And climate and water.

A Purdue University scientist has shown man-made changes to the landscape have affected Indian monsoon rains, suggesting that land-use decisions play an important role in climate change. . .

Analysis of soil moisture showed that before monsoon rains came, the northwest region had become as much as 300 percent wetter in recent years relative to the past 30 years, which has been attributed to irrigation from groundwater to sustain intensified agricultural production. This wetter surface causes cooling that weakens the strength of low pressure necessary for monsoons to progress into northern India. . .

"In this case, you need a warm, dry surface to advance the monsoon," said Niyogi, whose findings were published in the journal Water Resources Research. "Because of increased irrigation, you now have a wet, green area, which does not allow the monsoon to reach far enough north." . . .

Urban areas, on the other hand, are being pounded with rain when it comes. Niyogi said there have been storms in some urban areas that drop as much as 37 inches of rainfall in a single day.

Analysis of the areas that have received increases in heavy seasonal rainfall, based on Indian Meteorological Department and NASA satellite data, showed that those areas were experiencing fast urban growth. Areas where seasonal rainfall decreased were determined to have slow or no urban growth.

"You only see these types of heavy rainfall events in those areas with heavy urbanization," said Niyogi, whose research on the urban effect was published in the International Journal of Climatology. "The more urbanization spreads in those areas, the more of these heavy rain issues we'll see and the more flooding will become a problem."

Niyogi said there are two theories on why that's happening. The first says that urban landscapes create heat, which extends into the atmosphere and energizes storms. The second theory is that pollution created in urban settings interacts with passing clouds and increases rainfall. . .

The next step in this research is to examine landscapes in the United States to see if development has affected weather patterns historically.

It's not clear from this article but if moisture is being wrung out of the air passing over urban areas that too could account for the monsoons being weaker by the time they reach rural agricultural areas. This would fit the data that the total amount of water falling hasn't changed, but that it falls in different places than in the fairly recent past.

Something like that happens with mountains: the rain shadow effect. Wet air piles up against the mountains that block its way. The warmer air that rises and so gets across the barrier cools, is unable to hold so much water, and drops it before passing over the peaks. One side of the mountains is extra wet and the other side is extra dry compared to what it might be if the land was flat.

I'd be interested in an analysis that accounted for the path of monsoon moisture as it passes over land to see if there was a relationship of this sort. If the urban areas are in effect scraping the water out of the passing air - and flooding themselves in the process - it puts a very different spin on the observed events, and would require very different steps to mitigate those adverse consequences.