We identified Tanzania as a hotspot in 2007 and again in 2010. Yet Tanzania still has unprecedented levels of poaching, to the point that some of the most important elephant populations in Africa have declined by over 65%.
As the largest land mammal, African elephants are keystone species in their environment. We are currently losing up to 50,000 African elephants to poaching each year with fewer than 470,000 remaining. That's one-tenth of the population per year.
This loss rate must be contained. Stopping demand for ivory is too slow a process to do it alone. We urgently need to stop the killing. Loss of elephants is already causing serious ecological and economic damage as well as threatening national security with organized crime across Africa and abroad.
We addressed this problem by using DNA to determine the geographic origin of large ivory seizures, each weighing over half a ton. These large seizures, each worth more than US$1 million, bear the signature of large transnational organized crime (TOC) syndicates. Large seizures also represent about 70% of all ivory smuggled by weight.
All but one of the large seizures we examined that occurred over the last decade came from just two areas. Targeting these two areas for law enforcement could stop the largest amount of poaching-related mortality in Africa and choke at the source the criminal networks that allow this TOC to operate.
How we did it
We statistically assigned the genetic makeup, or genotype, of ivory to a DNA reference map we assembled from elephant samples collected across Africa over the past 15 years. We primarily acquired the DNA for the map from elephant dung samples, taking advantage of their ease of collection.
We simultaneously developed methods to extract the same DNA from ivory, allowing us to determine the origin of ivory by statistically matching the ivory genotypes to the DNA reference map.
We collected over 1,500 unique elephant samples from across Africa, using a sampling method that maximizes the chances that each sample came from a separate family. We extracted and amplified DNA from up to 16 hypervariable genetic markers from each sample, giving us high precision in assigning ivory to its place of origin.
We increased assignment accuracy by developing novel software to create a smoothed continuous DNA reference map from these samples, reflecting genetic differences among elephants the occurred over space and time.
We showed that the method is highly accurate by blindly assigning each of the reference samples, testing how close we could get to their true origin. We were able to assign any Africa elephant sample to less than 300 kilometers (186 miles) from its true origin from anywhere in Africa, often even closer. Most protected areas are farther apart than that distance.
We then examined 28 large tusk seizures that occurred between 1996 and 2014. The seizing country provided the samples, which were either subsampled by them, by INTERPOL or by us using methods we designed to assure they were representative - that is, that all the locations that contributed ivory to that seizure were adequately represented among the provided samples.
A small piece of ivory, the size of a large coin, was then cut from the base of each selected seized tusk and shipped to our lab to determine its origin.
In matching tusk seizures to our DNA sample map, we learned that poachers return to same area repeatedly to fill their quotas and that ivory tends to be shipped from a different country from where it was poached.
We also learned that the number of major hotspots being poached to supply large criminal syndicates are remarkably few and slow to change. This makes hotspots from the recent past excellent predictors of future hotspots.
Since 2006, these hotspots were concentrated in just two areas, one for forest elephants and the other for savanna elephants.
The forest elephant hotspot occurred within or in close proximity to the Tridom (Tri-National Dja-Odzala-Minkébé) protected ecosystem that spans northeastern Gabon, northwestern Republic of Congo, and southeastern Cameroon, and the adjacent Dzanga Sangha Reserve in southwestern Central African Republic.
The savanna elephant hotspot was concentrated in and around southern Tanzania, especially the Selous Game Reserve but including the adjacent Niassa Game Reserve in northern Mozambique, eventually extending north to Ruaha National Park and the adjacent Rungwa Game Reserve.
What's particularly sad is that my lab and others first identified Tanzania, and the Selous in particular, as a major hotspot in 2007 and again in 2010. Yet, Tanzania still has unprecedented levels of poaching, to the point that some of the most important elephant populations in Africa have declined by over 65%.
The magnitude of this sustained poaching, occurring in the same area for over a decade, could not have occurred without high-level corruption. Dealers had to be able to obtain the ivory and readily move it out of the country without detection over the past decade.
Equally of interest, the multi-ton Malaysia seizure we analyzed from December 2012 contained large amounts of ivory from both hotspots, suggesting a link between the major operatives in these two hotspots.
Wildlife Crime is now the fourth or fifth largest TOC, worth $20 billion if you include fish and timber. Fighting any of these TOCs is a challenge: the networks that allow them to operate are complex.
Targeting this very small number of hotspots, however, could stop a huge portion of the killing. It could also choke at the source the major flow of ivory from entering into elaborate criminal networks where it becomes far more difficult and expensive to trace.
Unlike transit countries, source hotspots can't change very quickly. They require large numbers of elephants and considerable infrastructure to move the ivory out of the country without detection. This infrastructure must also be developed in the next source country before it can become a major hotspot on the scale we have identified. Thus, our methods should be able to detect this.
Our lab already has the needed precision and ability to process seizures as they come in. We can accurately assign a large seizure within three weeks from the time it enters our lab. We merely require that the seizing countries promptly provide us subsamples from their recent large seizures for geographic assignment.
This was made much easier in 2013, when delegates of the Convention on International Trade in Endangered Species (CITES) unanimously passed Decision 16.83, urging all seizing countries to turn over samples from their large seizures for origin analysis within 90 days of the seizure.
Since then, we have been collaborating with ICCWC (INTERPOL, UNODC, CITES, World Customs Organization, World Bank) and the US Department of State's Bureau of International Narcotics and Law Enforcement Affairs (INL) to help us acquire these large seizures on a timely basis.
Over 90% of recent seizures since 2013 are either in our lab, en route or promised. If we continue to acquire seizures at this pace, we will surely be able to detect any changes in hotspots on a real-time basis. This approach is applicable to other species and TOCs as well.
Samuel Wasser is Research Professor of Biology at Center for Conservation Biology at University of Washington.