A preliminary study to predict the spatial distribution of dung from beef cattle in a slope-grazed pasture.
A preliminary study to predict the spatial distribution of dung from beef cattle in a slope-grazed pasture.
复制标题
预测坡地放牧牧场肉牛粪便空间分布的初步研究。
DOI：
10.1111/grs.12077
复制
发表时间：
2015
影响因子：
1.3
通讯作者：
H.J.
中科院分区：
文献类型：
--
作者：
Yoshitoshi;R.;Watanabe;N.;Yasuda;T.;Kawamura;K.;Sakanoue;S.;Lim;J. and Lee;H.J.
本研究的目的是预测牛粪的空间分布的贝叶斯估计的基础上，广义线性混合模型（GLMM）和内在条件自回归（CAR）模型使用牧草绿色生物量（GBM）和距离水槽（Dw）在一个矩形的斜坡放牧牧场（0.85公顷）在北海道，日本。在20头日本黑牛进行为期4天的放牧试验（2010年6月14日至18日）后，我们在围场中设置了10 × 10 m的网格单元（共85个单元），并计算了每个单元中的粪便沉积量（Nd）。95%后验概率区间表明，较大的Nd分布在GBM较高的区域和靠近水槽的区域。CAR模型的偏差信息准则（DIC）值（DIC = 291.5）低于GLMM模型（DIC = 502.6），建议作为一个更好的模型。
This study aimed to predict the spatial distribution of cattle dung based on Bayesian estimation with a generalized linear mixed model (GLMM) and an intrinsic conditional autoregressive (CAR) model using herbage green biomass (GBM) and distance from a water trough (Dw) in a rectangular slope‐grazed pasture (0.85 ha) in Hokkaido, Japan. After a 4‐day grazing trial (June 14–18, 2010) with 20 Japanese Black cattle, we set 10 × 10 m grid cells (total of 85 cells) in the paddock and counted the number of dung deposits (Nd) in each cell. The 95% posterior probability intervals indicated that a greater Ndwas distributed in areas with a higher GBM and those located closer to the water trough. Lower value of deviance information criterion (DIC) was obtained in CAR model (DIC = 291.5) than in GLMM (DIC = 502.6) to be suggested as a better model.