You have misrepresented a very complicated mathematical model when you concluded that our research "shows TNR programs have no impact on feral cat populations." This was not the published conclusion of the paper.
The goals of TNR are multifaceted and include:
1) prevention of the birth of feral kittens (75% of which may not survive to adulthood)
2) improvement of the welfare of the cats that live in managed colonies
3) reduction of the size of targeted colonies
4) reduction of cat populations in targeted neighborhoods
5) reduction of cat populations in larger regions
Our group and other scientists have published numerous papers in scientific journals that have proven the success of TNR in accomplishing goals 1-4; some of these are referenced in the current paper if you are not familiar with them. These are well-documented significant impacts of TNR on feral cat populations.
Not surprisingly, the new model did not predict that the scale and duration of the two TNR programs in the study would produce a statistically detectable impact at the larger county-wide level. You incorrectly state that "From 1992 to 2003, 14,452 cats were trapped, neutered, and released out of an estimated 240,690 feral cats in San Diego County, CA. From 1998 to 2004, 11,822 cats were trapped, neutered, and released out of an estimated 36,398 feral cats in Alachua County, FL." The number of cats trapped is the accumulated total over the years whereas the estimated population sizes are approximately the same each year (but not necessarily the same cats). Thus, it is improper to suggest that 6-32% of the total population was trapped as you state. In reality, only 0.5-6% of the total county cat population was trapped each year. This may seem like a small point, but it is critical to avoid misuse of the model. The model predicts that increasing the trapping percentage to 14-19% annually would successfully reduce cat population growth. Thus, TNR is not a failed concept, it simply needs to be practiced on a larger scale.
It is not uncommon for lay persons to have difficulty understanding the concept of models. Models are mathematical predictions of what might happen when certain conditions exist in the environment. Models provide a starting point for developing programs and public policy. However models are not the same as an actual study of the program or policy outcome. Perhaps one of the most graphic example of models gone wrong occurred in the presidential election in Florida. Even though the models were built by experienced analysts using up-to-the-minute data, they predicted the wrong outcome of the election. Similarly, a recently published model on feral cat control (Anderson MC, et al, JAVMA 2004) predicted that unmanaged cat populations would increase exponentially. This would be a population growth pattern that is virtually unheard of in any mammalian species. Since most feral cat populations remain unmanaged, the Anderson model would predict that we would literally be knee-deep or worse in cats, which is clearly not the case.
All this means that models provide a great starting place for addnaged, the Anderson model would predict that we would literally be knee-deep or worse in cats, which is clearly not the case.
All this means that models provide a great starting place for addressing problems, but they should be replaced as quickly as possible with real studies that test the actual (instead of theoretical) outcomes of control programs.
BTW, have you seen the recently published model which predicts that reducing cat populations would actually cause more harm to birds due to a resulting increase in rat populations (Fan M, et al, Bull Math Biol, 2005)? Interesting!
Mark Twain said, "Get your facts first, then you can distort them as you please." While this may be a strategy in the war against feral cats, I hope that you will not intentionally distort the growing body of scientific literature that we desperately need to make informed decisions about managing feral cats.
Julie Levy, DVM, PhD, ACVIM