Genetic parameters for dairy calf and replacement heifer wellness traits and their association with cow longevity and health indicators in Holstein cattle

High mortality and involuntary culling rates cause great economic losses to the worldwide dairy cattle industry. However, there is low emphasis on wellness traits in replacement animals (dairy calves and replacement heifers) during their development stages in modern dairy cattle breeding programs. Therefore, the main objectives of this study were to estimate ge-netic parameters of wellness traits in replacement cattle (replacement wellness traits) and obtain their genetic correlations with 12 cow health and longevity traits in the Chinese Holstein population. Seven replacement wellness traits were analyzed, including birth weight, survival from 3 to 60 d (Sur1), survival from 61 to 365 d (Sur2), survival from 366 d to the first calving (Sur3), calf diarrhea, calf pneumonia, and calf serum total protein (STP). Single and bivariate animal models were employed to estimate (co)variance components using the data from 189,980 Holstein cattle. The genetic correlations between replacement wellness traits and cow longevity, health traits were calculated by employing bivariate models, including 6 longevity traits and 6 health traits (clinical mastitis, metritis, ketosis, displaced abomasum, milk fever, and hoof health or hoof disease). The estimated heritabilities (± SE) were 0.335 (± 0.008), 0.088 (± 0.005), 0.166 (± 0.006), 0.102 (±0 .006), 0.048 (± 0.003), 0.063 (± 0.004), and 0.170 (± 0.019) for birth weight, Sur1, Sur2, Sur3, pneumonia, diarrhea, and STP, respectively. The majority of the genetic correlations among the 7 replacement wellness traits were negligible. The genetic correlations among Sur1, Sur2, and Sur3 ranged from 0.112 (Sur1 and Sur3) to 0.445 (Sur1 and Sur2) when fitting a linear model (estimates in the observed scale), and from 0.560 (Sur1 and Sur3) to 0.773 (Sur1 and Sur2) when fitting a threshold model (estimates in the liability scale). The genetic correlations between replacement wellness and cow longevity were low (absolute value lower than 0.30), but some of them were significantly different from zero. Compared with other replacement wellness traits, Sur3 and STP had relatively high genetic correlations with cow longevity. Replacement wellness traits are heritable and can be improved through direct genetic and genomic selection. The results from the current study will contribute for better balancing dairy cattle breeding goals to genetically improve dairy cattle wellness in the period from birth to first calving.


INTRODUCTION
Cow replacement costs represent one of the largest financial components in dairy cattle farms.For instance, a study in 44 dairy cattle operations from 13 Pennsylvania counties showed that the total cost for raising a heifer from birth until calving averaged around $1,808 (Heinrichs et al., 2013).Estimates of expenses associated with rearing replacement heifers can range from 15 to 20% of the total milk production costs (Heinrichs, 1993;Karszes, 1994).However, many potential replacement heifers do not reach their first lactation due to premature death or involuntary culling.In our previous study (Zhang et al., 2019), it was found that the replacement mortality-culling rates (including voluntary and involuntary culling) from d 3 to 60, d 61 to 365, and d 366 to first calving were 5.5, 7.4, and 8.7%, respectively (Zhang et al., 2019).In various dairy cattle populations around the world, calf diarrhea (DIA) and pneumonia (PNE) are highly prevalent diseases.For example, a

Genetic parameters for dairy calf and replacement heifer wellness traits and their association with cow longevity and health indicators in Holstein cattle
high DIA risk up to 29% was reported in New York (Virtala et al., 1996).In Chinese Holstein cattle, our previous study found that DIA and PNE caused the highest number of calf and heifer involuntary culling or mortality until the first calving (Wathes et al., 2008;Zhang et al., 2019).Even if the replacement calves survive and recover from the disease, their performance as mature cows is usually compromised in some way, such as greater age at first calving and reduced survival through the first and second parities (Rossini, 2004).Therefore, cow replacement wellness (traits that indicate the health and welfare of young animals that will become dairy cows) is not only a serious welfare issue, but also closely related to the lifetime profit of the animals and direct economic losses to the dairy farms (Meyer et al., 2001;Ortiz-Pelaez et al., 2008).Despite the increasing emphasis on health and longevity traits in mature cows (Miglior et al., 2017;Brito et al., 2021), less importance has been given to replacement wellness traits in dairy cattle breeding programs.
Generally, replacement survivability has been defined as binary traits corresponding to the survival during various life stages from birth to first calving.In Iranian (Forutan et al., 2015), UK and US (Pritchard et al., 2013), Danish (Fuerst-Waltl and Sørensen, 2010), and Dutch (Harbers et al., 2002) Holstein cattle; Austrian (Fuerst-Waltl and Fuerst, 2012) and US (Erf et al., 1990) Brown Swiss; Danish Jersey (Norberg et al., 2013); and Spanish local beef cattle (Asturian Valley breed; Goyache et al., 2003) populations, the heritability of replacement survival ranged from 0.01 to 0.14.In US Holstein (Gonzalez-Peña et al., 2019) and Jersey (Gonzalez-Peña et al., 2020) cattle, genomic evaluation on replacement survivability defined as a binary trait between 2 and 365 d of age has been performed with a reliability of 47.3 to 50.5%.More recently, alternative approaches for defining survival and longevity have also been proposed based on random regression models in beef cattle (Oliveira et al., 2020).
Passive immunity is important for calf health and survival, and serum total protein (STP) can be used as an indicator of passive transfer of immunity from cows to calves (Haagen et al., 2021a).In recent years, more Chinese dairy farms realized the importance of passive transfer of immunity and monitored the health management of calves using STP measured by refractometers.When the calf is born, colostrum with sufficient IgG content (colostrum quality) is essential for obtaining passive immunity through absorption of maternal antibodies from colostrum.The quality of the colostrum provided to calves should be included in the evaluation model of STP, and it is generally assessed in Chinese dairy farms based on the Brix value (Biel-mann et al., 2010) using a Brix meter.Although the heritability estimates of STP in Danish Red heifers at the age of 4 to 38 mo (0.09; Jensen and Christensen, 1975) and in Holstein cows (0.20;Cecchinato et al., 2018) have been reported, only one genetic analysis study was found for calf STP using large data sets, and the heritability of calf STP ranged from 0.06 (0.01) to 0.08 (0.02) (Haagen et al., 2021a).Birth weight (BIW) is an important trait, usually measured by the farmers, that indicates the growth and development of the calf during the fetal period.However, its genetic correlation with health and survivability of the calves after birth is still unclear.
The estimation of genetic parameters is essential for genetic and genomic selection, especially when developing novel breeding goals.However, genetic parameters for wellness traits in dairy calves and replacement heifers are still uncommon, and there is a lack of genetic correlation estimates among replacement wellness traits (survivability, disease resistance, and STP) and between wellness traits in replacement young animals and longevity and health indicators in mature cows.In this context, the main objectives of this study were: (1) to obtain heritability and genetic correlation estimates of 7 replacement wellness traits, including 3 survivability traits in different life stages, DIA, PNE, STP, and BIW; and (2) to estimate the genetic correlations between these replacement wellness traits and 12 indicators of cow longevity and health in Chinese Holstein cattle.The findings of this study will help breeders to refine their breeding goals for improving the production efficiency and animal welfare in dairy cattle populations, and provide a reference for further studies on the genetic background of wellness traits in dairy calves and replacement heifers.

Data
Phenotype for Replacement Wellness Traits.Phenotypic records of birth, calving, health, and culling or death variables were collected in 239,043 Holstein females born from 1999 to 2020 in a total of 31 herds located in Beijing (n = 22), Hebei (n = 2), Tianjin (n = 2), and Yunnan, Henan, Heilongjiang, Jilin, and Inner-Mongolia (one herd in each of these other locations).The individual records were downloaded from the farm management software.The STP was measured in 21,722 calves born from 2016 to 2020.The STP was tested by the producers at the farm using the blood sample collected in calves with 2 to 3 d of age.The cow colostrum was collected immediately after calving and the Brix value (Bielmann et al., 2010) was measured on-farm using a Brix meter.The Brix value of colostrum provided to the calves was obtained for each individual with the STP phenotype.
Seven replacement wellness traits were analyzed, including BIW, survival from 3 to 60 d (Sur1), survival from 61 to 365 d (Sur2), survival from 366 d to first calving (Sur3), DIA, PNE, and STP.Survival traits were defined as binary traits, in which a value of 0 was assigned to animals that left (including the voluntary and involuntary culling) the herds before the age threshold established and 1 to those that survived up to the next life stage.DIA and PNE were also defined as binary traits, with a value of 1 if the calf had the corresponding health issue at any time point during the first 120 d after birth and 0 otherwise, regardless of how many times the disease incidence or treatment was recorded.Diarrhea and PNE records were collected by the farm veterinarian.The individuals with no birth date or pedigree information were removed from further analyses.Furthermore, only BIW and STP records within the range of the mean ± 3 SD were kept in the data set.After data editing, the number of observations available for BIW, Sur1, Sur2, Sur3, DIA, PNE,and STP were 117,460,184,202,168,552,141,521,184,563,184,563,and 16,293,respectively.Phenotypes for Cow Longevity and Health Traits.Six cow longevity traits were defined in the same dairy population with replacement survivability traits, including the period from the first calving to the end of the first (Lon11), second (Lon12), third (Lon13), fourth (Lon14), and fifth parity (Lon15), and productive life (PL), all measured in days.For the cow longevity traits, the traits' definition and data set editing were the same as employed by Zhang et al. (2021).The number of animals for the 6 cow longevity traits ranged from 114,554 (Lon11) to 89,675 (Lon15), and the number of individuals with phenotypes of both replacement wellness and cow longevity trait ranged from 1,035 (STP and PL) to 114,554 (Sur1 and Lon11; Supplemental Table S1, https://figshare.com/articles/dataset/Additional_file_of_JDS_pa-per_21450/19640754; Zhang, 2022).
Six cow health traits were evaluated in this study, including clinical mastitis, metritis, ketosis, displaced abomasum, milk fever, and hoof disease.The cow health traits were defined in lactations 1, 2, and 3 as binary traits with a value of 1 indicating if a cow had at least one health problem at any time point of the corresponding lactation, and 0 otherwise.The number of individuals with phenotypic records for health traits measured in lactation 1, 2, and 3 was 118,022, 88,216, and 54,936, respectively.The number of individuals with phenotypic records for both replacement wellness and cow longevity traits ranged from 287 (STP and cow health traits in the third lactation) to 99,294 (Sur1 and cow health traits in the first lactation; Supplemental Table S1).
Pedigree.The pedigree was provided by the Dairy Association of China (Beijing, China).Animals with phenotypic records were traced back up to 16 generations, and 85.23 and 60.15% of the animals with phenotypic records had sire and grandsire recorded, respectively.There were no animals with both parents unknown.The final pedigree included 289,016 females and 6,010 males born from 1933 to 2020.

Statistical Analyses
Phenotypic Analyses for Replacement Wellness Traits.The GLM procedure of the SAS software (version 9.1; SAS Institute, 2004) was used to test the significance of nongenetic effects on BIW and STP, including herd-birth year and birth season.A binomial logistic regression was used to analyze the effect of birth season on risk of PNE and DIA using the LOGISTIC procedure implemented in the SAS software (version 9.1;SAS Institute, 2004).
Genetic Analyses for Replacement Wellness Traits.Variance and co-variance components for 7 replacement wellness traits were estimated using the average information restricted maximum likelihood algorithm implemented in the DMU software (Madsen et al., 2006).Heritabilities were estimated using singletrait linear animal models (linear model) for 7 replacement wellness traits.For the binary traits Sur1, Sur2, Sur3, PNE, and DIA, we also used generalized linear mixed models linked to the binary trait with a logit (logit model) and probit (probit model) link function to estimate variance components.The residual variance in probit and logit model was set to 1 (Su et al., 2008).
In this study, the genetic correlations among the replacement wellness traits were obtained based on bivariate linear animal models and bivariate threshold (or linear threshold) animal model.For binary trait, only the probit link function was used in evaluation model to estimate genetic correlation.The model fitted for BIW, Sur1, Sur2, Sur3, DIA, and PNE was whereas the model for STP was where y is the phenotypic records for replacement wellness traits; hby the fixed effect of herd-birth year (in-cluding 306, 363, 363, 363, 363, 356, and 43 levels for BIW, DIA, PNE, Sur1, Sur2, Sur3, and STP, respectively); bs is the fixed effect of birth season, including spring (March to May), summer (June to August), fall (September to November), and winter (December to February), as referred in the study of Yin et al. (2019); a is the additive genetic random effect; brix is the Brix value for colostrum drunk by the calves, which reveal the quality of colostrum influencing passive immunity transfer; β is the fixed regression coefficient, and e is the random residual effects.It was assumed that ( ) where A is the matrix of additive genetic relationships constructed based on pedigree information, σ a 2 is the additive genetic variance, I is an identity matrix, and σ e 2 is the residual variance.
Genetic Correlations with Cow Longevity and Health Traits.Bivariate linear animal models and linear threshold models were fitted to estimate the genetic correlations of calf and heifer wellness traits with cow longevity and health traits, respectively.For the binary traits, only the probit link function was used in the model to estimate the genetic correlations.The effects included in the model for replacement wellness traits were the same described for model 1 (for BIW, Sur1, Sur2, Sur3, DIA, and PNE) and model 2 (for STP).The effects of age at first calving, herd-birth year, birth year-season, and random additive genetic effects were included in the evaluation model for cow longevity traits (including PL, Lon11, Lon12, Lon13, Lon14, and Lon15), which are detailed in Zhang et al. (2021).The model for the cow health traits (including clinical mastitis, metritis, ketosis, displaced abomasum, milk fever, and hoof disease in the first, second, and third parity) was where y is the phenotypes for the cow health traits; hcy is the fixed effect of herd-calving year (the combination of herd and calving year); cs is the fixed effect of calving season (including 4 levels; its definition was the same as the birth season in model 1); and a and e are the random additive genetic and random residual effects, respectively, as defined in models 1 and 2.

Descriptive Statistics
The descriptive statistics for BIW and STP are presented in Table 1.The average BIW was 39.00 (± 4.33) kg and ranged from 24 to 53 kg.The average STP was 7.44 (± 1.68) g/dL with a large coefficient of variation (CV; 22.58%).The morbidity of calves due to PNE and DIA was 3.77 and 12.01% in the first 120 d after birth, respectively.The frequency distribution of the calves age when first suffering from PNE and DIA is shown in Figure 1.The average age of occurrence for the first PNE and DIA event during the first 120 d after birth was 39.15 (± 33.92) d and 15.93 (± 21.80) d, respectively.The incidence of PNE was the highest after birth and then gradually decreased as calves got older.For DIA, the incidence was the highest right after birth, and a second peak was observed when the calves were one week old.Compared with PNE, most of the DIA events occurred within the first month after birth, and only 13.43% of the calves had a recorded DIA event between the second and fourth month of age, whereas 50.31% of the calves had PNE in this same period.
The descriptive statistics for Sur1, Sur2, and Sur3 are reported in Zhang et al. (2019).

Effects of Birth Season on Replacement Wellness Traits
In this study, birth season had a significant effect (P < 0.05) on both BIW and STP.The least squares means estimates of various levels and multiple comparisons based on Bonferroni t corrected are presented in Table 2.The calves born in the winter had the largest BIW, whereas the smallest ones were born in the summer.STP was highest in the spring and lowest in the fall season with a difference of 0.24 g/dL.
According to the Wald test (Chi-squared), birth season significantly (P < 0.01) influenced the risk of PNE and DIA during the first 120 d in Holstein calves.The results of the binomial logistic regression on PNE and DIA in Holstein calves are presented in Table 3.Among the 4 birth seasons, the risk of calf PNE born in spring was the lowest, and the risk of those calves born in the summer, fall, and winter was 1.13, 1.57, and 1.28 times greater than those born in the spring season.Calves born in the winter had the lowest risk of DIA.The risk of DIA ranged from 1.24 (winter) to 1.34 (summer) times greater than those born in the winter.

Genetic Parameters for Replacement Wellness Traits
The estimates of the variance components and heritabilities using the single-trait model for the 7 replacement wellness traits are shown in Table 4.Among the 7 wellness traits, BIW had the highest heritability (0.335).The survival traits were lowly to moderately heritable with heritability estimates ranging from 0.088 (Sur1) to 0.166 (Sur2) using linear model.Based on the linear model, PNE and DIA were lowly heritable (0.048-0.063) and the standard errors for all heritability estimates were low (<0.010,except for STP), suggesting that all the estimates were accurately obtained.In Different letter superscripts among different levels means significant difference (P < 0.05), and the same letter superscripts means no significant difference (P > 0.05).comparison to the linear models, the threshold models yielded higher heritability estimates for PNE, DIA, and Sur1, similar estimates for Sur2 (except for the probit model); and lower estimates for Sur3.
The genetic correlation among the 7 replacement wellness traits are presented in Table 5.Except for Sur1 and STP, all genetic correlations between BIW with the other wellness traits (Sur2, Sur3, PNE, and DIA) were negligible.Birth weight had a low genetic correlation with Sur1 (0.169 when fitting a linear model and 0.236 for the linear threshold model).Low to moderate genetic correlations were found among Sur1, Sur2, and Sur3, which ranged from 0.112 (Sur1 and Sur3) to 0.445 (Sur1 and Sur2) by linear model, and from 0.560 (Sur1 and Sur3) to 0.773 (Sur1 and Sur2) based on the threshold model.Diarrhea is lowly genetically correlated (0.154 or 0.496, according to the different model) with PNE and calves more susceptible to DIA tend to have a higher risk of PNE.Furthermore, low genetic correlations were found between STP and other replacement wellness traits (absolute value lower than 0.20).

Genetic Correlation Between Replacement Wellness and Cow Longevity and Health
The genetic correlations between calf and heifer wellness traits and cow longevity traits based on a bivariate linear animal model and linear threshold animal model are presented in Tables 6 and 7, respectively.Low and positive genetic correlations were observed between BIW and cow longevity traits, which ranged from 0.048 (Lon11) to 0.150 (PL) with a significant difference from zero.Sur3 had the highest genetic correlations with cow longevity traits, ranging from 0.017 (PL) to 0.277 (Lon12) based on a linear model, and from 0.088 (PL) to 0.413 (Lon12) based on a linear threshold model.Low genetic correlations were observed for PNE (close to zero) and DIA (0.117-0.143 by linear model, and 0.154-0.184by linear threshold model) with cow longevity traits.Among the 7 calf and heifer wellness traits, STP had relatively high genetic correlation with cow longevity, which ranged from 0.151 (Lon11) to 0.291 (PL), indicating that calves with higher STP will have a longer PL.For Sur1, Sur2, Sur3, PNE, and DIA, the For linear traits BIW and STP, genetic correlation was estimated using bivariate linear animal model.
estimates of genetic correlation based on a linear model were similar to those from the linear threshold models, suggesting that linear models can also be used for binary traits.The genetic correlations between calf and heifer wellness and cow health traits are presented in Supplemental Table S2 (https://figshare.com/articles/dataset/Additional_file_of_JDS_paper_21450/19640754).Most genetic correlations were not well estimated due to the small sample size and were not significantly different than zero.

DISCUSSION
More recently, replacement wellness traits are being gradually included in dairy cattle selection indices around the world.For example, young stock survival (replacement survivability) traits have a weight of 3.8% (Jersey) to 6.9% (Red cattle) in the Nordic Total Merit (www .nordicebv.info/ ) index for Nordic Holstein, Jersey, and Red cattle, which are higher than that of cow longevity (2.2% for Red cattle to 3.4% for Jersey).Since 2018, calf health traits have been introduced into the RZG (Relativ-Zuchtwert Gesamt, German Total Merit index) index for German Holstein, Jersey, Red cattle, and dual-purpose cattle, including calf survival from d 3 to 458 (www .vit.de/).In the US dairy cattle industry, genetic and genomic evaluations for heifer livability (presenting the livability percentage of an animal's female offspring from 2 d up to 18 mo after birth) have been performed for Holstein and Jersey since December 2020 (www .uscdcb.com/), which has a weight of 0.8% in the Lifetime Net Merit selection index.To our knowledge, this is the first study to explore the genetic parameters of replacement wellness traits using large-scale data sets in Chinese Holstein cattle.
In our previous study (Zhang et al., 2019), we found that the combined mortality rate of dairy calves and replacement heifers in Chinese Holstein cattle was 21.2% and diseases related to digestive (e.g., DIA), respiratory (e.g., PNE), and circulatory systems, and   reproductive disorders (infertility based on nonreturn rate) were the main death or culling reasons.In this study, the population parameters (descriptive statistics) of DIA and PNE were further investigated, in which 12.01% of calves were recorded as having at least one occurrence of DIA.In Holstein calves, a wide range of DIA incidence from 10% (Svensson et al., 2003) to 29% (Virtala et al., 1996) has been reported during the first 3 mo of the calf life.Diarrhea can be caused by both infectious agents and noninfectious factors such as the housing environment, colostrum intake, feed management, breed, and cow-level factors (Vinet et al., 2018;Condon et al., 2021).Therefore, large variations among populations are expected.The incidence risk of PNE in the Chinese Holstein population was lower than the reported value (6.4% within the first 5 mo) in Irish beef cattle (Condon et al., 2021).In the current study, the decreased trend of the incidence risk for PNE and DIA as the calf age agrees with the findings for Charolais beef cattle (Vinet et al., 2018).The first month after birth is crucial for calves due to the high incidence of health issues.The average STP (7.44 g/dL) in this study was slightly higher than the reported mean range (5.06-7.16g/dL) in US (Donovan et al., 1986;Paré et al., 1993;Villarroel et al., 2013;Haagen et al., 2021a), Canadian (Wilm et al., 2018), and Italian (Agnes et al., 1993) Holstein cattle.Furthermore, the CV (22.58%) of STP in the current study was also much larger than the literature values (around 10%; Tóthová et al., 2016).The average BIW (39.00 kg) in this study is within the reported mean range (38.8-43.4Kg) of previous studies in Holstein populations (Coffey et al., 2006;Koçak et al., 2007;Linden et al., 2009).From 2005 to 2020, a downtrend (from 40.58 to 37.84 kg) was observed for BIW in Holstein population involved in the present study (data not shown), which may be an indirect response to selection for calving ease.In recent years, genetic analyses of calf health traits have gradually become an important topic in dairy cattle breeding, but genetic parameters of PNE (or respiratory disease, RESP) and DIA have been estimated in few dairy and beef cattle populations.As there are limited records for calf health and PNE is usually the most frequent respiratory disease, all respiratory events (including PNE) are usually combined into RESP in some studies.Both DIA and PNE have low heritability in Chinese Holstein cattle.Previous studies reported heritabilities for DIA and PNE (or RESP) of less than 0.10 (0.012-0.084 for DIA and 0.042-0.090for PNE or RESP) in US Jersey (Gonzalez-Peña et al., 2020) and Holstein (Henderson et al., 2011;Mc-Corquodale et al., 2013;Gonzalez-Peña et al., 2019;Haagen et al., 2021b), German Holstein (Mahmoud et al., 2017), and French Charolais beef cattle (Vinet et al., 2018), which are in agreement with the findings of the present study.
Since the 1970s, genetic parameters of replacement survival traits have been estimated in various dairy cattle populations over time (Bar-Anan et al., 1976;Erf et al., 1990;Hansen et al., 2003;Haagen et al., 2021b;Neupane et al., 2021;Weller et al., 2021), with a large variability in the estimates, ranging from 0.001 in Danish Holstein (Hansen et al., 2003) to 0.120 in Indian Jersey crossbred animals (Pathak et al., 2018).The heritability estimates for Sur1 and Sur3 are within the literature range (0.001-0.120), whereas Sur2 has a higher estimate than the other 2 traits.In a Spanish local beef cattle population (Goyache et al., 2003), a heritability estimate of 0.142 was reported for calf survival at weaning, which is similar to the estimate for Sur2 observed in our study.The differences on the population genetic background, statistical models, trait definition based on life stage, and frequencies of dead or culled animals across studies contribute to the variation observed in the heritability estimates for replacement survival indicators.
In the current study, a moderate heritability estimate (0.17) was found for STP in Chinese Holsteins, whereas lower estimates (0.02-0.08) were previously reported in Danish Red cattle (Jensen and Christensen, 1975) and US Holstein cattle (Donovan et al., 1986;Haagen et al., 2021a).With exception of the study in the US Holstein population (Haagen et al., 2021a), small data sets were used for the other studies (758-2,105 calves).Furthermore, the variation in animal age when STP was measured might have influenced the heritability estimates in these studies considering the effect of age on STP (Agnes et al., 1993;Villarroel et al., 2013;Tóthová et al., 2016).For instance, the age of the animals evaluated were 4 to 38 mo, 1 to 3 d, and 1 to 12 d in the studies by Jensen and Christensen (1975), Haagen et al. (2021a), andCordero-Solorzano et al. (2021), respectively.A moderate heritability estimate (0.20) was found for cow STP in Italian Holstein cattle (Cecchinato et al., 2018), which is similar to the estimate for calf STP from this study.In the present study, the estimated heritability of BIW was high (0.335 ± 0.008) but within the range (0.12-0.53) of literature reports in Turkish (Koçak et al., 2007), British (Coffey et al., 2006), and North American (Olson et al., 2009) Holstein cattle.
Low to moderate genetic correlations were observed among Sur1, Sur2, and Sur3, which agrees with the literature (Forutan et al., 2015;Pathak et al., 2018).There are differences on the replacement survival trait definitions across studies, and, as expected, higher genetic correlations are observed between survival indica-Zhang et al.: GENETIC ANALYSES OF CALF AND HEIFER WELLNESS tors based on similar or adjacent life stages (Harbers et al., 2002;Hansen et al., 2003;Fuerst-Waltl and Sørensen, 2010;Pathak et al., 2018).The low genetic correlation between Sur1 and Sur3 indicates that different gene sets influence replacement mortality in early and late age periods.Furthermore, Sur1, Sur2, and Sur3 should be combined into a survival subindex to improve replacement survivability.
To our best knowledge, only one study reported the genetic correlation between STP and calf survivability up to 365 d (0.19 ± 0.23; Haagen et al., 2021a), and low genetic correlations ranging from −0.186 with Sur3 to 0.170 with DIA were found in the current study.Although few studies indicated that lower STP is associated with greater incidence of calf DIA and PNE (or RESP;McCorquodale et al., 2013), this association with health traits needs to be further explored in future studies.Furthermore, it was reported that both high and low STP at birth was associated with an onset of DIA at an earlier age for calves (Paré et al., 1993).In US Holstein and Jersey populations, there was no significant difference in STP (P = 0.924) between healthy calves and those that developed scours, PNE, or both (Villarroel et al., 2013).The STP can be used to monitor if calves are receiving enough colostrum to achieve sufficient transfer of immunity.However, it might not be the most efficient indicator of calf health for breeding purposes.
The low genetic correlations found between PNE and DIA, health (PNE and DIA), and survival traits (Sur1, Sur2, and Sur3) are similar to the approximate genetic correlations from 0.226 (DIA and RESP) to 0.293 (calf livability and DIA) reported by Gonzalez-Peña et al. (2020) in US Jersey cattle.Furthermore, moderate approximate genetic correlations ranging from 0.464 (DIA and calf livability) to 0.697 (RESP and calf livability) were observed between DIA, RESP, and calf livability in US Holstein cattle (Gonzalez-Peña et al., 2019).
The method proposed by Calo et al. (1973) has been widely used to obtain approximate genetic correlations between replacement wellness traits and cow traits in various populations (Gonzalez-Peña et al., 2019, 2020;Haagen et al., 2021a;Neupane et al., 2021;Weller et al., 2021).However, in this study, direct genetic correlations were calculated, as they are more appropriate than approximate values.In general, replacement wellness, cow longevity, and health traits are lowly genetically correlated, which agrees with previous literature reports based on approximate genetic correlations in US Jersey (Gonzalez-Peña et al., 2020) and Holstein (Gonzalez-Peña et al., 2019) cattle.Based on these results, it is important to implement direct genetic selection in dairy replacement cattle, and cow traits as indirect genetic responses are expected to be small.For binary traits Sur1, Sur2, Sur3, PNE, and DIA, the genetic parameters were estimated using both linear and threshold models in this study, and the differences between estimates from different models, which was is consistent with the literature (Forutan et al., 2015;Malchiodi et al., 2017).Although the threshold model was appropriate for fitting binary trait, considering the high EBV rank correlations between linear and threshold as reported by Fuerst-Waltl and Fuerst (2010), we suggest that linear model can be used for genetic evaluation of above binary traits.

CONCLUSIONS
Except for BIW, replacement wellness traits are lowly to moderately heritable, with estimates ranging from 0.048 (PNE) to 0.165 (Sur2) when fitting linear models, and from 0.056 to 0.070 (Sur2) to 0.126 to 0.168 (Sur3) when fitting threshold models.Due to the low genetic correlations with cow longevity, replacement wellness traits should be incorporated in dairy cattle breeding programs for directly improving animal welfare and the profitability of dairy cattle farms.Based on the genetic parameters and effects on production system of the 7 replacement wellness traits evaluated here, we suggest the inclusion of replacement survival from 3 to 60 d (Sur1), survival from 61 to 365 d (Sur2), survival from 366 d to the first calving (Sur3), calf diarrhea (DIA) and PNE in dairy cattle genetic selection schemes.
2 PL = productive life; Lon11 = number of days from first calving to the end of the first parity; Lon12 = number of days from first calving to the end of the second parity; Lon13 = number of days from the first calving to the end of the third parity; Lon14 = number of days from the first calving to the end of the fourth parity; Lon15 = number of days from the first calving to the end of the fifth parity.BIW = birth weight; Sur1 = calf survival from 3 to 60 d; Sur2 = calf and heifer survival from 61 to 365 d; Sur3 = heifer survival from 366 d to first calving; PNE = calf pneumonia during the first 120 d after birth; DIA = calf diarrhea during the first 120 d after birth; STP = calf serum total protein.*Indicates significant genetic correlations.

Table 1 .
Zhang et al.: GENETIC ANALYSES OF CALF AND HEIFER WELLNESS Descriptive statistics of birth weight (BIW) and serum total protein (STP) of Holstein calves Figure 1.The age distribution of calves when they first suffered from pneumonia or diarrhea within the first 120 d after birth.

Table 2 .
Zhang et al.: GENETIC ANALYSES OF CALF AND HEIFER WELLNESS The effect of birth season on calf birth weight (BIW, kg) and serum total protein (STP, g/dL) in Holstein cattle

Table 3 .
The effect of birth season on the risk of calf pneumonia (PNE) and diarrhea (DIA) during the first 120 d after birth in Holstein cattle

Table 4 .
Zhang et al.:GENETIC ANALYSES OF CALF AND HEIFER WELLNESS Estimates of variance components and heritability (h 2 ) of replacement wellness traits in Holstein cattle

Table 5 .
Genetic correlations among replacement wellness traits based on bivariate linear animal models (below the diagonal) and bivariate threshold (or linear threshold) animal model (above the diagonal) in Holstein cattle d to first calving; PNE = calf pneumonia during the first 120 d after birth; DIA = calf diarrhea during the first 120 d after birth; STP = calf serum total protein.2

Table 6 .
Zhang et al.:GENETIC ANALYSES OF CALF AND HEIFER WELLNESS Genetic correlations between calf and heifer wellness traits and cow longevity traits based on bivariate linear animal models in Holstein cattle 1,2 Genetic correlations were considered significantly different from zero if they deviated by more than 1.645 SE units from zero, where the value 1.645 corresponds to a one-sided 5% cut-off point of the normal distribution. 1

Table 7 .
Genetic correlations between calf and heifer wellness traits and cow longevity traits based on linear threshold animal models in Holstein cattle1,2