Most air pollution monitors have repeatedly underestimated levels of toxic lead in the air, according to an internal Environmental Protection Agency memo released yesterday by an environmental group and a member of Congress.

The memo, obtained under a Freedom of Information Act request and released jointly by the Natural Resources Defense Council and Rep. Toby Moffett (D-Conn.), provides further ammunition to critics of EPA's proposal to relax restrictions on lead in gasoline.

The memorandum said that the lead level in most areas is probably higher than tests indicate because monitors were located too far from roads, in areas with little traffic or at elevations considerably higher than ground level.

The memo also concluded that there appears to be "a very strong correlation" between the amount of lead in gasoline and the amount of lead in the air, something EPA has not publicly acknowledged.

The memo was dated Jan. 27, 1982, a month before EPA announced it was thinking of relaxing the lead level standard.

Responding, EPA spokesman Byron Nelson said, "This issue is being looked at as part of the decision-making process. It is only one of many issues being looked at. The review is currently ongoing and absolutely no decision has been made on lead." Nelson declined further comment because he had not seen the memo. EPA employes involved with the issue could not be reached.

Lead has long been known to have adverse health effects, particularly on children. But the lead industry and some refineries have argued that the 1979 restrictions gradually reducing gasoline lead levels have so improved the air quality that there are no longer health reasons for retaining the regulations. They also complained that the standards were extremely costly.

Reacting to industry pleas, the Presidential Task Force on Regulatory Relief directed EPA last August to relax the lead level standard. EPA rejected that approach as political suicide and instead announced in February that it was considering several proposals ranging from elimination of the standards to retention of the current rule.

The memo on the monitoring problems was written by Robert Kenney, chief of EPA's state and local control programs section, in anticipation of the agency's announcement.

Kenney said he was asked to review the impact that weaker lead standards would have on lead concentrations in the air. But he said the quality of the data made that almost impossible.

"The vast majority of lead monitoring which has been done in the past has been at sites which were not designed to measure maximum lead concentations," Kenney wrote. Except for a few newer sites, most monitoring stations were designed to read other pollutants and do not meet the lead-monitoring criteria for siting or quality assurance established by the agency last fall.

Furthermore, stations meeting those criteria did not have to begin their monitoring until this month and do not have to begin reporting to EPA until the end of the year.

"I am stunned that the administration has misled the public on the serious nature of the nation's lead pollution problem," said Moffett, chairman of the House Government Operations subcommittee on the environment. He said it would be "irresponsible" for the Reagan administration to weaken the lead rule.

High levels of lead affect children's nervous systems, kidneys and brains. Low-level exposure has been linked to learning disabilities. About 500,000 children or 4 percent of those between the ages of six months and 5 years have high levels of lead in their blood, according to the National Center for Health Statistics. That figure jumps to 18 percent for urban minority children.