Do Feed Plants Provide Sufficient Sodium, Calcium and Magnesium to Sika Deer in Japan? An Analysis Using Global Plant Trait Data

Simple Summary Selective culling of female deer is effective in reducing overabundant sika deer (Cervus nippon) populations. Sodium (Na), calcium (Ca) and magnesium (Mg) can act as an attractant to selectively cull female sika deer due to the differences in mineral requirements. Here, using global plant trait data and a published list of sika deer feed plants in Japan, we estimated whether feed plants provide sika deer sufficient Na, Ca and Mg, and compared the results between males and females. Our analysis showed that sufficient Na and Ca were not always provided, especially when intakes were low and deer were large. Na deficiency was more intense for lactating females than males, whereas Ca deficiency was more intense for males. We suggest that Na and Ca could be useful for developing effective culling methods. Especially, Na could be useful for selective culling of female sika deer during the lactating period. Abstract Deficient minerals in overabundant populations could act as an attractant to cull sika deer (Cervus nippon). Because selective culling of female deer is reported to be effective in reducing sika deer populations, it is particularly important to clarify the differences in mineral requirements between males and females. Here, using global plant trait data and a published list of sika deer feed plants in Japan, we estimated whether feed plants provide sika deer sufficient sodium (Na), calcium (Ca) and magnesium (Mg), and compared the results between males and females. An analysis of 191 feed plant species suggested that feed plants can provide sufficient Mg, whereas sufficient Na and Ca is not always provided, especially when the intake is small or the deer large. Na deficiency was more intense for lactating females than males, suggesting that Na can be an effective attractant for selectively culling female deer. In summary, this study demonstrated that sika deer in Japan might require extra Na and Ca sources in addition to feed plants, and therefore these minerals could be useful for developing effective culling methods.


Introduction
Ungulate overpopulation is an important issue in the northern hemisphere [1], including northern America [2] and Europe [3]. Ungulate populations are regulated by bottom-up, top-down and abiotic factors [4]. A lack of certain nutrients/elements may limit ungulate population growth, especially at a high population density.
Sodium (Na), calcium (Ca) and magnesium (Mg) may be limiting elements for ungulates at a high population density because livestock are often deficient in these elements, especially Na, which is supplemented artificially [5]. A mineral deficiency can also occur in nature based on the fact that various animals use salt licks [6][7][8]. Geophagy (i.e., eating soil) is common in ungulates [9][10][11], probably for mineral supplementation. 2 of 10 Various species were reported to lick natural springs, presumably for obtaining minerals [12]. Ceacero et al. (2014) [13] reported that Iberian red deer (Cervus elaphus) feed on seaweeds, possibly to acquire minerals (but see Ceacero et al., 2015 [14]). The use of salt licks has often been attributed to a requirement for Na [8,9,15], which is important for maintaining osmotic pressure [16]. However, it may also be due to a requirement for other minerals, such as Ca, which is a major component of bones [17], and Mg, which has important roles in enzyme activity and bone development [18]. Several studies have argued that natural salt licks are important sources of Ca and Mg [19].
Sika deer (Cervus nippon) are native to eastern Asia and have been introduced into many parts of the world, such as Europe, North America and New Zealand [20]. In some areas, the deer populations have increased enough that they cause serious browsing damage to natural and anthropogenic environments [20,21]. In Japan, especially, the sika deer population has increased dramatically in recent decades [22]. According to the Ministry of the Environment (2015), the distribution of sika deer expanded more than 2.5 times from 1978 to 2015 [23]. Accordingly, sika deer are being culled in many parts of Japan to reduce overabundant populations, and the development of effective culling methods is desirable. Since this substantial increase in the sika deer population and distribution and subsequent overbrowsing could lead to mineral deficiencies in sika deer, we considered that the deficient minerals in overabundant populations act as an attractant to cull deer. Indeed, sika deer also lick salt [24]. Furthermore, it is particularly important to clarify the differences in mineral deficiency between males and females because selective culling of female deer is reported to be effective in reducing sika deer populations [25,26]. However, no study has evaluated whether sika deer in Japan obtain sufficient minerals, such as Na, Ca and Mg, through feed plants by comparing the minerals provided by feed plants and the mineral requirements of sika deer. Mineral deficiencies between male and female deer have also not been compared.
In this study, we evaluated whether feed plants can provide sika deer sufficient Na, Ca and Mg by comparing the minerals provided by feed plants, calculated from leaf mineral contents and mineral requirements of sika deer, simulated by several scenarios of daily dry matter intake (DMI), body weight (BW) and sex. Since the feed plant species composition of wild sika deer in their natural state is currently undeterminable, we investigated whether each plant species could provide sufficient minerals to sika deer when sika deer ingested only that plant species. It should be noted that this approach may be misleading because this approach did not consider the feed plant species composition of wild sika deer and therefore is unable to accurately estimate the amount of minerals provided by feed plants. However, this approach is valuable to roughly estimate the amount of mineral deficiency in wild sika deer under natural conditions, which is the first attempt in Japan.

Dataset
A dataset of leaf nutrient contents of sika deer feed plants in Japan was constructed using plant trait data from the TRY Plant Trait Database [27] and a list of feed plant species of sika deer in Japan [28]. From TRY, we extracted the leaf Na, Ca and Mg contents per leaf dry mass (Trait IDs 260, 252 and 257, respectively) for plant species previously documented as sika deer feed in Japan [28]. In other words, plants not recorded as feed plants, including toxic plants [14], were not included in the analysis. The final dataset included 3179 values for 191 plant species (Table S1) from five original publications [29][30][31][32][33]. The leaf mineral content data were averaged for each species, and the averaged values were used for further analysis.

Analysis
Using the leaf mineral concentration data for sika deer feed plants in Japan, we simulated whether each plant species could provide sufficient minerals for sika deer. We considered four sika deer body weight (BW) scenarios based on the BW range of sika Animals 2023, 13, 1044 3 of 10 deer in Japan (i.e., 30, 60, 90 and 120 kg) and four daily dry matter intake (DMI) scenarios considering a wider range than reported from feeding experiments (i.e., 0.5, 1.5, 3 and 4.5 kg) [34][35][36]. Unrealistic scenarios (BW 30 kg with DMI 3 or 4.5 kg; BW 60 kg with DMI 4.5 kg; and BW 120 kg with DMI 0.5 kg) were excluded from our analysis. Mineral intakes provided by leaves, which were calculated by multiplying DMI and leaf mineral concentration, were compared with the mineral requirement of sika deer calculated as follows. The Na maintenance requirement was calculated based on the following equation for Cervus [17]: where Na_req_maintain is the Na requirement for maintenance and BW is body weight (kg). The Na requirement for males was calculated as Na_req_maintain plus the requirement for antlers, assuming that antler growth in sika deer requires 150 days, as follows [17]: The equation for determining the Na requirement of females during lactation [17] was modified for sika deer as follows: where Na_milk_sika is the Na content in sika deer milk (1.00 g L −1 , [37]). We also calculated the Na requirement of females during the non-lactating period (equal to Na_req_maintain) by omitting the term expressing the Na requirement for milk production from the equation. The Na requirement of females during the gestation period is shown in Figure S1. The above equations do not include the Na requirements for body growth, assuming that body growth is limited by a lack of feed plants. This is because we set conservative thresholds to avoid overestimating the mineral deficiency. The maintenance Mg requirement (Mg_req_maintain) was calculated based on the following equation for goats [17], because equations for goats were recommended to calculate values for cervids (equations for cervids were not available) [17]: Without considering Mg requirement for body growth, the Mg requirement for males was assumed to be equal to Mg_req_maintain [17]. The Mg requirement for lactating females was calculated using equations prepared for goats as recommended [17], but modified as follows: where Mg_milk_sika is the Mg content of sika deer milk (0.0819 g L −1 [37]). The Mg requirement of females during the non-lactating period was considered equal to the Mg requirement of males. The Mg requirement of females during the gestation period was shown in Figure S2. The maintenance Ca requirement (Ca_req_maintain) was calculated based on the following equation for Cervus [17]: The Ca requirement for males was calculated as Ca_req_maintain plus the Ca requirement for antlers as [17]: Animals 2023, 13, 1044 4 of 10 The Ca requirement for lactating females was calculated as Ca_req_maintain plus the Ca requirement for lactation [17]. We modified the equation by replacing the milk Ca concentration of the Cervus with that of sika deer [37]: where Ca_milk_sika is the Ca content of sika deer milk (1.62 g L −1 [37]). The Ca requirements for body growth were also not included in the same manner as Na and Mg. The Ca requirement of females during the non-lactating period (equal to Ca_req_maintain) was also calculated by omitting the term expressing the Ca requirement for milk production from the equation. The Ca requirement of females during the gestation period was shown in Figure S3.

Results
The sika deer feed plant dataset used for this analysis showed large variance ( Figure 1). The average leaf contents of Na, Ca and Mg were 0.47 ± 0.82, 15.3 ± 8.4 and 3.5 ± 1.8 (g element kg dry matter −1 ± standard deviation), respectively. Ca_req_male = Ca_req_maintain + (0.05 × BW)/0.39 (g day −1 ) The Ca requirement for lactating females was calculated as Ca_req_main Ca requirement for lactation [17]. We modified the equation by replacing the centration of the Cervus with that of sika deer [37]: where Ca_milk_sika is the Ca content of sika deer milk (1.62 g L −1 [37]). Th ments for body growth were also not included in the same manner as Na an requirement of females during the non-lactating period (equal to Ca_req_m also calculated by omitting the term expressing the Ca requirement for mil from the equation. The Ca requirement of females during the gestation perio in Figure S3.

Results
The sika deer feed plant dataset used for this analysis showed large var 1). The average leaf contents of Na, Ca and Mg were 0.47 ± 0.82, 15.3 ± 8.4 an element kg dry matter −1 ± standard deviation), respectively. Our analysis showed that the Na requirements were rarely achieved by especially under the scenarios with a smaller DMI and larger BW (Figure 2 the 12 scenarios, less than 50% of the feed plant species provided sufficien 2a,c,d,f-h,j-l). The results also indicated that a Na deficiency would be mo lactating females than males (Figure 2), as the Na requirement for females w double that of males ( Figure 2). Our finding that Na requirements often exce sion by feed plants contrasted to that for Mg; the leaf Mg contents were hi quired in most of the feed plants in our analysis, except for the scenarios w tions of large BW and small DMI (Figure 3c,f), indicating that Mg deficiency The fulfillment of the Ca requirement depended largely on the scenario (Figu DMI resulted in insufficient Ca provision in a large portion of the feed plan pecially for greater BWs (Figure 4a,c,f,g,j), whereas Ca requirements were m portion of the feed plant species when the DMI was ≥3 kg (Figure 4). The Ca Our analysis showed that the Na requirements were rarely achieved by feed plants, especially under the scenarios with a smaller DMI and larger BW (Figure 2). In most of the 12 scenarios, less than 50% of the feed plant species provided sufficient Na (Figure 2a,c,d,f-h,j-l). The results also indicated that a Na deficiency would be more intense for lactating females than males (Figure 2), as the Na requirement for females was more than double that of males ( Figure 2). Our finding that Na requirements often exceed the provision by feed plants contrasted to that for Mg; the leaf Mg contents were higher than required in most of the feed plants in our analysis, except for the scenarios with combinations of large BW and small DMI (Figure 3c,f), indicating that Mg deficiency is less likely. The fulfillment of the Ca requirement depended largely on the scenario (Figure 4). A lower DMI resulted in insufficient Ca provision in a large portion of the feed plant species, especially for greater BWs (Figure 4a,c,f,g,j), whereas Ca requirements were met in a large portion of the feed plant species when the DMI was ≥3 kg (Figure 4). The Ca requirement for males was larger than that of females (Figure 4). Dashed line indicates Na requirement of females during the non-lactating period, which is equal to Na maintenance requirement for both males and females. In the figures, the percentage is the proportion of feed plant species that meet the Na requirements.

Discussion
Our results suggest that feed plants in Japan are unlikely to provide sufficient Na, which agrees with the traditional view that Na is the main deficient mineral leading to the use of salt licks [8,9,15]. By comparing several salt solutions as attractants experimentally, Fraser and Reardon (1980) demonstrated that Na, but not K, Ca or Mg, attracted moose (Alces alces) and white-tailed deer (Odocoileus virginianus), indicating that Na was the primary attractant at their study site [6]. Sika deer were also reported to be attracted by Na [15,24,38]. The larger Na deficiency for lactating females than males (Figure 2) is consistent with reports that female sika deer were attracted by salt, especially during pregnancy or lactation [15,24].
This study suggests the importance of Ca as a potentially limiting element of sika deer in Japan. Although natural salt licks have been attributed to Na acquisition, several authors have reported that Ca could also be a salt lick attractant [19]. Kitahara et al. (2005) reported that sika deer in Hokkaido, Japan may be Ca-deficient [39]. Tsujii and Tokumoto (1997) reported that several natural salt licks in Japan did not have high Na contents [40], but they tended to have high Ca, Fe, Mg and Mn contents. Miyazaki and Kohara (2017) reported that sika deer licked an antifreeze containing calcium chloride (CaCl 2 ) as the main component, proposing that antifreezes accelerate the overpopulation of sika deer in Japan by providing Ca [41]. Our simulation supported the hypothesis that sika deer in Japan are potentially deficient in Ca, demonstrating that Ca requirements may not be met by feed plants, at least in certain situations (Figure 4). This study also raised the question of how sika deer acquire minerals in high-density areas with small amounts of feed plants, if they do not have access to salt licking sites.
Our finding might be beneficial for efficiently culling sika deer using minerals as attractants. Culling is one of the important management tools for overabundant sika deer populations [25,42], and suitable attractants have been verified to increase culling efficiency [43,44]. We suggest that Na can be an effective attractant for selectively culling female deer, which has been reported to be effective in reducing sika deer populations [25,26], especially during the lactating period ( Figure 2). We also suggest that Ca could be a potential attractant for sika deer (Figure 4), as well as Na, which has been used as an effective attractant for hunting sika deer [24]. However, since Ca deficiency is stronger in male deer, Ca may not be an effective attractant for selective culling of females. Given the different mineral requirement patterns observed for males and females (Figures 2 and 4) and potential seasonal changes in the effects of minerals to attract males and females [15], the combination of multiple mineral usage depending on seasons might be more effective for sika deer culling.
In summary, our results indicate that feed plants are unlikely to always provide sufficient Na and Ca, especially under current conditions in Japan, where sika deer overpopulation has intensified herbivorous pressure and the choice of feed plant species is often limited. Furthermore, we suggested that Na can be an effective attractant for selectively culling female deer, whereas Ca could be an attractant for male deer. Nevertheless, this study has several limitations. First, the data were obtained from a global database, and thus the analysis might not be applicable to a specific region. In the future study, data should be collected on a local scale, as plant mineral content is affected by soil conditions and climate. Second, this approach did not consider the feed plant species composition of wild sika deer and therefore is unable to accurately estimate the amount of minerals provided by feed plants. Quantitative data on the feed plant species composition of wild sika deer is necessary to more accurately estimate the mineral supply to sika deer. Third, the equations used to calculate the Na, Ca and Mg requirements were not constructed for sika deer, although a minor modification was made for sika deer (see Section 2 Materials and Methods). Equations for sika deer are required for a more robust estimation. Despite these limitations, this study demonstrates that sika deer in Japan might require extra Na and Ca sources in addition to feed plants, and therefore these minerals could be useful for developing effective culling methods.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/ani13061044/s1, Figure S1: Comparison between Na provision via feed plants and the Na requirement of sika deer under the 12 scenarios; Figure S2: Comparison between Mg provision via feed plants and the Mg requirement of sika deer under the 12 scenarios; Figure S3: Comparison between Ca provision via feed plants and the Ca requirement of sika deer under the 12 scenarios; Table S1: Feed plant species included in the present study.