Abstract
Keenjhar Lake, Pakistan's largest freshwater lake and an important Ramsar site, provides habitat for internationally important water birds. Annually, 385,000 people visit the lake. The lake is threatened by a variety of causes, including industrial and agricultural pollution. To support its sustainable management and conservation, the lake's recreational value is estimated using an individual travel cost model. Randomly selected visitors are interviewed during peak season about their recreational travel behavior and perception of lake conditions. Key issues in travel cost modeling are addressed, including the opportunity cost of time, group travel, substitution and income effects, and endogenous stratification and truncation due to on-site sampling. Poisson and negative binomial regression models produce similar results. We find significant over-dispersion, and therefore, use the more conservative truncated negative binomial model results to estimate consumer surplus. The value of this assessment method for resource managers is illustrated by comparing the consumer surplus with existing pricing and budgeting mechanisms. The annual flow of benefits from lake recreation appears to be almost 50 times higher than the average entrance fee paid by the predominantly higher-income segments visiting the lake, suggesting scope for increasing fees and reallocating government budgets to finance the necessary lake protection measures.
使用一种旅行花费模型来估计巴基斯坦最大的淡水湖在支持可持续性旅游管理中的娱乐价值
Keenjhar湖,巴基斯坦最大的淡水湖和一个重要的拉姆萨湿地,为国际性的重要的淡水鸟类提供了栖息地。每年有385000游客参观该湖。该湖被一系列的因素所威胁,包括工业和农业的污染。为了支持它的可持续性管理和保护,该湖的娱乐价值通过一个单独的旅行费用模型来评估。随机选择的参观者在旺季的时候被采访关于他们的娱乐性旅行行为和对湖的状态的看法。在旅行花费模式化中的主要因素都有被讨论到,包括时间的机会成本,组团旅行,替代品和收入的影响,和由场地抽样造成的内源性分层和截断。用泊松分布和负面的二项式回归模式产生了相近的结果。我们发现了重要的偏大离差,因此使用了更加保守的截断负面二项式模式结果来估计消费者盈余。该评估方法对资源经理的价值通过对比消费者盈余和已有的价格和预算机制来显示出来。湖泊娱乐每年的收益量显示为高收入人群参观湖泊平均门票收入的50倍,这建议提高入场费和重新安排政府预算以资助重要的湖泊保护措施。
Notes
1. In 2010, one Pakistan Rupee (PKR) equaled, on average, approximately US$ 0.012.
2. Note that this is not the same test as the likelihood ratio test result shown in . The χ2-distributed likelihood ratio test in under the heading “Model summary statistics” indicates that the null hypothesis of zero-coefficient estimates for all variables is convincingly rejected at the 1% level. Hence, the estimated models presented in are all highly significant.
3. In this case, the dispersion is a function of the expected mean of the counts for each observation.