Page 17 - Marieke Poppe

1 General introduction

with a less severe negative energy balance (Sölkner and Fuchs, 1987; Dekkers et al.,
1998). Furthermore, data is becoming available on feed intake which is used for
selection for feed efficiency (Berry et al., 2014; Pryce et al., 2014; Martin et al.,
2021b). Automatic recording of e.g. body condition score is an emerging area of
study as well (e.g. Salau et al., 2014; Hansen et al., 2018; Song et al., 2019). These
new sources of data may offer an opportunity for genetically improving energy
balance (Martin et al., 2021a).
A third method, specifically for dairy cattle, was to select for more robust
conformation (Hamoen et al., 2009). A cow with robust conformation is a cow with
intermediate body condition score, body depth, rump width and chest width. The
robustness score has been applied in The Netherlands in the overall conformation
score, but was removed again due to lack of clarity of the trait for farmers
(Veeteelt, 2015).
The last proposed method to improve robustness was selection for lower
environmental sensitivity (Knap, 2005; Veerkamp et al., 2013). Two types of
environmental sensitivity exist: macro- and micro-environmental sensitivity (for a
review, see Iung et al. (2020)). Macro-environmental sensitivity refers to how
sensitive performance is to a measurable aspect of the environment, such as
ambient temperature (Berghof et al., 2019b). Macro-environmental sensitivity can
be improved by applying reaction norm models. A reaction norm model is a genetic
model where the trait of interest is regressed on an environmental descriptor, such
as temperature-humidity index. With this model, animals receive a breeding value
for the level and the slope of the trait. A steeper slope means that the animal is
more environmentally sensitive. Many studies have been performed on selection
for macro-environmental sensitivity (e.g. Calus et al., 2005; Bohmanova et al.,
2008; Mulder et al., 2013a; Rashidi et al., 2014; Herrero-Medrano et al., 2015;
Carabaño et al., 2017; Nguyen et al., 2017b). Nevertheless, to my knowledge such a
model is in practice only being applied for heat tolerance in dairy cattle in Australia
(Nguyen et al., 2017b).
Micro-environmental sensitivity refers to how sensitive performance is to
temporary changes within the environment, and is therefore similar to the
definition of resilience used in this thesis. Studies that focused on selection for
micro-environmental sensitivity used advanced statistical models on variability of
available traits, within family groups or within animals. The hypothesis for family
groups, is that family groups with high variability in performance within them, have
high micro-environmental sensitivity (or low uniformity). The hypothesis for
individual animals, is that a trait measured multiple times on the same animal is
more variable for animals with high than with low micro-environmental sensitivity.

15