Calculation of the sterile-wild male ratio
I'm confused by the explanations in Carrieri's report about how to estimate the sterile-wild male ratio depending on the circumstances.
-
"There are three alternative methods", he says (p. 15). The preferred one is a direct calculation with MRR studies. But he does not specify how to perform the estimate and he actually uses a different method in the MRR_template, even though it is certainty a MRR study.
From the FAO/OIE guidelines,
The sterile to wild ratio generally drops quickly within 200m from the release point [...] To obtain a good estimation of the competitiveness in the release area, it is thus necessary to release sterile males homogeneously [...] over the monitored area
I'm suspecting that a direct estimate requires a areal (line) release, although it is not said explicitly and the last sentence was in the context of egg fertility. So, I'm not sure. Please confirm.
If that was the case, the estimate would be directly the ratio between the total number of sterile males and the total number of wild males across all traps in the SIT area.
-
The second method, is through the sex-ratio and is the one exemplified in the MRR_template. This is based on the comparison the sex ratios in the control and SIT areas. However, in the MRR_template there is no control area.
Instead, the sex ratio in the control area uses the data from the SIT area excluding the sterile males. So it does not really requires a control area? Is this an approximation? Would it be better to use a real control area? Furthermore, since counts of sterile and wild males are used for computing the sex ratios, actually the final result is exactly the same as directly computing the ratio of sterile to wild male counts, which was not supposed to be applicable since the spatial distribution is not homogeneous. This looks as a very convoluted way of doing the same thing as in point 1.
-
The third estimation method is by using relative estimates of population. I didn't look into it but is uses the coefficients of a regression model. What I want to know is when this method should be used.
-
More conceptually, I don't even understand what is the target parameter. Since we know that the sterile-wild male ratio is something that decays with distance and with time since release, what exactly are we trying to estimate with this parameter?