Worldwide Regulations of Standard Values of Pesticides for Human Health Risk Control: A Review

The impact of pesticide residues on human health is a worldwide problem, as human exposure to pesticides can occur through ingestion, inhalation, and dermal contact. Regulatory jurisdictions have promulgated the standard values for pesticides in residential soil, air, drinking water, and agricultural commodity for years. Until now, more than 19,400 pesticide soil regulatory guidance values (RGVs) and 5400 pesticide drinking water maximum concentration levels (MCLs) have been regulated by 54 and 102 nations, respectively. Over 90 nations have provided pesticide agricultural commodity maximum residue limits (MRLs) for at least one of the 12 most commonly consumed agricultural foods. A total of 22 pesticides have been regulated with more than 100 soil RGVs, and 25 pesticides have more than 100 drinking water MCLs. This research indicates that those RGVs and MCLs for an individual pesticide could vary over seven (DDT drinking water MCLs), eight (Lindane soil RGVs), or even nine (Dieldrin soil RGVs) orders of magnitude. Human health risk uncertainty bounds and the implied total exposure mass burden model were applied to analyze the most commonly regulated and used pesticides for human health risk control. For the top 27 commonly regulated pesticides in soil, there are at least 300 RGVs (8% of the total) that are above all of the computed upper bounds for human health risk uncertainty. For the top 29 most-commonly regulated pesticides in drinking water, at least 172 drinking water MCLs (5% of the total) exceed the computed upper bounds for human health risk uncertainty; while for the 14 most widely used pesticides, there are at least 310 computed implied dose limits (28.0% of the total) that are above the acceptable daily intake values. The results show that some worldwide standard values were not derived conservatively enough to avoid human health risk by the pesticides, and that some values were not computed comprehensively by considering all major human exposure pathways.


Introduction
Pesticides are broadly applied in numerous agricultural, commercial, residential, and industrial applications to control and kill pests. They help society fight disease and increase agricultural productivity; however, pesticides can be transported into the air, water, soil, and biomass after numerous applications and can cause risks to the ecosystem and to human health. The impact of pesticide residues on human health is a worldwide problem, as human exposure to pesticides can occur through the ingestion of pesticide-contaminated water, food, or residential surface soil, the inhalation of pesticide-contaminated air, soil dust, or industrial vapor, and dermal contact with pesticide-contaminated water (swimming, showering, or raining), air, agricultural commodities, or soil. Worldwide jurisdictions have been working on regulating pesticide standard values for residential surface soil, residential air, drinking water, surface water, groundwater, and food for years.
Pesticide soil regulatory guidance values (RGVs) are applied by worldwide soil jurisdictions to control pesticide pollution in residential surface soil. Pesticide soil RGVs specified the maximum amount of a pesticide which might be present in the soil without prompting regulatory responses, such as surface or groundwater contamination by the transport of pesticides from surface soil, ecological risk, and adverse human health effects by exposure to soil pesticides. The most concerned and conservative pesticide soil RGVs are provided for residential surface soil, where children can be exposed to soil pesticides by the ingestion of soil, the inhalation of soil dust, or dermal contact. Although many worldwide regulatory jurisdictions have provided the RGVs in soil to protect human health, there is a lack of agreement on the pesticides that need to be regulated, as well as the magnitude of the pesticide soil RGVs which should be applied to a certain pesticide. For some of the most frequently regulated pesticides, the RGVs vary to above six orders of magnitude (i.e. 1,000,000) [1]. This variability implies that worldwide soil regulatory jurisdictions have hugely different views on the criteria which cause significant human health risks by residential surface soil pesticides. Other studies have also investigated soil RGVs, but have had their evaluations restricted to less-extensive sets of jurisdictions, such as the U.S. and European nations [2][3][4][5][6][7][8].
Drinking water supplies might be contaminated by pesticides as pesticide can be transferred into surface water or ground water systems, which are usually considered as important drinking water sources. The pesticides found in drinking water may have a potential impact on human health, depending on the amount and the toxicity of the pesticides and the frequency/length of human exposure to the contaminated drinking water. Similar to the pesticide soil RGVs, pesticide drinking water maximum concentration levels (MCLs) are also established by worldwide regulatory jurisdictions to specify the maximum allowable concentration of pesticides in drinking water to protect human health. The results indicate that there is little agreement on both how the pesticides should be regulated and what the magnitude of the MCLs applied to a certain pesticide should be among worldwide drinking water regulatory jurisdictions. The analysis [1] demonstrated that the MCLs of the most commonly regulated pesticides often vary by five, six, or even seven orders of magnitude, which also indicated that some extremely large pesticide MCL values are unlikely to protect human health. The California Department of Public Health (2013) [9] compared the MCLs to the U.S. Environmental Protection Agency (EPA) standard. Among others [10,11], Bamidele (2015) studied the MCLs among the Canadian, European Union (EU), World Health Organization (WHO), U.S., and Nigerian national standards.
Since pesticides are directly applied on crops, fruits, and vegetables in most agricultural applications, infants, children, and adults can be exposed to pesticides by the ingestion of those pesticide-contaminated foods. Pesticide maximum residue limits (MRLs), which specify the maximum concentration of a pesticide that can exist in certain agricultural commodities, were regulated by many nations to promote good agricultural practice (GAP). Because food consumption varies by season, geology, culture, personal habit, economic status and crop availability, which all have impacts on human exposure to pesticides, it is crucial and challenging to develop pesticide MRLs for numerous agricultural commodities to protect human health. To evaluate the effectiveness of the protection from the pesticide MRLs for various agricultural foods, estimations of the most commonly consumed agricultural commodities and the ingestion rate are necessary. Therefore, an implied exposure dose (IED) was introduced in this research to convert the pesticide MRLs in the most consumed agricultural commodities into the pesticide exposure mass burden, and to compare with the toxicology data. The results indicated that many MRLs for the most widely used pesticides were set too high to protect human health.
Pesticides can exist in residential air by the evaporation of volatile and semi-volatile pesticides, such as organochlorine pesticides, from crops and residential surface soil. In addition, pesticides can be blown away from agricultural fields by the wind, and some fumigants (e.g., bromomethane) are released into the air in a gaseous form. Therefore, the regulation of pesticide standard values in the residential air is necessary to control human health risks through inhalation and dermal contact exposures, especially for volatile and semi-volatile pesticides. However, few worldwide jurisdictions have regulated pesticide air standard values, which means that people around the world are probably not protected by the pesticide air regulations, especially for some farmers and workers who frequently work in the agricultural field.
Human exposures to pesticide may also occurs through swimming in rivers, lakes, or pools where the water has been contaminated by pesticides, taking a shower when the water is being pumped from pesticide-contaminated ground water, getting wet from pesticide-contaminated rainwater, or handling pesticide-related products during work. Since the regulation of pesticides in these scenarios is important to control human health risks, and since these scenarios are the four most frequent human exposure pathways for pesticides, the pesticide standard values in major human exposure pathways which include residential soil, drinking water, agricultural commodities, and residential air were discussed in this research.
The standard values for pesticides for each human exposure pathway should be derived by laboratory toxicology data and human health risk models. Some uncertainty and marginal safety factors should also be applied to allow for additional exposure possibilities. As pesticide exposures in major exposure scenarios always happen simultaneously, it is necessary to derive and allocate pesticide standard values in major exposure pathways comprehensively. Therefore, the implied maximum dose limit (IMDL) was applied to compute the total maximum exposure mass burden for a certain pesticide from the pesticide standard values of the national jurisdictions in all major exposure pathways. The objectives of this research review are to evaluate current worldwide pesticide standard values in major exposures, examine whether those standard values can protect human health, and help environmental regulatory jurisdictions to rationalize their pesticide standard values by a scope of worldwide efforts.

Materials
The materials for this research review include two main parts; one is the pesticide which had been regulated with a certain residential soil RGV, drinking water MCL, agricultural commodity MRL, or residential air MCL, and the other is the worldwide regulatory jurisdiction which had promulgated a certain pesticide standard value in any of those major exposure pathways.

Pesticide
The World Health Organization (WHO) (2017) [12] defined a pesticide as a chemical compound that is used to kill pests, including insects, rodents, fungi and unwanted plants (weeds). The Food and Agriculture Organization (FAO) of the United Nations defined a pesticide as any substance or mixture of substances intended for preventing, destroying, or controlling any pest, including vectors of human or animal diseases, unwanted species of plants or animals causing harm during, or otherwise interfering with, the production, processing, storage, or marketing of food, agricultural commodities, wood and wood products, or animal feedstuffs, or which may be administered to animals for the control of insects, arachnids or other pests in or on their bodies.
Pesticides can be classified by target groups as acaricides, avicides, bactericides, herbicides, fungicides, insecticides, repellents, virucides, and so on. According to the chemical compositions of the active ingredients, pesticides can be categorized into four main groups: carbamates, organochlorines, organophosphorus, and pyrethrin and pyrethroids. WHO (2009) [13] classified pesticides by hazard as an extremely hazardous pesticide, a highly hazardous pesticide, a moderately hazardous pesticide, a slightly hazardous pesticide, and a pesticide which is unlikely to present an acute hazard. In addition, based on the mode of formulation, pesticides can be classified as emulsifiable concentrates, wettable powders, granules, baits, dust, and fumigants [14].

CAS No.
Because of the complex chemical structures and the chemical complexity of pesticides and their active ingredients, pesticides are often regulated by their trade names instead of the chemical nomenclature conventions. For example, the National Institute of Standards and Technology (NIST) (2011) [15] listed 120 names for the pesticide "Lindane", which include 30 chemical nomenclature names and 90 other trade names. Even worldwide jurisdictions have regulated pesticides by their local trade names in foreign languages, which has made it difficult to identify pesticides by their "names". The Chemical Abstracts Service Registry Number (CAS No.) by NIST and the common name by the International Union of Pure and Applied Chemistry (IUPCA) were applied to identify pesticides in this research review, as each pesticide was assigned a unique CAS No. and a common name. It would be helpful and convenient to the public if a jurisdiction were to use the CAS No. to identify regulated pesticides; however, the CAS No. is not available for most worldwide jurisdictions beyond Europe and North America.

Sources for Worldwide Pesticide Soil RGVs and Drinking Water MCLs
Pesticide soil RGVs and drinking water MCLs were directly taken from the regulatory jurisdictions and most were obtained from the official government websites, which are the primary sources. Sometimes the primary source was not available when the official documents were too old to have an online version when access was needed for the document, when the official website was under maintenance, and when the jurisdiction needed to be purchased. For these cases, secondary sources such as newspapers, annual reports, research articles, conferences, and government statements were applied to obtain the pesticide standard values. Since many international regulatory jurisdictions were written in foreign languages, the Google translate online tool was used to translate the foreign language documents into English. The reference websites for the worldwide and U.S. pesticide soil RGVs and drinking water MCLs and the pesticide standard values used in this study were provided in supplement materials (Tables S1-S4 and Database). When web addresses and online documents become out of date and inactive, keywords from the document title would help to address the new web location by using web search engines. Table 1 lists the worldwide nations, regions, territories, and organizations which had provided the pesticide soil or drinking water regulatory jurisdictions, their sources, the numbers of pesticide standard values, and the languages applied. A total of 4590 pesticide soil RGVs were identified by 108 international jurisdictions from 54 United Nation (UN) members, three multi-national organizations, and two non-UN members outside of the U.S. At least 3534 pesticide drinking water MCLs were identified from 130 international jurisdictions from 102 UN members, four multi-national organizations, and two non-UN members outside of the U.S. There were more nations which provided pesticide standard values in drinking water than for residential soil, indicating that more nations focused on pesticide regulations for drinking water. Pesticides RGVs and MCLs were also identified by the U.S. jurisdictions from national organizations, states, cities, U.S. territories, and Autonomous Native American Tribes (see Table 2). There were 14,831 pesticides RGVs identified from 66 U.S. soil regulatory jurisdictions, including 46 of the 50 states, six national organizations, five regions (cities and counties), two U.S. territories, and seven Autonomous Native American Tribes. In addition, a total of 1940 pesticide MCLs were identified by 61 U.S. drinking water jurisdictions, from 48 of the 50 states, three national organizations, and two U.S. territories. Only four states, including North Dakota, South Carolina, South Dakota and Utah, did not provide pesticide soil RGVs, and two states-Georgia and Washington-did not provide pesticide drinking water MCLs.

Sources for Worldwide Pesticide Agricultural Commodity MRLs
Pesticide agricultural commodity MRLs of worldwide nations were collected by the global MRL database (2014) [16], which included at least 90 nations and multi-national organizations (see Table 3). Since hundreds of agricultural commodities and pesticides were regulated and collected in this database, only the currently most widely used pesticides and the most consumed agricultural commodities were used in this research to compute the IED. The most widely used pesticides and consumed foods were selected based on worldwide pesticide usage and the consumption of foods, which was investigated for the nations where agricultural plays a significant role, such as China [17,18], India [19], the Philippines [20], Germany [21], the United Kingdom [22], Canada [23,24], the U.S. [25][26][27], Mexico [28], Costa Rica [29], Brazil [30,31], New Zealand [32], Australia [33,34], and South Africa [35]. The 14 current most widely used pesticides selected in this research review were 2,4-D, Aldicarb, Atrazine, Chlorothalonil, Chlorpyriphos, Diazinon, Dicamba, Diuron, Glyphosate, Malathion, Mancozeb, MCPA, Metolachlor, and Trifluralin. The most consumed agricultural commodities selected in this study were classified into four groups: grain crops (corn, wheat, and rice), vegetable crops (tomato, onion, and potato), fruit crops (apple, bananas, grape, and orange), and drinks (coffee bean, and tea leaves). Although pesticides can be transported and accumulate into meat and dairy products such as beef and fish, agricultural commodities are often exposed to pesticides directly, due to application methods such as spraying. The amount of pesticide accumulated in livestock always depends on the living environment, feeding stuff, and the metabolism of the animals. Compared to the pesticide exposure from the meat consumption, the pesticide exposure risk from agricultural commodities is much higher due to its broader application. Thus, pesticide exposure from the consumption of meat and dairy products will not be discussed in this study; however, marginal safety factors should be accounted for to allow additional exposures.

Sources for Pesticide Residential Air MCLs
Few jurisdictions in the world had regulated pesticide residential air MCLs; only the U.S. [36] regulated and derived the pesticide air MCLs systematically. Cancer and non-cancer human health risk models were developed by the U.S Environmental Protection Agency (USEPA), and the MCLs for 43 volatile and semi-volatile pesticides were promulgated. It is necessary to regulate pesticide air MCLs as the human health risks are raised by inhalation, skin contact, and even eye contact for pesticide-contaminated air, especially for farmers and workers who work in forests and the agricultural field. The regulation of pesticides in the air can also protect the ecosystem, wildlife, and livestock. Because of the lack of information available for worldwide pesticide air MCLs, regulatory pesticide standard values for residential air are omitted in this study.

Statistical Analysis of Pesticide Standard Values
Since most data sets for pesticide standard values resemble a log-transformed random distribution [37][38][39][40][41], the cumulative distribution function was applied to plot the pesticide standard values and to compare it with the cumulative distribution of a log-normal random variable with identical statistics, arithmetic, mean, and standard deviations. The cumulative distribution function was used to illustrate how worldwide pesticide standard values dispersed over the span of values and the ranking of the standard values for each jurisdiction. The empirical cumulative distributions generated from pesticide standard value sets were generated as follows, where X r is a random value for a pesticide RGV, MCL, IED, IMDL, or the number of pesticide standard values which a jurisdiction had provided, X i is the known value for the same pesticide, and ni is the integer rank of X i in the N known values.
To examine if the degree of the empirical distribution resembled the log-normal random distribution of a Pearson (r) correlation, the analysis was conducted as follows, where E(X i ) is the probability computed from the empirical distribution, and F(X i ) is the probability calculated for the log-normal cumulative distribution.
Since some jurisdictions shared the same standard value for a certain pesticide, non-random values always appeared in empirical distributions. A data value cluster was introduced in this study and defined as a data interval (X i − X i+Y ), containing Y values which did not occur randomly. The probability (P c ) of a randomly occurring data value cluster was quantified by the binomial probability function as follows, where F(X i ) and F(X i+Y ) were computed from the log-normal cumulative distribution. A probability of less than 0.001 indicated that the data value cluster did not occur randomly.

Human Health Risk Models
Many regulatory jurisdictions developed standard values for non-genotoxic pesticides, "thresholded toxicants" [42], or "systemic toxicants" [43] based on the acceptable daily intake (ADI), tolerable daily intake (TDI), or reference dose (RfD) [43]. The standard values are defined as the level of toxicant exposure on a daily or weekly basis without adverse health effects or an appreciable health risk over a lifetime. The ADI or TDI was usually converted from the no observed adverse effect level (NOAEL), or the lowest observed effect level (LOEL) by laboratory animal experiments by applying an uncertainty factor which allows for interspecies differences and human variability, expressed as follows, ADI (TDI, or RfD) = NOAEL (or LOEL) UF (4) where UF is the uncertainty factor and usually applies to a 100-fold increase [42]. Based on the toxicology data, the pesticide soil RGVs should be developed based on exposure scenarios. To evaluate if the pesticide soil RGVs are conservative enough to protect human health, human health cancer and non-cancer risk models (Equations (5)- (12)) developed by the USEPA [36] were applied in this research to compute cancer and non-cancer risk uncertainty bounds. The USEPA models considered all pesticide soil exposure scenarios, such as soil ingestion, soil dust inhalation, and soil dermal contact. There is an uncertainty when using the USEPA models to examine soil RGVs. This does not mean that the USEPA models are universal; every jurisdiction can develop and conduct an RGV risk assessment and evaluation based on their situations. The USEPA summarized the toxicity and chemical-specific information of pesticides, which included the chronic oral slope factor (CSF o ) (kg-mg/day), the fraction of contaminant absorbed in gastrointestinal tract (GIABS) (unitless), the fraction of contaminant absorbed dermally from soil (ABS d ) (unitless), the chronic inhalation unit risk (IUR) (m 3 /ug), the volatilization factor (VF s ) (m 3 /kg), the chronic oral reference dose (RfD o ) (mg/kg-day), and the chronic inhalation reference concentration (RfC) (mg/m 3 ) [36]. The range of the exposure coefficients [44] applied by the U.S. states were defined in the following equations.
The pesticide soil RGV (mg/kg) equation derived by cancer risk equations was expressed as follows, where the RSL (mg/kg) is the reginal screen level derived by the cancer risk equations of soil ingestion (Equation (6)), soil dermal contact (Equation (7)), and soil dust inhalation (Equation (8)).
DFS adj -Resident soil dermal contact factor (360.8 mg-year/kg-day), (253, 1257) PEF w -Particulate emission factor (1.4 × 10 9 m 3 /kg), (7.8 × 10 7 -6.6 × 10 9 ) ED-Exposure duration (30 years) ET-Exposure time (24 h/day), (2, 24) The pesticide soil RGV (mg/kg) non-cancer risk equations were expressed as follows, SA c -Soil surface area for child (2800 cm 2 ), (1750, 2960) AF c -Soil adhesion factor for child (0.2 mg/cm 2 ), (0.2, 1.0) Pesticide drinking water MCLs and human health uncertainty risk bounds were based on the exposure scenario of ingestion. The MCL uncertainty risk bounds depend on the variation of the parameters in Equation (13), where HW is human body weight (kg), which is referred to as an average adult human weight in many jurisdictions. Some jurisdictions, such as the Australian Government [45], applied 70 kg and others, such as WHO [46], used 60 kg. PF is a proportion factor which quantifies the portion of the total pesticide exposure that is allocated to the drinking water ingestion pathway, which usually ranges from 0.1 to 1.0 [46]. V is the daily drinking water intake rate (L/day), and for most worldwide jurisdictions, 2.0 L/day was used, while for some nations with a cold climate, 1.5 L/day was applied [47].

Pesticide Agricultural Commodity Implied Exposure Dose (IED)
Pesticide IED was introduced to compute the implied pesticide mass burden from the most consumed agricultural commodities, and was compared to the ADI value of the same pesticide. The agricultural food intake rates were estimated in Equation (14) by the U.S. Department of Agriculture (USDA) [48] annual food consumption database. It should be noticed that agricultural commodity consumptions vary by season, geology, culture, personal habit, economic status and crop availability. The application of the USDA is one of the methods to estimate consumption rates. Table 3 lists the estimated agricultural food intake rate.
where FIR j is the estimated intake rate of the agricultural commodity j, M j is the total mass of agricultural commodity j consumed annually in the U.S. (kg/year), and P is the U.S. population (318.9 million for 2014).
The IED for the most widely used pesticides was expressed as follows, where the IED i is the implied exposure dose computed for the pesticides i (mg/kg-day), HW is the average adult human weight (kg), and EF is the exposure factor (unit less). The calculated results in this study were based on an average adult weight of 70 kg and an EF of 1.0 [49].

Implied Maximum Dose Limit
Since human exposure to pesticides always occurs in different exposure pathways simultaneously, it is necessary to develop pesticide standard values comprehensively by considering all possible human exposure scenarios. The IMDL was applied to examine the pesticide standard values in major exposure pathways by computing the implied maximum pesticide mass burden from all major exposures; this was then compared with the ADI value of the same pesticide. The IMDL could be an indicator to assess whether the pesticide standard values were developed comprehensively and conservatively enough to protect human health in all major exposure scenarios. The IMDL was developed based on human exposure models in the following equations. Since there is little information about the worldwide pesticide air standards, the pesticide exposure from the residential air was omitted.
For drinking water: For residential soil: For agricultural commodities: where IDL is the implied dose limit (mg/kg-day) computed from drinking water, soil, and agricultural commodities. For the IDL soil , since the soil dust inhalation pathway contribution is extremely low compared to soil ingestion and soil dermal contact exposure, the soil inhalation exposure was omitted for the IDL soil calculation. If more than one PSV for a certain pesticide was regulated by a nation in one major exposure, different IMDLs were calculated by combining different IDLs with others.  Table 1 for details) and the information regarding MCL numbers from these jurisdictions was not shown in Figure 1. The U.S. Army Public Health Command had provided the maximum number of drinking water MCLs, which is 520. Jurisdictions from Morocco, South Africa, Tanzania, Thailand, and Tunisia only regulated one pesticide MCL. There are two drinking water MCL numbers data clusters. The cluster at 24 is made up of 35 values from the province of Ontario (Canada) and 34 U.S. related jurisdictions. The cluster at 36 is made up of 17 values from the WHO, Albania, Antigua and Barbuda, Bahamas, Belize, Bhutan, State of San Paolo (Brazil), Fiji, Japan, Kiribati, Kuwait, Nauru, Sudan, Tonga, Tuvalu, Vanuatu, and Vietnam. The arithmetic mean of the MCL numbers is 36, and this value is larger than the median and geometric mean, which are 24 and 23, respectively, because the arithmetic mean is skewed by some large values such as 520 (U.S. Army Public Health Command). In general, worldwide jurisdictions had regulated more pesticide soil RGVs than drinking water MCLs, because usually pesticides are directly applied to agricultural land, forest, and home gardens, which make the pesticides accumulate in the surface soil first. However, there are more nations regulating pesticides standard values in drinking water (102 nations) than in soil (54 nations), which indicates that more nations focus on the regulation of pesticides in drinking water.  23, respectively, because the arithmetic mean is skewed by some large values such as 520 (U.S. Army Public Health Command). In general, worldwide jurisdictions had regulated more pesticide soil RGVs than drinking water MCLs, because usually pesticides are directly applied to agricultural land, forest, and home gardens, which make the pesticides accumulate in the surface soil first. However, there are more nations regulating pesticides standard values in drinking water (102 nations) than in soil (54 nations), which indicates that more nations focus on the regulation of pesticides in drinking water. Since hundreds of different pesticides have been registered for use and nations may apply different pesticides based on their situations, 25 pesticides were generally selected to examine if the worldwide nations had provided enough pesticide regulation information based on historical and current usage. These 25 selected pesticides include 14 current widely used pesticides: 2,4-D, Aldicarb, Atrazine, Chlorothalonil, Chlorpyriphos, Diazinon, Dicamba, Diuron, Glyphosate, Malathion, Mancozeb, MCPA , Metolachlor, and Trifluralin, and 11 historically largely used pesticides (the Stockholm Convention POP): Aldrin, Chlordane, DDT, Dieldrin, Endrin, Heptachlor, Toxaphene, Lindane, Endosulfan, Pentachlorophenol, and Bromomethane. Figure 2 illustrates the geographic distribution of nations on regulating the 25 selected pesticides in residential surface soil. A total of 49 nations have regulated the soil RGV for at least one of these 25 pesticides. Only national jurisdictions were applied, and, if a nation had more than one national jurisdiction, the better performing one was selected. For example, the U.S. EPA regulated the RGVs for all of these 25 pesticides, and the U.S. ASTDR provided the RGVs for only eight of the selected pesticides. Therefore, the U.S. EPA was selected here as the U.S. national representative jurisdiction. In Figure 2, a nation with a darker red color means that this nation had regulated more selected pesticides for soil. The Czech Republic, New Zealand, Slovakia, and the U.S. regulated soil RGVs for all the selected pesticides. Malaysia provided the RGV for 24 selected pesticides and the Bahamas regulated for 20 pesticides. The arithmetic mean for the number of selected pesticides which had been regulated with the soil RGV is 9. Some nations in Africa, Asia, and South America did not provide any soil pesticide standard values for the selected pesticides. Some multinational organizations, including both the EU and WHO, regulated the soil RGVs for eight of the selected pesticides. Figure 3 illustrates the geographic distribution of nations regulating the 25 selected pesticides in drinking water. There are 97 nations which provided the pesticide drinking water MCL for at least one of the 25 selected pesticides. Most of the European nations had regulated the MCLs for all selected pesticides, because these nations applied EU  Figure 2 illustrates the geographic distribution of nations on regulating the 25 selected pesticides in residential surface soil. A total of 49 nations have regulated the soil RGV for at least one of these 25 pesticides. Only national jurisdictions were applied, and, if a nation had more than one national jurisdiction, the better performing one was selected. For example, the U.S. EPA regulated the RGVs for all of these 25 pesticides, and the U.S. ASTDR provided the RGVs for only eight of the selected pesticides. Therefore, the U.S. EPA was selected here as the U.S. national representative jurisdiction. In Figure 2, a nation with a darker red color means that this nation had regulated more selected pesticides for soil. The Czech Republic, New Zealand, Slovakia, and the U.S. regulated soil RGVs for all the selected pesticides. Malaysia provided the RGV for 24 selected pesticides and the Bahamas regulated for 20 pesticides. The arithmetic mean for the number of selected pesticides which had been regulated with the soil RGV is 9. Some nations in Africa, Asia, and South America did not provide any soil pesticide standard values for the selected pesticides. Some multinational organizations, including both the EU and WHO, regulated the soil RGVs for eight of the selected pesticides. Figure 3 illustrates the geographic distribution of nations regulating the 25 selected pesticides in drinking water. There are 97 nations which provided the pesticide drinking water MCL for at least one of the 25 selected pesticides. Most of the European nations had regulated the MCLs for all selected pesticides, because these nations applied EU standards, which provided individual and total standards for any pesticide. Australia provided the MCLs for 22 selected pesticides and Iraq regulated for 21. Both the U.S. and China regulated the MCLs for 9 of the 25 selected pesticides, while Bangladesh, South Korea, and Morocco had only provided the MCLs for one of the selected pesticides. Some nations in Africa and Asia did not provide any drinking water pesticide standard values for the selected pesticides. The arithmetic mean for the number of selected pesticides which had been regulated with the drinking water MCL is 16. In terms of the multinational organizations, the EU provided all MCLs for these 25 pesticides and the WHO regulated for 13.

The Most Commonly Regulated Pesticides in Soil and Drinking Water
The most commonly regulated pesticides in this study were defined as pesticides with over 100 soil RGVs or drinking water MCLs. Table 4 summarizes the commonly regulated pesticides by the common name under which they were most often regulated in jurisdictions, CAS No, occurrence

The Most Commonly Regulated Pesticides in Soil and Drinking Water
The most commonly regulated pesticides in this study were defined as pesticides with over 100 soil RGVs or drinking water MCLs. Table 4 summarizes the commonly regulated pesticides by the common name under which they were most often regulated in jurisdictions, CAS No, occurrence

The Most Commonly Regulated Pesticides in Soil and Drinking Water
The most commonly regulated pesticides in this study were defined as pesticides with over 100 soil RGVs or drinking water MCLs. Table 4 summarizes the commonly regulated pesticides by the common name under which they were most often regulated in jurisdictions, CAS No, occurrence frequency (in U.S. jurisdictions and worldwide jurisdictions), the lowest and highest values, and the log orders of variation (LOV, LOV = log {highest value/lowest value}) over which the RGVs and MCLs were scattered. There are 39 most commonly regulated pesticides with either soil RGVs or drinking water MCLs regulated above 100. DDT is the most frequently regulated pesticide in soil with 319 RGVs, made up of 140 RGVs from the U.S. related jurisdictions and 179 RGVs from the jurisdictions outside of the U.S., while 2,4-D is the most frequently regulated pesticides in drinking water with 180 MCLs, including 59 U.S. MCLs and 121 worldwide MCLs. Pesticides including Endosulfan, α-HCH, β-HCH, Bromomethane, and o-Cresol were regulated by over 100 RGVs in soil but less than 13 MCLs in drinking water. There were 125 drinking water MCLs regulated for DBCP, while 31 DBCP soil RGVs were promulgated with only one RGV from the jurisdictions outside of the U.S. There were 22 and 25 pesticides which had been regulated with over 100 soil RGVs and drinking water MCLs, respectively. For these 39 most commonly regulated pesticides, the state of Idaho had specified the lowest soil RGVs for at least 7 pesticides, and both Oregon and Serbia provided the lowest RGVs for at least four of the most commonly regulated pesticides. Texas specified the highest soil RGVs for at least 14 of the commonly regulated pesticides, Guam (U.S.) provided the highest RGVs for five pesticides, and the U.S. Military jurisdiction regulated for four pesticides. Regardinf drinking water, the EU and the jurisdictions which applied the EU standards specified the lowest drinking water MCLs for at least 21 of the most commonly regulated pesticides, and Wyoming provided the lowest MCLs for at least 9 pesticides. The EU provided the pesticide drinking water MCLs quite conservatively, because the jurisdiction specified 0.0001 mg/L for an individual pesticide. On the other hand, the U.S. Military jurisdiction promulgated the highest MCLs for at least 12 pesticides; this is probably because the jurisdiction derived the MCL based on adult body weight and short time exposure conditions. Vietnam regulated the highest MCLs for 9 of the commonly regulated pesticides, and Mexico provided the highest MCLs for 5.
The LOV values for the 50 most commonly regulated pesticides in soil range from 2.41 (Terbufos) to 9.89 (Dieldrin). The arithmetic mean of these LOV values is 6.08. The average LOV value for the 22 most commonly regulated pesticides in soil is 6.97, and the average for the rest of the pesticides is 5.38. The LOV values for the 50 most commonly regulated pesticides in drinking water range from 1.70 (Epichlorohydrin) to 7.11 (DDT). The arithmetic mean of the drinking water pesticides LOV values is 4.63. For the 25 most commonly regulated pesticides in drinking water, the average LOV value is 4.93, and the average LOV value for the rest is 4.33. Figure 4 illustrates the individual and running average LOV values for the 50 most commonly regulated pesticides in soil and drinking water. In general, the individual and running average LOV values are smaller when the number of RGVs and MCLs decreases, and the LOV values of pesticides drinking water MCLs are less than soil RGVs.

Analysis of Soil RGVs for the Commonly Regulated Pesticides
2,4-D and Glyphosate are two most widely used pesticides today and are two commonly regulated pesticides in residential surface soil. Figure 5

Analysis of Soil RGVs for the Commonly Regulated Pesticides
2,4-D and Glyphosate are two most widely used pesticides today and are two commonly regulated pesticides in residential surface soil. Figure 5    The cancer risk uncertainty bounds for 2,4-D and Glyphosate were not derived, because the U.S. EPA did not consider the cancer risk for 2,4-D and Glyphosate RGV calculations. The non-cancer risk uncertainty bounds derived for 2,4-D soil RGVs were (45.0, 3690.0) mg/kg. Six values (4.1% of the total RGVs) are above the non-cancer uncertainty upper bound, 111 values (75.5%) fall within the uncertainty bounds, and 30 values (20.4%) are below the lower non-cancer risk uncertainty bound. For Glyphosate, the computed non-cancer risk uncertainty bounds were (3000.0, 32,000.0) mg/kg. Only one RGV is above the uncertainty upper bound, 44 values (47.3% of the total RGVs) fall within the bounds, and 48 values (51.6%) are below the lower bound.
The arithmetic mean of the 2,4-D RGVs is 867.0 mg/kg, which is exceeded by only 30 RGVs. This value is heavily affected by some large RGVs, such as 12,000.0 mg/kg (Oregon) at the high end of the data distribution. The geometric mean and median are 122.0 mg/kg and 630.0 mg/kg, respectively, which are better measures of the central tendency of the distribution. For Glyphosate, the arithmetic mean of the RGVs is 4715.7 mg/kg, with 45 RGVs larger than this value; this is because it is heavily affected by large RGVs, such as 36,000.0 mg/kg (Texas). The geometric mean and median are 545.8 mg/kg and 1500.0 mg/kg, respectively, which are better measures of the central tendency of the distribution. Table 5 provides the statistical summary information of the RGVs for the 27 commonly regulated pesticides. The weighted average of the Pearson coefficients by the RGV number is 0.958, and nearly all the pesticides RGVs have Pearson coefficients above 0.900, except Picloram (0.878), which indicates that the soil RGVs for these commonly regulated pesticides are well dispersed. Some pesticides, such as Chlordane, Lindane, and Atrazine, have coefficients very close to 1.00. For these pesticides, there are at least 1632 RGVs (42% of the total) in the non-random data clusters, and the data clusters are usually made up of the jurisdictions from U.S., EU, Spain, Canada, and Australia, which means that the national jurisdictions in these nations had provided leadership in soil pesticide regulations. There are at least 300 soil RGVs (8% of the total) above all of the computed human health uncertainty upper bounds, which indicates that, for these 27 commonly regulated pesticides in soil, at least 300 RGVs cannot protect human health.   Table 6 provides the statistical summary information of the MCLs for the 29 commonly regulated pesticides. The weighted average of the Pearson coefficients by the MCL number is 0.904, and nearly all the pesticides MCLs have Pearson coefficients above 0.850, except Aldrin (0.817), Chlorpyrifos (0.811), and Dieldrin (0.803), which indicates that the drinking water MCLs for these 29 commonly regulated pesticides are well dispersed. Some pesticides, such as Heptachlor and Pentachlorophenol, have coefficients above 0.950. For these 29 pesticides, there are at least 2980 MCLs (80.7% of the total) in the non-random data clusters, and the data clusters are usually made up of the jurisdictions from U.S., EU, and WHO, which means that these national and multination jurisdictions had provided leadership in drinking water pesticide regulations. There are at least 172 drinking water MCLs (5% of the total) above the computed human health uncertainty upper bounds, which indicates that, for these 29 commonly regulated pesticides in drinking water, at least 172 MCLs cannot protect human health.  Table 6 provides the statistical summary information of the MCLs for the 29 commonly regulated pesticides. The weighted average of the Pearson coefficients by the MCL number is 0.904, and nearly all the pesticides MCLs have Pearson coefficients above 0.850, except Aldrin (0.817), Chlorpyrifos (0.811), and Dieldrin (0.803), which indicates that the drinking water MCLs for these 29 commonly regulated pesticides are well dispersed. Some pesticides, such as Heptachlor and Pentachlorophenol, have coefficients above 0.950. For these 29 pesticides, there are at least 2980 MCLs (80.7% of the total) in the non-random data clusters, and the data clusters are usually made up of the jurisdictions from U.S., EU, and WHO, which means that these national and multination jurisdictions had provided leadership in drinking water pesticide regulations. There are at least 172 drinking water MCLs (5% of the total) above the computed human health uncertainty upper bounds, which indicates that, for these 29 commonly regulated pesticides in drinking water, at least 172 MCLs cannot protect human health.   Figure 7 illustrates the 91 2,4-D and 91 Glyphosate agricultural commodities' IEDs which were plotted in the empirical cumulative distribution forms and compared to the log-normal random variable cumulative distributions with identical statistics. The 2,4-D IEDs span 1.33 orders of magnitude (0.00054, 0.011) mg/kg-day with the Pearson coefficient of 0.761. For Glyphosate, the IEDs span 3.67 orders of magnitude (0.0000386, 0.18) mg/kg-day with the Pearson coefficient of 0.907. Those correlations were heavily influenced by several large IED data clusters. The 2,4-D food IED set has two large non-random data clusters at 0.0072 mg/kg-day (32 values) and 0.0069 mg/kg-day (31 values). For Glyphosate, there are also two large non-random data clusters at 0.17 mg/kg-day (31 values) and 0.033 mg/kg-day (33 values Figure 7 illustrates the 91 2,4-D and 91 Glyphosate agricultural commodities' IEDs which were plotted in the empirical cumulative distribution forms and compared to the log-normal random variable cumulative distributions with identical statistics. The 2,4-D IEDs span 1.33 orders of magnitude (0.00054, 0.011) mg/kg-day with the Pearson coefficient of 0.761. For Glyphosate, the IEDs span 3.67 orders of magnitude (0.0000386, 0.18) mg/kg-day with the Pearson coefficient of 0.907. Those correlations were heavily influenced by several large IED data clusters. The 2,4-D food IED set has two large non-random data clusters at 0.0072 mg/kg-day (32 values) and 0.0069 mg/kg-day (31 values). For Glyphosate, there are also two large non-random data clusters at 0.17 mg/kg-day (31 values) and 0.033 mg/kg-day (33 values). The ADI value of 2,4-D is 0.01 mg/kg-day. Only one IED computed from Russia is above the ADI value. Although most IEDs computed from the most commonly consumed agricultural commodities are within the safety levels (i.e. lower than the 2,4-D ADI value), this does not account for other foods (meat or dairy product) and other exposure pathways. For example, the 2,4-D IED computed from the U.S. and Mexico is 0.009 mg/kg-day, which is nearly 90% of the 2,4-D ADI value, which indicates that there is little safety margin left for other foods or exposure pathways. For Glyphosate, 41 IEDs computed from the most commonly consumed agricultural commodities exceed the ADI value of 0.1 mg/kg-day, and this does not account for other foods or exposure pathways either. Most nations only regulated the MRLs for four of the 12 most commonly consumed agricultural commodities (see Table  3). Table 7 provides the statistical summary information of the IEDs for the 14 widely used pesticides. The weighted average of the Pearson coefficients by the IED number is 0.825. Some pesticides, such as Diazinon, Malathion, and Mancozeb, have Pearson coefficients around 0.950, while some other pesticides, such as Dicamba and MCPA, have coefficients less than 0.700, because the correlations are heavily affected by large IED data clusters. For these 14 pesticides, there are at least 804 IEDs (72.7% of the total) in the non-random data clusters, and the data clusters are usually led by EU and WHO jurisdictions. There are at least 310 computed IEDs (28.0% of the total) above the ADI values, which indicates that, for these 14 widely used pesticides, at least 310 IEDs computed from the agricultural commodities MRLs cannot protect human health. The ADI value of 2,4-D is 0.01 mg/kg-day. Only one IED computed from Russia is above the ADI value. Although most IEDs computed from the most commonly consumed agricultural commodities are within the safety levels (i.e. lower than the 2,4-D ADI value), this does not account for other foods (meat or dairy product) and other exposure pathways. For example, the 2,4-D IED computed from the U.S. and Mexico is 0.009 mg/kg-day, which is nearly 90% of the 2,4-D ADI value, which indicates that there is little safety margin left for other foods or exposure pathways. For Glyphosate, 41 IEDs computed from the most commonly consumed agricultural commodities exceed the ADI value of 0.1 mg/kg-day, and this does not account for other foods or exposure pathways either. Most nations only regulated the MRLs for four of the 12 most commonly consumed agricultural commodities (see Table 3). Table 7 provides the statistical summary information of the IEDs for the 14 widely used pesticides. The weighted average of the Pearson coefficients by the IED number is 0.825. Some pesticides, such as Diazinon, Malathion, and Mancozeb, have Pearson coefficients around 0.950, while some other pesticides, such as Dicamba and MCPA, have coefficients less than 0.700, because the correlations are heavily affected by large IED data clusters. For these 14 pesticides, there are at least 804 IEDs (72.7% of the total) in the non-random data clusters, and the data clusters are usually led by EU and WHO jurisdictions. There are at least 310 computed IEDs (28.0% of the total) above the ADI values, which indicates that, for these 14 widely used pesticides, at least 310 IEDs computed from the agricultural commodities MRLs cannot protect human health. Figure 8 illustrates the 145 2,4-D IMDLs computed from three major exposure pathways which were plotted in the empirical cumulative distribution form and compared to the log-normal random variable cumulative distribution with identical statistics. A total of 17 IMDLs were computed from three exposure pathways, 53 values were calculated from two of the major exposure pathways, and 75 values were computed from one of the major exposure pathways. The 2,4-D IMDLs span 6.70 orders of magnitude (1.73 × 10 −7 , 8.66 × 10 −1 ) mg/kg-day, and are well dispersed over this span with the Pearson coefficient of 0.881. The correlation was heavily influenced by three IMDL data clusters at 0.00718-0.00781 mg/kg-day ( Figure 8 illustrates the 145 2,4-D IMDLs computed from three major exposure pathways which were plotted in the empirical cumulative distribution form and compared to the log-normal random variable cumulative distribution with identical statistics. A total of 17 IMDLs were computed from three exposure pathways, 53 values were calculated from two of the major exposure pathways, and 75 values were computed from one of the major exposure pathways. The 2,4-D IMDLs span 6.70 orders of magnitude (1.73 × 10 -7 , 8.66 × 10 -1 ) mg/kg-day, and are well dispersed over this span with the Pearson coefficient of 0.881. The correlation was heavily influenced by three IMDL data clusters at 0.00718-0.00781 mg/kg-day (33 values), 0.00695 mg/kg-day (17 values), and 0.000857 mg/kg-day (21 values). The maximum IMDL value is 0.866 mg/kg-day, computed from the Vietnam national drinking water MCL, agricultural commodity MRLs, and soil RGV (2008). The minimum IMDL value is 1.73 × 10 -7 mg/kg-day, computed from the Armenia, Georgia, and Moldova soil jurisdictions. A total of 13 IMDLs (9.0% of the total number) are above the 2,4-D ADI value, 10 of the IMDLs were computed from three major exposure pathways, and three of them were computed from two of the major exposure pathways. Although 91% of the 2,4-D IMDLs computed from the national jurisdictions are below the ADI value, 95.5% of the IMDLs were computed from either two of the major exposure pathways or one exposure pathway, which indicates that many nations did not provide enough information to protect human health against 2,4-D exposure. Figure 9 illustrates the 115 Glyphosate IMDLs computed from three major exposure pathways which were plotted in the empirical cumulative distribution form and compared to the log-normal random variable cumulative distribution with identical statistics. Four IMDLs were computed from three exposure pathways, 36 values were calculated from two of the major exposure pathways, and 75 values were computed from one of the major exposure pathways. The Glyphosate IMDLs span 6.11 orders of magnitude (1.43 × 10 −7 , 1.86× 10 −1 ), and are well dispersed over this span with the Pearson coefficient of 0.854. The correlation was heavily influenced by three IMDL data clusters at 0.166 mg/kg-day (35 values), 0.0335 mg/kg-day (30 values), and 2.86 × 10 −6 mg/kg-day (8 values). The maximum IMDL value is 0.866 mg/kg-day, computed from the Vietnam national drinking water MCL, agricultural commodity MRLs, and soil RGV (2008). The minimum IMDL value is 1.73 × 10 −7 mg/kg-day, computed from the Armenia, Georgia, and Moldova soil jurisdictions. A total of 13 IMDLs (9.0% of the total number) are above the 2,4-D ADI value, 10 of the IMDLs were computed from three major exposure pathways, and three of them were computed from two of the major exposure pathways. Although 91% of the 2,4-D IMDLs computed from the national jurisdictions are below the ADI value, 95.5% of the IMDLs were computed from either two of the major exposure pathways or one exposure pathway, which indicates that many nations did not provide enough information to protect human health against 2,4-D exposure. Figure 9 illustrates the 115 Glyphosate IMDLs computed from three major exposure pathways which were plotted in the empirical cumulative distribution form and compared to the log-normal random variable cumulative distribution with identical statistics. Four IMDLs were computed from three exposure pathways, 36 values were calculated from two of the major exposure pathways, and 75 values were computed from one of the major exposure pathways. The Glyphosate IMDLs span 6.11 orders of magnitude (1.43 × 10 −7 , 1.86× 10 −1 ), and are well dispersed over this span with the Pearson coefficient of 0.854. The correlation was heavily influenced by three IMDL data clusters at 0.166 mg/kg-day (35 values), 0.0335 mg/kg-day (30 values), and 2.86 × 10 −6 mg/kg-day (8 values Figure 9. Glyphosate IMDLs computed from soil, drinking water, and the commonly consumed agricultural commodities plotted in the empirical cumulative distribution form and compared to the log-normal random variable cumulative distribution with identical statistics.

Analysis of the IMDLs for the Widely Used Pesticides
The maximum IMDL value is 0.186 mg/kg-day, computed from the Guatemala national drinking water MCL and food MRL. The minimum IMDL value is 1.43 × 10 −7 mg/kg-day, computed from the Iraq drinking water MCL only. A total of 42 (27.1% of the total) IMDLs are above the Glyphosate ADI value. Although 72.9% of the Glyphosate IMDLs computed from the national jurisdictions are below the ADI value, most of the IMDLs were computed from either two of the major exposure pathways or one exposure pathway, which indicates that many nations did not provide enough information to protect human health against Glyphosate exposure. Table 8 provides the statistical summary information of the IMDLs for the 14 widely used pesticides. The weighted average of the Pearson coefficients by the IMDL number is 0.868. Aldicarb has a Pearson coefficient of 0.977, while Malathion has the low coefficient of 0.688, because the correlation is heavily affected by large IMDL data clusters. For these 14 pesticides, there are at least 931 IEDs (60.3 % of the total) in the non-random data clusters, and the data clusters are usually led by EU and WHO jurisdictions. There are at least 373 computed IMDLs (24.1% of the total) above the ADI values, which indicates that, for these 14 widely used pesticides, at least 373 IMDLs computed from the national jurisdictions cannot protect human health. The maximum IMDL value is 0.186 mg/kg-day, computed from the Guatemala national drinking water MCL and food MRL. The minimum IMDL value is 1.43 × 10 −7 mg/kg-day, computed from the Iraq drinking water MCL only. A total of 42 (27.1% of the total) IMDLs are above the Glyphosate ADI value. Although 72.9% of the Glyphosate IMDLs computed from the national jurisdictions are below the ADI value, most of the IMDLs were computed from either two of the major exposure pathways or one exposure pathway, which indicates that many nations did not provide enough information to protect human health against Glyphosate exposure. Table 8 provides the statistical summary information of the IMDLs for the 14 widely used pesticides. The weighted average of the Pearson coefficients by the IMDL number is 0.868. Aldicarb has a Pearson coefficient of 0.977, while Malathion has the low coefficient of 0.688, because the correlation is heavily affected by large IMDL data clusters. For these 14 pesticides, there are at least 931 IEDs (60.3 % of the total) in the non-random data clusters, and the data clusters are usually led by EU and WHO jurisdictions. There are at least 373 computed IMDLs (24.1% of the total) above the ADI values, which indicates that, for these 14 widely used pesticides, at least 373 IMDLs computed from the national jurisdictions cannot protect human health.

Conclusions
Over 145 worldwide soil jurisdictions from 54 nations (27.7% of the total nations) and 171 drinking water jurisdictions from 102 nations (52.3%) had provided the pesticide standard values. Over 90 nations (46.2%) had regulated the pesticide agricultural commodities MRLs which were collected in the global MRL database (2014) [16]. Only the U.S. had systematically promulgated the pesticide air MCLs. This research shows that many nations are lacking pesticide standard values for the major exposure pathways, especially nations in Africa, Asia, and South America. For nations which provided the pesticide standard values, human health risk uncertainty bounds and the implied total exposure mass burden method were introduced to examine whether current standard values in the major exposure pathways can protect human health. The results indicate that some pesticide standard values were derived too high to protect human health, even in a single exposure pathway, such as soil. This is because some jurisdictions derived the pesticide standard values without considering other exposure probabilities, did not derive them comprehensively and conservatively enough for human health risk control. In addition, the standard values for most commonly regulated pesticides often vary over seven, eight, or nine orders of magnitude, indicating that there is little agreement on the regulation of pesticide standard values among worldwide jurisdictions. This study will help worldwide pesticide regulatory jurisdictions to rationalize their standard values and provide some references for nations which do not yet have pesticide standard values.
Supplementary Materials: The following are available online at www.mdpi.com/1660-4601/14/7/826/s1, Table S1: Worldwide pesticide soil regulatory jurisdictions references and dates accessed, Table S2: Worldwide pesticide drinking water regulatory jurisdictions references and dates accessed, Table S3: U.S. pesticide soil regulatory jurisdictions references and dates accessed, Table S4: U.S. pesticide drinking water regulatory jurisdictions references and dates accessed, Database: worldwide pesticide standard values.