A revised calculation method for correcting population density in the field: a case of Plutella xylostella (Lepidoptera: Plutellidae)

In this paper, we introduce a revised model to estimate the corrected initial number of each life stage in field life tables of insect pest populations, with a case example of the diamondback moth, Plutella xylostella L. (Lepidoptera: Plutellidae). To validate our revised model, we used life table data of P. xylostella from 9 field surveys conducted by the corresponding author in 1996. In these surveys, P. xylostella infestations had been monitored in fields of flowering Chinese cabbage, long white radish, butter cabbage, Shanghai pak choi, and Chinese kale to investigate the impact of different pest management methods on the pest populations. In the past, the numbers of individuals per life stage on any given sampling date were calculated based on sampling data of that particular date and an average temperature determined for the entire sampling period. In several cases, this approach produced unrealistic survival rates (above 100%) for certain life stages in the resulting life tables. This problem continued even after the model was adjusted by calculating the duration of each life stage based on the actual temperature measured on each sampling date (instead of using one average temperature for the entire period). With temperature being an important factor that (i) affects the development time of insect life stages and (ii) can be determined easily during field surveys, we previously hypothesized that including corrected initial numbers of each life stage on each sampling date based on an average of the temperature between the given sampling date and its preceding sampling date will result in realistic and precise life tables. Furthermore, to estimate accurate survival rates, we here hypothesized that (i) it is important to adjust and correct the numbers of life stages on a given sampling date by including the numbers of preceding life stages from a preceding sampling date (not from the given sampling date) in the model; and (ii) the development time of the preceding life stage will determine which sampling date needs to be included in the calculation. By re-constructing and comparing 9 life tables of P. xylostella populations according to the previous and the revised new model, we confirmed these hypotheses. The revised model will allow a precise and realistic evaluation of control efforts against diamondback moth and other insect pest infestations in agriculture.