The main objection to developing the oil resources in the Arctic National Wildlife Refuge (ANWR) is the potential impact on the Porcupine Caribou Herd that migrates through the area. The arguments are made in terms of broad generalities, along the lines of “caribou are in the same area, so they are threatened.” If one tries to quantify the threat, is turns out that perhaps 1% of the herd might be impacted in some way, but the actual reduction in the herd size is probably a good deal less.

The part of ANWR at issue is called Area 1002. The Herd of about 200,000 animals has about 2.8 million acres of calving area of which 43%, 1.2 million acres, is somewhere within the 1002 area where drilling to occur. Area 1002 has 1.5 million acres of which 2000 acres is proposed to be used for drilling operations. The claim is that “40% of the herd is threatened by the operations.”

We need some concepts and terminology to analyze the situation. Consider, “A car entering Manhattan threatens 20% of the population of New York City.” This is a true statement. Cars run into pedestrians and hit other cars, so they pose a threat to any potential pedestrian or driver on Manhattan. About 20% of the population of New York City is in the potential victim class and on Manhattan. I didn’t say where the car was on Manhattan, so the threat is posed as general one. If I tell you the car is emerging from the Midtown Tunnel and crossing the island, then the folks in Harlem are safe, and we can lower our estimate of the threatened population. I didn’t specify the level of threat. The car might transit the island without hitting anything. But unquestionably it might hit someone, so there is a threat.

By comparison, the statement “A terrorist poison gas attack threatens …” has the same structure. It is a true statement. The difference between the car and the poison gas is a quantitative one, not a logical one. Evaluating the respective threats depends upon depends upon the probability of the threat being realized and the consequences if it occurs. The probability of the car hitting someone is the probability of a car entering the island times the probability of hitting someone. For terrorism, it is the probability of an attack being launched times the probability that attack will not be foiled by authorities.

Back to the caribou, the threat can only be logically evaluated quantitatively. Is ANWR development more like a car passing through a city, or more like a terrorist attack?

We can safely assume that caribou will not have calves right on top of a drilling platform or other occupied area. Thus 0.057% of the calving area would be directly affected by drilling operations. However, it’s also not likely that a caribou would choose a calving site extremely close to an occupied area. There is research that says that caribou populations drop by two-thirds within two kilometers of a traveled road. We can also assume that caribou do not care about roads that have no traffic.

All of the heavy transport and drilling operations must be done in the winter when the ground is frozen solid. That is because the soft ground will not support the weight of the transported equipment. There are no caribou in Area 1002 in the winter. Thus the long access roads and the roads interconnecting platforms are not at issue. However, there would be some traffic during the summer to support maintenance and operations. Thus the affected area will be larger than the drilling area.

The only estimate I’ve seen from opponents implies that all of Area 1002 will be crossed by roads 3 km apart, thereby chasing out all caribou. That is unreasonable. There are plenty of pictures of caribou quite near the Prudhoe Bay facility, and during the time that Prudhoe Bay has been operating, the caribou population in that area has grown by a factor of five. The growth was natural, and it is possible that it might have grown more if the Prudhoe Bay operations were not there. What is precluded is any possibility that caribou are affected many miles from the facility, because if that were true the population would have been decimated, which has not happened. The ANWR drilling area is much smaller than that at Prudhoe Bay, due to advances in technology that permit drilling out horizontally from a central platform.

Being fairly generous to opponents, let’s suppose that the affected area is fifteen times the actual occupied area. That would make the affected area about 1% of the calving area.

What happens to the caribou that will not have calves in the affected area? The other 99% of the area is available, so they will go into the unaffected area. Opponents suppose that the drilling operations will occur on what is the most prime area, and that the other 99% is not as hospitable. They suppose that as a result, the rate of calf survival will drop. If, for example, the survival rate of calves dropped by 30%, then the surviving birth rate of the herd would drop from X to 99.7% of X.

The natural survival rate of caribou fluctuates dramatically. Among the factors mentioned in the literature are vast swarms of mosquitoes that attack the herds, driving the animals to different locations depending upon the vagaries of the mosquito population. There are also long cycles of predation over which populations of wolf and prey can swing dramatically. Opponents make no attempt to estimate ANWR impact relative to the natural forces at work. Without a convincing data to the contrary, 0.3% decline in birthrate for the total population due to drilling is not plausibly significant. Even if it were the full 1%, it would not be significant.

Opponents go on to argue that while the oil we get out is temporary relief, the ecological damage is permanent. “All of the oil rigs will permanently scar the landscape.” Oil rigs are expensive items used temporarily for drilling holes, and they are never left behind after the drilling is complete. There is a possibility that pumps and piping might be left behind after the field is exhausted. That’s easily avoided by passing regulations that require it to be removed.

There might be minor oil spills in the drilling area, but such spills are readily cleaned. Keep in mind that the great disaster of the Exxon Valdez spill is not permanent. The areas are recovering. New technology and requirements for double-hulled tankers minimizes the tanker threat, but as long as we are importing oil there will be tankers from somewhere.

It’s now common for complex problems to be analyzed mathematically by computer simulation. Is there any rational set of assumptions that leads to to permanent damage to caribou herds? Even significant temporary reduction in herd size? I think if there were, we would have been told. Instead, everything is kept at the level of “sensitive habitat” and “scarring the landscape.” No rational case can be made at that level.

The reason we ought to drill is not to reduce the price of oil, but rather to reduce foreign trade deficits. That was discussed in a previous post.