolidated Rocks

Highly Fractured Rocks

Oil Reservoir Rocks

Fresh Sandstone

Fresh Limestone, Dolomite

Fresh Granite

Source: modified from Bear, 1972


Long Island is essentially a giant pile of sand.  On Long Island are acres and acres of lawn and many gold courses.  Private land owners to some extent, and gold courses use extremely large amounts of pesticides to treat their lawns.  A major if not the major chemical used for insect control on the gold courses is imidacloprid.  If you dump tons of water containing imidacloprid on a giant pile of sand some of it will seep through into the groundwater aquifer. 


The comparison to the situation in Long Island to that of individual hemlocks or hemlock groves being treated with imidacloprid to kill the Hemlock Wooly Adelgid (HWA) is unrealistic and invalid.  The amount of pesticide being used differs by several orders of magnitude as does the typical hydraulic conductivity of the soil sediment.  There are two scenarios areas in which there may be some potential for  imidacloprid to leach into the groundwater in trace amounts.  The first would be a scenario as exists on Long Island where the surface materials consist of well sorted sand and gravels.  These may be found in some outwash and till deposits, or on barrier islands, bars, or capes.  The other scenario are those in karst terrains.  Karst terrains are those in which the major land features are derived or significantly altered by dissolution of bedrock – essentially limestone and gypsum bedrock areas.  These if well developed have few or no surface runoff as all the water not evaporated or transpirated  flows into the bedrock through discrete crevices, sinkholes, or caves.  If imidacloprid were poured into such an opening it would reach the groundwater table.


Bayer, developer of imidacloprid reports:


“What happens to imidacloprid in the environment?

The success of imidacloprid as a crop protection product would have been unthinkable had its short and long term impact on the environment brought about any adverse effects or irreversible changes. Assuming a use pattern which guarantees the desired protective effects, the behavior in and between the compartments of the environment, biosphere, soil, water, and atmosphere, depends on physico-chemical and chemical properties defined principally by the chemical structure. Climatic differences and diversities of the soils must also be taken also into account. Which of the theoretical outcomes will predominate following application of a crop protection product, degradation, persistence, binding to soil, volatilisation, translocation into groundwater, runoff into surface due to rainfall after application, must be established either from physico-chemical data or by direct measurements.

In the case of imidacloprid it was proven beyond doubt that persistence of residues in soil due to repetitive application over several years; translocation into deeper soil horizons, groundwater, adjacent crops or surface waters; volatilisation; and transport through the air into other regions can be ruled out. This has been confirmed again and again by world-wide and long-term experience following its use in all major crops.


Degradation in soil

There is broad evidence from research at Bayer, as well as from independent sources that imidacloprid is degraded continuously though not very rapidly. Practical trials conducted under northern European conditions showed the half-life for dissipation to be less than six months.

Degradation ends with complete mineralization to carbon dioxide, though binding of intermediate degradates to soil also occurs. It is important to draw a line between relatively long lasting residence time and persistence in the soil. Imidacloprid cannot be classified as being persistent as it does not accumulate. Long-term trials under worst case conditions with the repeated use of imidacloprid over several years have demonstrated that maximum concentrations in soil will reach a plateau and will decline if no further applications occur.

Mobility in soil and leaching into groundwater

The translocation behaviour and particularly the leaching potential of a crop protection chemical from soil into groundwater is equivalent to its inclination for hydrophilic interactions or for interactions especially with water. Imidacloprid contains in its molecular structure substituents which cause a relatively high water solubility and a low affinity to hydrophobic structures found in ordinary organic matter. The parameters, which characterise this affinity, are the partition coefficient for the system octanol-water (Pow-value) and the soil adsorption coefficient normalised to the content of organic carbon (Koc-value). Pow and Koc-values are in a range where, translocation in soil and from soil is still negligible under ordinary conditions, but where the mobility is already sufficiently high for systemic action into the roots of plants or within plants for pest control.

Behaviour in water

Though imidacloprid is not intended to be applied directly in water, it nevertheless may enter water bodies due to spray drift or in extreme situations by runoff from treated fields after rainfall. It has been shown that no unacceptable harmful effects would occur under these circumstances as the substance will undergo complete elimination from water by photolytic reactions and by microbial activity. Though the substance is stable in sterile water in the dark, it decomposes readily under the influence of light. Biotic processes under the influence of microbes present in natural water and its sediments present another mechanism for the elimination of imidacloprid.”




The key points in this article is that the “Pow and Koc-values are in a range where, translocation in soil and from soil is still negligible under ordinary conditions, but where the mobility is already sufficiently high for systemic action into the roots of plants or within plants for pest control.”  The material does not translocate in the soils. It does not accumulate over time except when it is being continuously applied.  It begins to decline in concentrations as soon as applications of the chemical ceases. And finally it degrades rapidly in water in the presence of sunlight and bacteria. 


Any comparisons between the slight contamination found at long Island, essentially a giant sand and gravel pile on which tons of pesticides were dumped, and the treatment of individual trees with a ground injection application of a few ounces of imidacloprid in materials with generally magnitudes lower permeability are  inappropriate.


Edward Forrest Frank, Hydrogeologist

July 18, 2007







Nassau and Suffolk Counties with close to 3 million people are completely dependent on groundwater for all of their freshwater needs. As a result the hydrology of Long Island has been extensively studied. Long Island is completely surrounded by salt water.


The unconsolidated materials that overlie the bedrock constitute Long Island ’s groundwater reservoir. These materials can be classified into several hydrogeologic units on the basis of hydrologic properties summarized in the following table. Three major aquifers can be identified: an Upper Glacial aquifer at the top, the Magothy aquifer in the middle and a deep less accessible Lloyd aquifer lying just above the Paleozoic metamorphic basement rocks.


Table 1. Major hydrogeologic units of the Long Island groundwater reservoir.




Approximate Maximum Thickness

Water-Bearing Character

Upper glacial aquifer


Mainly sand and gravel of moderate to high permeability; also includes clayey deposits of low permeability.

Gardiners Clay


Clay, silty clay, and a little fine sand of low to very low permeability.

Jameco aquifer


Mainly medium to coarse sand of moderate to high permeability.

Magothy aquifer


Coarse to fine sand of moderate permeability; locally contains highly permeable gravel, and abundant silt and clay of low to very low permeability.

Raritan Clay


Clay of very low permeability; some silt and fine sand of low permeability.

Lloyd aquifer


Sand and gravel of moderate permeability; some clayey material of low permeability.


























Groundwater Hydrology


Precipitation enters the groundwater system by infiltration through the porous soil at Long Island’s surface. The field capacity of soils in the unsaturated zone is ~ 10-15%, and since the total porosity of the capillary fringe is ~ 30-35%, the specific yield of these materials is estimated to be on the order of 15-25%. A relatively high percentage of wet precipitation can percolate through the water table into the saturated zone.


The groundwater flow has vertical and horizontal velocity components. To simulate the flow, realistic estimates of the vertical and horizontal conductivities are required. Table 5 compiles conductivity values (inferred from pumping tests) which were adopted in a recent USGS study.


Table 5. Horizontal and vertical hydraulic conductivities.


Hydrostratigraphic Unit

Horizontal Hydraulic Conductivity (ft/day)

Vertical Hydraulic Conductivity (ft/day)



Upper glacial



















Confining units


not estimated



not estimated


The thing we need to remember here is that we are not talking about large scale applications in which large areas are drenched with imidacloprid, but directed application to hemlocks through ground injections or surface drenching at the base of selected scattered hemlocks. We are not talking golf courses or sub-urban lawns. I don't think bees being threatened will be a factor in the targeted applications as bees do not feed upon hemlock...

An off-list  response referred to this document:

"A toxicology extension service run by Oregon State, Cornell, and others ( suggests that:

Effects on Aquatic Organisms: The toxicity of imidacloprid to fish is moderately low. The 96-hour LC50 of imidacloprid is 211 mg/l for rainbow trout, 280 mg/l for carp, and 237 mg/l for golden orfe. In tests with the aquatic invertebrate Daphnia, the 48-hour EC50 (effective concentration to cause toxicity in 50% of the test organisms) was 85 mg/l (1[Kidd and James. 1991. The Agrochemicals Handbook]). Products containing imidacloprid may be very toxic to aquatic invertebrates. Breakdown of Chemical in Surface Water: The half-life in water is much greater than 31 days at pH 5, 7 and 9. No other information was found.

Here is data below on the concentrations observed in Long Island. As you can see the levels detected are at a maximum of 6.9 ppb. This maximum is 30,000x less than the 210 mg/l concentrations that had 50% toxicity IN 96 hours for trout and carp, and 10,000x less than the 50% toxicity for daphnia.

I expect that soil invertebrates in the immediate area of application will be killed by the pesticide. That is what the pesticide is designed to do. But since this pesticide is applied to just the area around the base of the trees, that will not effect the overall population of these invertebrates in the forest in general.

Ed Frank

Letter from Bayer

Imidacloprid groundwater monitoring on Long Island has been ongoing since 1998 Less than 1% of all samples taken had any detections of imidacloprid residues and all were low levels (<7 ppb). These low-level detects are consistent with modeling originally and independently conducted by Bayer, EPA and NYSDEC prior to state approval. Upon state approval, an agreement between Bayer and NYSDEC established mitigation triggers of 10 and 25 ppb based on the NYSDEC default of 50 ppb (this is compared to EPA’s informal MCL of 399)

Imidacloprid NYS DEC Letter - Status of Imidacloprid in New York State 10/03

To date, imidacloprid has been identified at low levels (parts per billion), in groundwater samples from approximately twenty monitoring and private wells in Nassau and Suffolk Counties. Most groundwater
detections have been very low, ranging from 0.1 ppb to 2.0 ppb. However, imidacloprid has been found in clusters of private wells down-gradient of farms, and a recent private well sample in Suffolk County contained 6.7 ppb imidacloprid. Additionally, imidacloprid has now been detected at a golf course monitoring well and at monitoring wells near trees that have been treated with imidacloprid by injection. ..    The United States Environmental Protection Agency (USEPA) Health Advisory Level (HAL) for imidacloprid is 399 ug/l.

Imidacloprid NYS DEC Letter - Registration of New Imidacloprid Products in New York State as Restricted-Use Products 10/04

The first detection of imidacloprid in a private homeowner well (far removed from the intended monitoring zone) was in April 2000. To date, imidacloprid has been detected at concentrations (0.2 to 7 ppb) in 12 
monitoring wells and 16 down gradient private homeowner wells.  Imidacloprid has also been recently detected at 0.24 ppb in two Suffolk County community water supply wells (85 feet and 90 feet deep).  Additionally, imidacloprid has now been detected at a golf course monitoring well (0.43 ppb) and at monitoring wells near trees (0.2 to 5.1 ppb) that have been treated with imidacloprid by trunk injection for the Asian Longhorned Beetle (ALB). 
***See Notes below


Re: Imidacloprid and Groundwater   Edward Frank
  Jul 19, 2007 02:47 PDT 

The last sentence quoted below refers to a concentration of imidacloprid of 0.2 to 5.1 ppb in monitoring wells near some trees that had been stem injected. I can't see how the imidacloprid can get out of the trees and back into the groundwater. Do any of you with a better grasp of tree physiology have an explanation for this? My first impression would be that the source of the contamination was not the trees that had been injected. Ideas?????

Ed Frank

Re: Imidacloprid and Groundwater   DON BERTOLETTE
  Jul 19, 2007 14:23 PDT 


Although it's a basic handling of a complex process, a quick I-net search yielded the following explanation (
Roots absorb water by a process known as osmosis, whereby water with a low concentration of minerals and nutrients passes through the root membrane, the 'Endoderm' towards the moisture within the rootsystem and the Stele which has a higher concentration of minerals and nutrients. It should be noted that if the water in the soil has a stronger solution of nutrients than the roots, usually as a result of overfeeding, water will be drawn from the roots back into the soil, depriving the tree of water. This process is known as reverse osmosis.
This may not explain the presence of high imachlorid (sp?) concentrations around trees that were stem injected, but does point out that fluids pass both up and down through root systems.
Plumes descending into the ground 80-90 feet don't originate from a single application of imachlorid by injection into stem/trunk of tree...

Noting the "with all other things being equal" approach to this chemicals singular impact on the environment (specifically the aquifer), the only worrisome item for me would be its potential (even noting its low affinity for bonding) in the vast milieu of chemicals to be found in the "sand pile" that is Long Island, or other similar such locations...


Re: Imidacloprid and Groundwater   edward coyle
  Jul 19, 2007 15:43 PDT 

Ed Frank,

I forwarded your page to a friend of mine, who is very much in the know, as
far as trees, pesticides, and all things arborist related. He attended a
meeting which included the DEC, and a bunch of other agencies, in which the
well contamination was discussed. I suspect it is the same well(s) and the
same source of contamination- and that is a greenhouse within ~50' of it.
When this fact was brought up by him at the meeting it was ignored, and he
was kicked under the table by someone he knew. As if to say," You simply do
not question the abilities and thoroughness of their study."

It is nothing new. Someone screwed up, and someone else is covering up. I'm
quite sure the data you presented is accurate. I am also sure even with the
contamination in the parts per billion it means nothing, except someone
dumped an excessive amount near an entrance to the aquifer. If it were done
continually until the concentration were 10,000 times greater- you might be
able to kill bugs.
However, this is the sort of thing that makes and shapes policy. 'Look what
happened in Long Island!', as the control measures are tightened. It is a
house built upon a poor foundation.

Ed Coyle