Where were the biblical Sedom and Amora located? South of the Dead Sea? Or north of it? I'll try to present arguments for both sides and to find a resolution. Keep in mind that I know nothing of archaeological finds or other such information which would affect the conclusions. In fact, my only resources are Tanach, a single journal article, and my geographic knowledge. But even within those sources, there's plenty to work with.
Arguments for the North
1. The initial description of Sedom (13:10) indicates very strongly that it is in the north. At this point, Abraham seems to be in the Beit El region. When he has to split up from Lot, Lot scans the horizon, notices Sedom, and decides to go east to there instead of north or south as Abraham suggested. From Beit El, the north end of the Dead Sea is easily visible and is directly to the east. In contrast, the south end of the sea is much further away to the southeast, and probably impossible to see from Beit El. Thus, the northern end seems a much better fit.
2. Even more convincingly, Lot's vision of the area can only correspond to the north end. The phrases "kulo mashkeh" and "kegan Hashem keeretz mitzrayim" bring up images of a lush forested area. Scientifically, it's hard to believe that the Dead Sea area ever looked nice and forested, or that a single rain-down of fire could permanently change the climate. Of course, if you read the verse carefully, you realize that that's not what it says. Egypt would be an uninhabitable desert if not for the river which flows through it. The Dead Sea area is also uninhabitable desert, but it too has a river flowing through it - the Jordan. The north end of the Dead Sea is "keeretz mitzrayim" - a desert with a life-sustaining river flowing through it, and "kulo mashkeh" - entirely irrigated from this river. This fits Lot's description perfectly, while the south end has no river, could never have been fertile, and thus is very hard to imagine as a setting for the story.
3. Not only that, but Lot specifically sets out for "kikar hayarden" - the plain of the Jordan river, which of course is in the north. Later on, "the kikar" is mentioned repeatedly as the site of Sedom, and presumably we are still talking about "kikar hayarden" in the north.
4. Biblical cross-references are ambiguous, with one exception. Breishit 19:22 tells us that Tzoar is a city close to Sedom. Devarim 34:3 seems to indicate that Tzoar is close to Jericho. If so, Tzoar and Sedom would be at the north end.
Arguments for the South
While the modern-day Sedom is south of the Dead Sea and most people assume that the ancient Sedom was there too, I'll ignore this widespread perception and try to focus on textual and geographic clues. Some are more conclusive, some less so.
1. The war between Chedarlaomer and Sedom was preceded (14:5-7) with a military campaign which took Chedarlaomer to a series of places. Several of these - Seir, Paran, Kadesh - seem to be south of the Dead Sea, while others are hard to locate. Nevertheless, if Sedom and its neighbors are north of the Dead Sea, this campaign seems rather out of the way and unrelated. If Sedom and company are to the south, then Sedom was located right in the middle of the campaign. The Sedom war and the previous campaign would then seem to fit together as parts of one larger conflict, which would explain why the Torah bothers to mention the campaign.
2. When Chedarlaomer finally battles Sedom, the king of Sedom is forced to flee, soon encountering the "be'erot chemar". Based on likely meanings of "chemar" and the geography of the Dead Sea area, it seems clear that the king of Sedom ran across and fell into a sinkhole created where the Dead Sea retreated and left behind salty terrain, unstable and prone to collapse. Many such sinkholes are common now at the south end of the Dead Sea, where due to human water use, the whole bottom third of the sea has dried up. Presumably such was also the case in Abraham's time (later on I'll present strong evidence for this). It thus makes sense to say the king of Sedom was fleeing near the southern end of the sea, when he fell into a sinkhole, and was later rescued.
3. Lot's wife was turned into a pillar of salt, and nowadays pillars and whole mountains of salt can be found south of the sea. This seems to indicate that she was "pilloried" in the south; though miracles by definition don't have to conform to nature.
4. The strongest argument for a southern Sedom may come from an article published in the Al Atar periodical, issue 12 (2003). The authors (Yoel Elitzur and Amos Frumkin) used geological data to calculate what the water level of the Dead Sea has been at each point over the last 4000 or so years. It turns out that there have been extensive variations in the water level. Most interestingly, in roughly 1200 B.C.E. the sea was at an unusual high peak of about 385 meters below sea level, while in 1600 B.C.E. the water level was much lower - perhaps 400 meters below, or comparable to the level today. As noted above, in the last century a similar change in water level has caused the bottom third of the sea to dry up entirely. Since the terrain was the same then as now, it's very probably that between 1600 and 1200 the southern third of the sea re-flooded as the water level rose.
This allows us to nicely explain a phrase which otherwise seems redundant and problematic. The battle takes place (14:3) in "the valley of Shidim, which is the salt sea". Why give both names? It seems that in Abraham's time (~1600 B.C.E.), the sea level was low and the area was a dry valley - "the valley of Shidim". By the time the Torah was given (~1200), the area had flooded. What was once a valley now became part of the Dead Sea. After giving the valley's original name, the Torah provides the "modern" name of the now flooded area, "the salt sea", so that readers in Moshe's time would know what is being referred to.
(Bible critics should note that the valley was once again dry from roughly 1000 to 100 B.C.E. :) )
In order for this re-flooding explanation to work, we must assume that the entire valley in question will flood and dry up again based on normal changes in the Dead Sea's level. This phenomenon has, of course, been observed in recent times at the southern area of the Dead Sea. But on the north, east, and west sides of the Dead Sea, the marine slopes are much steeper and only a sliver of land has been exposed as the water level has dropped. It's hard to imagine an entire valley with cities and agriculture fitting into this sliver. Thus, according to this theory, "emek hashidim" must be located at the southern end of the Dead Sea.
1. Most of the reasons I gave for placing Sedom (or at least the battle) in the south - the salty terrain, sinkholes, and the repeated flooding and drying of large areas of land - imply a southern location because similar phenomena are observed today only in the south. Nevertheless, if in ancient times a similar process had occurred at the northern end of the sea, then a northern location would be just as plausible. In fact, according to the Al Atar article previously quoted, if the Dead Sea were to rise to 380 meters below see level (an elevation not seen in modern times), then a substantial sliver of the Jordan river valley - roughly 6 by 3 kilometers - would become flooded. Were the water level to drop again, presumably the sliver would become salty and develop sinkholes. This area would then become an reasonable setting for the battle. Then it would be logical to say that both the city and the battle were located in the north.
2. All the reasons favoring the north relate to the city of Sedom, while all the reasons favoring the south (except one) relate to the battle with Chedarlaomer. If we assume that the battle took place some distance away from the city, then nearly all the of the arguments would be reconciled perfectly. I can think of no reason why the battle would have to be next to the city; the military action both immediately before and after our battle took place over a quite wide area and there is no reason to assume this battle must have been an exception. Thus, the city of Sedom would be at the north end of the Dead Sea, and the battle with Chedarlaomer would have taken place at the south end. Overall this seems like the best solution. Lot's wife, the only dissenting piece of evidence, can be explained by the obvious argument that miracles take place in whatever manner God desires.