The Hari arises in the Hindu Kush mountains and flows west through the northwest part of Afghanistan, through the very ancient city of Herat, and from there it turns north and makes part of the border between Afghanistan, Iran and Turkmenistan. There, the name changes and it is called the Tejen, or Tedzhen (depending on whose map you're reading). In times of drought, the river has been known to dry completely (the year 2000 it did). The river was called the Arius by the Greeks, and was mentioned in the Rigveda.
That is not, however, what makes the river interesting.
Have a look at this typical map of Afghanistan:
The Hari, or Harirud as it is shown on the map, is in the upper left hand corner. Here, I'll enlarge that corner of the map and thicken the line of the river so it is clearer:
Take note that the river rises up towards the corner of the map and then just ends. This is where the river reaches the Kara-Kum, or Garagum, Desert. On most maps, this is all that's shown. Sometimes, there's a little arrow. That's it. A sizeable river, the most important in the area, upon which more than two million people depend upon for water, and it just dries up in the desert.
Let's have a look at it with GoogleEarth:
What's shown above are the 'plumes' of irrigated land that occur at the ends of the Hari/Tejen and Murghab Rivers. The Murghab is also shown on the Afghanistan map above, it flows up to the city of Mary, or Merv (also dating back to ancient times). When I mapped this area for my world back in 2007, the Tejen plume was about a quarter of this size on GoogleEarth. The Mary plume hardly existed at all. Most of the development in this part of eastern Turkmenistan is very recent.
Here's a close-up to look at, before checking out the region on Google yourself:
The river is simply sliced up and sliced up until the ends disappear. A similar thing has been going on with the Syr Darya and Amu Darya rivers, both of which used to create the Aral Sea ... but the irrigation has become so extensive now that the Aral has ceased to exist. The water never reaches the basin where that sea used to be (and the land which the sea resides in is poisoned due to poor environmental practices of the former Soviet State).
It's fair to argue that both the towns of Tejen and Mary were likely irrigating the water of both rivers as far back as their origins, well before the time of Christ. There's no question the water use could not have been this extensive ... which means the rivers back in ancient days (perhaps even into the 17th century, where my world takes place) reached much further into the Kara-Kum desert. There are structures in the sand north of Mary that suggests the Murghab simply spread itself onto the sand until the last free-moving water grew thin enough to make a mud plume. That would be an interesting and dangerous feature in a desert - a thin sheen of water, flowing across a flat during the late spring, and rich in sink holes and quicksand through much of the remaining year, with mudbowls where the water managed to pool for several months before completely drying up.
In all honesty, what must these places have seemed like to people who could never get above them and look down?
Northwest of the Hari/Tejen river there are dried lake bowls, some of them enormous, that suggests that perhaps before human beings the Hari might have created a smaller version of the Aral Sea. I haven't been able to find much online about it. Part of the outflow of the Amu Darya, which emerges from between Afghanistan and Tadzhikistan, rising in the Pamirs/Kashmir ranges, has been diverted by Uzbekistan farmers into a formerly dried bed called Sarygamysh. All that wikipedia says is that this lake was allowed to dry in the 17th century ... but I wonder if, 5,000 years ago, the Hari was able to flow this far north. This ancient map shows the Murghab flowing into the Amu Darya, without any sign of the Hari/Tejen at all:
(pity, but the map has been deleted from the net and I don't have a copy)
But ancient maps are notoriously unreliable. Still, I'm curious. Are there any geologists/hydrologists who have trekked across Turkmenistan, who might be in the house?