Which Science? Rethinking the Salt River Wild Horse Debate

For the past few months, I have watched people on both sides of the Salt River wild horse debate accuse the other side of ignoring—or even opposing—“the science.”

It has left me shaking my head.

Then I realized something: much of this disagreement is not really about whether science matters. It is about what people mean when they say “the science.”

One side wants to emphasize genetics: How many horses are necessary to maintain a genetically viable population over time? When it says “no scientific study” has been done and that there is “no science,” what it really means is “no genetic study.”

This is why the other side in this debate shakes its head. What do you mean, no scientific study? No science? They then point to the various scientific studies that have already been completed.

Strip away all of the rhetoric, and I think the two arguments look something like this:

Here is the argument presented by the Friends:

We need a genetic study to determine what population is necessary to maintain the long-term genetic viability of the herd. Until we have that information, we do not know how far the population can safely be reduced. Once science establishes the genetic requirements of the herd, we will have the information necessary to determine whether—and how far—horse numbers can be reduced.

Here is the argument presented by the other side:

We already have substantial scientific information about the condition and carrying capacity of the range. That science indicates that the land cannot sustainably support the present number of horses. A genetic study may provide useful information about the herd, but genetics cannot override the ecological limitations of the land. If the range can sustainably support only a certain number of horses, discovering that a larger population would be genetically preferable does not somehow create more forage.

In other words, both sides are appealing to science. They are simply privileging different scientific questions.

And I think both are making the same fundamental mistake.

There are not one, but three scientific questions that must be answered—and they have to be understood in relationship to one another: range, fertility, and genetics.

Here is how I understand them to be related.

The foundation: Range

On this point, I actually agree with an important part of the argument made by Robin Silver and his coauthors in their recent editorial: the carrying capacity of the land has to be foundational to any discussion of horse numbers.

Where I part company with Silver et al. is not over the importance of carrying capacity. It is over whether we currently have sufficient range science to treat the existing carrying-capacity estimates as the answer.

I do not believe we do.

The existing estimates quoted by Silver et al. are based upon a particular definition of the range and a series of assumptions about forage availability, utilization, horse diets, and allowable use. Those calculations may ultimately prove broadly correct. They may not.

When I say we need a comprehensive “range study,” I mean much more than simply measuring some vegetation and calculating how many horses it might feed. I mean a systematic assessment of the entire ecological system in which the horses are being managed.

First, we need to define the land being studied. Then researchers would systematically sample vegetation across that range—not simply a few locations—to determine how much forage is actually available, what kinds of forage are present, its nutritional value, and how availability changes by location and season.

They would also examine water availability, drought and rainfall patterns, soil and vegetation condition, riparian areas, invasive species, and evidence of grazing pressure and ecological change.

Just as importantly, the horses are not the only animals using that landscape. A comprehensive study would need to consider forage use and ecological impacts from horses alongside wildlife and livestock, rather than automatically attributing every observed condition to horses.

Researchers would then follow those conditions over time. A snapshot taken during one season or one unusually wet or dry year cannot necessarily tell us what that landscape can sustainably support over the long term.

Only after assembling that information would researchers model carrying capacity under different conditions and management assumptions: What can this range support during an average year? During drought? How much forage should remain ungrazed to protect vegetation and soils? What happens under different horse populations?

And importantly, I would not expect a good range study to announce, “The scientifically correct number is 73 horses.”

I would expect something more like: under these assumptions, with this amount of land available, and under these environmental conditions, a population within this range creates these ecological risks; under a different population or different assumptions, the risks change.

So, yes: range is the foundation. In that limited sense, Silver et al. and I agree.

The middle: Fertility

Once we understand the range, the next question is whether we can maintain a relatively stable horse population within its ecological constraints.

That makes fertility especially important on the Salt River, where PZP has been used extensively for years. The fact that there is no publicly available record identifying which mares have been treated, how frequently or for how many consecutive years they have been treated, combined with the absence of an independent scientific evaluation of the long-term effects of the fertility-control program on the herd, raises significant questions that need to be addressed.

And this is where I think the work being done by Paul Martin deserves serious attention.

Martin has spent considerable time examining the available records concerning PZP use on the Salt River herd, including information contained in management reports submitted to AZDA. He has raised important questions about the number of consecutive PZP treatments received by individual mares, whether and when mares have been given periods without treatment, how frequently mares intended to remain reproductively available have nevertheless been treated, and what the scientific literature tells us about delayed return to fertility or permanent infertility following repeated treatment.

He is asking questions that can—and should—be answered scientifically.

Fertility management is itself a scientific question.

It does little good to identify a sustainable population if the fertility-management system being used to maintain that population may itself change the demographic and genetic viability of the herd.

The top: Genetics

Finally, there is genetics.

Here I think the Friends of the Salt River Wild Horses and other advocates who pushed for a genetic study raised an important scientific question: before making long-term decisions about this herd, we need much better information about its genetic health and what population characteristics are necessary to maintain genetic viability over generations.

Governor Hobbs ultimately agreed that this question deserved study. Her administration worked with the Arizona Board of Regents to secure funding, and the Regents have now approved a $325,000, two-year research project involving Northern Arizona University and the University of Arizona. The study will examine the herd’s genetic makeup, inbreeding, population viability, and other factors intended to inform future management decisions.

I strongly support that study.

The genetics study can help us understand the genetic consequences of different population sizes and management choices. It can tell us about genetic diversity, relatedness, effective population size, inbreeding risk, and the likely long-term consequences of different management scenarios.

What I do not expect it to do is hand Arizona a single magic number and say: “This is how many horses there should be.”

Even if it identifies a minimum population necessary to maintain an acceptable level of genetic diversity, we still have to put that finding alongside the other two scientific questions.

Can the available range support that population?

Can fertility be managed in a way that maintains that population and preserves sufficient reproductive contribution across the herd?

And what happens if the answers conflict?

Suppose the genetics research indicates that 150 horses are necessary to maintain an acceptable level of long-term genetic viability, while a comprehensive range study concludes that the land currently available can sustainably support only 75.

Which scientific study wins?

Neither.

At that point, science has done exactly what we asked it to do. It has identified the risks and consequences associated with different choices.

And that takes us right back to policy.

So Where Does This Leave Us?

We need to determine what land should be available to the horses and then conduct a current, comprehensive assessment of what that range can sustainably support. We need an independent examination of the long-term demographic, reproductive, and genetic consequences of years of intensive fertility management. And we need the genetic research that Governor Hobbs and the Arizona Board of Regents have now initiated.

Range. Fertility. Genetics.

Together, those three provide the scientific foundation we need.

But they may not give us the same answer.

If the range can support fewer horses than genetics tells us are desirable, someone will have to decide what to do. Do we make more land available? Accept greater ecological pressure? Accept greater genetic risk? Change the fertility-management strategy? Consider genetic intervention?

Science can tell us about the probable consequences of those choices.

It cannot choose among them for us.

And that brings me back to where I started. Silver and his coauthors are right that carrying capacity matters. The Friends and other advocates are right that genetics matters. Paul Martin is right that the long-term consequences of fertility management deserve serious scientific examination.

The mistake comes when any one of those questions is treated as though it alone constitutes “the science.”

It doesn’t.

Let’s do all of the science. Let’s do it independently, transparently, and well. Then let’s have the much harder—and much more honest—public conversation about what we are going to do with what it tells us.