PREY DISTRIBUTION AS A FACTOR DETERMINING THE CHOICE OF OPTIMAL FORAGING STRATEGY
PREY DISTRIBUTION AS A FACTOR DETERMINING THE CHOICE OF OPTIMAL FORAGING STRATEGY
复制标题
DOI：
10.1086/283754
复制
发表时间：
1981-01-01
影响因子：
2.9
通讯作者：
YAMAMURA, N
中科院分区：
文献类型：
--
作者：
IWASA, Y;HIGASHI, M;YAMAMURA, N
在确定性连续模型不合适的情况下，采用随机离散模型拟合实验情况，研究了捕食理论中的最优斑块利用问题。他们考虑了三种基本策略，在何时离开捕食者觅食的区域方面有所不同；即(1)经过了一段固定的时间，(2)捕获了固定数量的猎物，(3)连续两次捕获之间的间隔超过了一段固定的时间。在特定条件下，这些固定数量中的每一个都有一些优化相关策略的值。将优化后的策略进行比较，以确定最有效的策略。在其他条件相同的情况下，策略的排名主要取决于不同斑块之间猎物分布的类型；例如，当猎物分布的方差足够大时，策略3是最好的，而当方差很小时，策略3往往是最差的。这些策略没有使用完整的信息来估计斑块中剩余猎物的数量；两次连续攻击之间的每一个时间间隔都可能被更老练的捕食者利用。统计决策理论表明，在随机搜索假设下，斑块中已捕获的猎物数量n和总觅食时间t是估计未开发猎物数量的最小充分统计量。在随机搜索的假设下，估计量r最多只是n和t的函数，对时间区间分布的详细了解无关紧要。当估计量r (n,t) <某一优化所采用策略的值时，老练捕食者应该离开补丁。估计量r取决于猎物的分布：如果猎物是传染性分布的，每次捕获一次，剩余猎物的估计值就会上升，如果猎物是有规律分布的，则估计值会下降。但是，r可能是n或t单独的函数，这取决于猎物斑块间分布的类型。特别地，r是n的函数，只有对于完全正则分布，最佳策略被简化为基本策略2。如果分布是随机的（泊松），则r仅是t的函数。那么最佳策略就是简单的初级策略1。观察下的捕食者是否真的按照猎物的分布表现出最佳行为，应该通过绘制捕获的数量与捕食者在每个斑块中停留的时间长度来揭示。
The optimal-patch-use problem in predation theory is investigated by use of a stochastic discrete model to match experimental situations when deterministic continuous models are inappropriate. Three elementary strategies are considered, differing in when to leave the patch in which the predator has been foraging; namely, (1) a fixed time has passed, (2) a fixed number of prey has been captured and (3) the interval between 2 successive catches has exceeded a fixed time. Each of these fixed quantities has some value that optimizes the strategy concerned, given certain conditions. The optimized strategies are compared to determine the most efficient. The ranking of the strategies depends critically on the type of prey distribution between patches, other things being equal; e.g., strategy 3, the best when the variance of prey distribution is sufficiently high, tends to be the worst when the variance is minimal. These strategies do not use full information for estimating the number of prey remaining in the patch; every time interval between 2 successive cathces might be used by a more sophisticated predator. The statistical decision theory reveals that the number of prey already taken (n) and the total time spent foraging (t) in the patch are the minimal sufficient statistics for estimating the number of unexploited prey under the random-search assumption. Under the assumption of random search, the estimator r is a function of n and t only at most, and detailed knowledge of the distribution of the time intervals in immaterial. The sophisticated predator should leave the patch when the estimator r (n,t) is < a certain value that optimizes the strategy adopted. The estimator r depends on the prey distribution: if prey are distributed contagiously, estimated value of the remaining prey jumps up each time a capture is made, but it steps down if prey are distrubuted regularly. But, r may be a function of n or t alone, depending on the type of the between-patch distribution of prey. In particular, r is a function of n only for a completely regular distribution, and the best strategy is reduced to elementary strategy 2. If the distribution is random (Poisson), r is a function of t only. Then the best strategy becomes simple elementary strategy 1. Whether the predator under observation actually behaves optimally in accordance with the distribution of prey should be revealed by plotting the number of captures against the length of period for which the predator stayed in each patch.