The new rules would require only modest effort from heavy coal states such as Indiana, Ohio, West Virginia and Kentucky while requiring big cuts from Texas, Louisiana, and New York, according to an analysis from Bloomberg New Energy Finance research company.

The Bloomberg analysis also found the plan would allow for slight emissions rises in some states.

“The bottom line is pretty clear. There are some major differences between what states are being asked to do in terms of being asked to reduce their total CO2 emissions on a volume basis,” said Ethan Zindler, head of policy analysis for BNEF.

The Environmental Protection Agency said from the outset that the plan was “customised” according to the energy mix and reductions potential in each state. But the agency said those targets would result in a national average of a 30% cut in emissions from 2005 levels by 2030.

But the differences between the expectations for states under the rules – and the extreme complexity of the plan – have exposed the EPA to charges of picking winners and losers.

Analysts said they were still in the early stages of interpreting the agency's figures and methodology.

But they said they were struck by the differences between the targets for the individual states.

Some of the apparent winners under the proposed EPA regulations are also some of the most coal-heavy states such as West Virginia and Kentucky. Kentucky is allowed a modest rise of 4% in emissions by 2030, according to the Bloomberg analysis.

North Dakota and Ohio would also see relatively little change in emissions by 2030 under the EPA plan.

California, Missouri, and Nebraska would also be allowed to increase emissions between 12% and 16%.

But Minnesota, New York, Oklahoma, Texas, Arkansas and Louisiana were all assigned targets above the 30% national average. The plan calls for Texas to cut power plant emissions by 44% and Louisiana by 68%, according to the Bloomberg analysis.

The EPA did not respond to specific questions on whether states would be allowed a rise in emissions under the plan.

"The 2030 goal for each state is tailored based on what can be reasonably achieved from the various emission reduction strategies available to each state,” EPA spokesperson Enesta Jones said in an email. “The difference in state goals reflects the diversity among states’ current generation portfolios. Most importantly, a lower state goal does not necessarily mean 'doing more' – for example, some states have already made commitments for coming years that will put them on track to meet their state goals."

The regulations unveiled on Monday would allow states to come up with their own plans for achieving their target emissions rate – the amount of carbon dioxide emitted for every megawatt hour of electricity generated.

The EPA said states could choose to improve existing plants, switch to cleaner burning natural gas, expand their use of renewables or encourage customers to use less electricity.

The plans also take into account scheduled construction and retirement of power plants in each state. The EPA said it set its high target for Washington state by taking into account that its own coal-fired power plant was due to retire before 2030.

Vermont is not affected by the rules because it does not have any fossil fuel plants. Washington DC has one small coal plant, which supplies Congress, but it is too small to fall under the new rules.

However, some analysts said on Tuesday they were struggling to understand how the EPA derived its targets for the individual states.

“Some states are winners and some states are clear losers and whether you are a winner or a loser doesn't really necessarily depend on how much C02 you are emitting,” said Andrew Weissman, a senior energy advisor at Haynes and Boone law firm. “It depends on how this very complex methodology applies.”

He said his early analysis suggested that some states – such as New York which is already part of a regional cap-and-trade market – were facing challenging targets, while others got off relatively lightly. “It seems that some states get off very easily and with far more modest burdens than what one might expect,” he said.

"It is not clear that this program has a whole lot of bite for a lot of very high carbon emitting states in the rust belt area, and by contrast it appears that it has a much heavier impact along the southern tier of the country including the south-east and Texas.”

He said his analysis did not indicate a rise in emissions in any state.

Another analysis by the climate change centre at Georgetown University Law School also saw vast differences between the emissions reductions that would be required by the individual states.

The Georgetown analysis did not indicate an emissions rise in any state.

The rule will now come under a 120-day comment period before it is finalised next year. In some cases, states will have until 2018 to finalise their plans – and those plans may not come into force until 2020.