The river which flows through the Indian states of Haryana, Punjab and Rajasthan and onwards into Pakistan had dried up in the latest Pleistocene. Therefore it may not have been the main water source for the later mid-Holocene Harappan settlements in that region and also likely was not the river Sarasvati described in Rig-Ved.
sarasvati river Map
That’s the gist from three separate studies on the Sarasvati river basin presented at a meeting of the American Geophysical Union and summarized in Science.

The finding that the river Sarasvati likely dried up or had a drastically reduced flow in the latest Pleistocene is not that surprising from the perspective of the broader climatic regime existing from Late Pleistocene to Holocene. However, it is a surprising finding if you consider another theory of Sarasvati water flow – the glacially fed Sarasvati – which I will come to later.

First, the climate change perspective.

After the Last Glacial Maximum, monsoonal strength over the Indian subcontinent has fluctuated over a millennial scale resulting in dry and wet phases lasting a few thousand years each.

A late Pleistocene arid phase affecting north west India is well documented by various studies on sediments and landscapes in Rajasthan, Gujarat and western Maharashtra. See this paper by Mishra and Rajguru and references therein. The studies by Gupta et al and Maemoku et al which suggest that between 15,000 years and 10,000 years ago the Sarasvati basin shows signs of drying up, likely reflect this phase of aridity.
This late Pleistocene arid phase was followed by an early Holocene wet phase reflecting a stronger Indian monsoon. Then by mid Holocene another episode of aridity set in which coincided with the ultimate demise of the Harappan civilization.

I don’t want to comment here on the implications of these studies regarding water dependency of the Harappan civilization or the Vedic Sarasvati problem.

Instead I want to point out that these new geological findings by Gupta et al and Maemoku et al do hint at an answer to another interesting geological question.. whether the Sarasvati was a glacially fed river in the past.

It has been hypothesized that the Sutlej and the Yamuna, both glacially fed rivers, flowed into the Sarasvati during the Holocene and by late Holocene around 2500 B.C to 1800 B.C avulsed or shifted course to their present positions. So along with climate change, rivers changing courses is another explanation given for the reduced water flow of the Sarasvati by late Holocene.

I think the time frame over which this problem has been thought out has a bearing on the choice between the climate change and river avulsion theories. If you only consider the Holocene and one episode of a river drying up (a larger Sarasvati drying up by 1800 B.C) then a Late Holocene river avulsion event seems as plausible an explanation as a climate change one.

But if more studies confirm Gupta’s and Moemuko’s finding that the Sarasvati had indeed dried up in the late Pleistocene then that would weaken the theory that the Yamuna and Sutlej flowed into the Sarasvati until the late Holocene.

A glacially fed Sarasvati would likely not have dried up even during the late Pleistocene arid phase. Summer melting of the large Himalayan glaciers built up through the Last Glacial Maximum would have maintained enough water flow in the Sarasvati throughout the year.
The simplest explanation of these alternating phases of river drying and rejuvenation through the late Pleistocene into the Holocene is that the Sutlej and the Yamuna never flowed into the Sarasvati. The Sarasvati was not glacially fed, but by streams originating in the lower and sub Himalayas, and its fortunes always depended on the strength of the Indian monsoons.

One scenario of rivers shifting courses that does fit this extended late Pleistocene-Holocene time frame depends on the Sutlej and / or the Yamuna getting diverted away from the Sarasvati much earlier, around 14,000 years ago or so. The drying of the Sarasvati brought out by these two studies then reflect a combination of river avulsion and climate change.

Either way, these new findings make it cumbersome to support the theory of a glacially fed Sarasvati river.