The FAO blithely projects a doubling of the population of farmed animals but such a doubling would presumably require a doubling of feedstuffs for those animals.  The world is already using half of the available habitable land (71% of all land) and large quantities of ocean animals to feed the existing farmed animal population.  If twice as many farmed animals have to be fed, then presumably that would require using all of the habitable land for agriculture (to produce feed for the greater number of animals.  That leaves no land for wild animals or forests (or other human needs).  This is undoubtedly a simplistic calculation but doubling the number of farmed animals would certainly have disastrous consequences for biodiversity and current land-based carbon sinks.

I would like to see a projection that looks at future animal populations plus the options for feeding those animals from existing land and marine sources.  

In addition, just because someone projects that the world will reach a particular condition does not mean that such a condition is inevitable.  At one point, population scientists projected that the world population would peak at around 12.4 billion people.  

However, their projections have fallen. Forecasts now suggest that there will be 8.9 billion people by 2100 (1 billion more than today).  That number could be driven even lower if the world was better at preventing a larger portion of unintended pregnancies.  

Andrew Rowan, D.Phil.

President, WellBeing International

I empathize with Akash Kulgod's efforts to estimate the owned and street dog population of India.  There are no good surveys of the national dog population.  The government does publish estimates of the dog population but they are hopelessly wrong.  However, there is an interesting method that could produce relatively accurate estimates of India's dog population.  It turns out that the relative population of dogs in India (that is the number of dogs per 100 or 1,000 people) varies inversely with human density and one can capture that variation by plotting dogs per 100 (or 1,000) people against log human density in a particular community.  For example, a survey in the state of Haryana reported that there were approximately 25 million dogs in the state.  The trendline equation for the State of Haryana is y (dogs per 1,000 people) = -100 X (log human density per sqkm) + 440.   A survey of dogs in Punjab produced data that gave a trendline equation of y = -32.5 X + 158.4.  In Bangalore, the equation is y = - 49 X + 250.  In Jamshedpur, the equation is y = -43.6 X + 204.   Therefore, it is likely that one could develop a reasonably accurate estimate of the number of dogs in India using a range of human density values and estimates of human populations in different landscapes across India.  

However, the Indian pet dog population is currently growing very rapidly and so it may be necessary to undertake a few careful surveys of pet and street dog populations to update the above equations.