台灣島境內高山疊起，地形起伏變化大，超過3,000公尺的高峰將近280座。由於山區的自然生態與景觀資源豐富，成為引人入勝的休閒去處。統計顯示近年來登山活動日益蓬勃發展，成為假日的熱門休閒運動之一。然而，台灣的山岳環境與氣候存在許多潛在的危險因子，使得登山者往往需要承擔高度的風險。因此如何保障登山活動的安全性實屬當務之急。除了完善的登山計畫與充分的山岳資訊外，如何在最短時間內找到失蹤、迷途或陷入困難的登山者，也可以有效地減少登山意外事件的傷亡。 本研究蒐集了近10年台灣中級山及高山範圍內，中長程探勘型登山記錄，擷取有用與登山者步行速率相關的變數(例如人數、體能狀態、疲勞度、經驗技術、團對結構、背負量、糧食...等等)，並配合各山徑的地形與環境(如坡度、高度、崎嶇度、植被情形..等等)，進行複迴歸行分析，得到「一般及上坡路段」與「陡下路段」兩組方程式構成之「台灣高山地區登山步行速率推估模式」。由於研究區域的範圍係傳統登山路線以外的罕行山徑以及荒野山區，與迷途或失蹤型山難的登山者面對之環境類似，因此本研究建立之統計模式，可以應用於山難事件的搜救中，提供救援人員估計並預測迷途或失蹤者自走失點隨時間的可能活動範圍，提高搜救工作之效率，減少登山活動所造成的傷亡。

The most prominent landscape in Taiwan is the island-wide high mountain terrains. There are more than 280 mountains over 3000 meters in altitude. The magnificence of mountain scenery and the diverse ecology attract many people for mountaineering. The statistics shows mountain climbing is increasing in recent years as a very popular activity in holiday. However, the geologic and climatic condition in the high mountain area is hazardous to those who climb mountains. In addition to the proper mountain plan and information for every member of the team, one way to minimize the accidents is to locate the climber in lost or in trouble as soon as possible. In this study, a multi-regression model is established for pace velocity in the high mountains above 1000m in altitude, based on 10 years log by climbers and database of topography and vegetation along the trails. This model consists of two equations with respect to the conditions of "uphill" and "downhill". The accident of missing is inevitable occurred in the mountainous area. Therefore, the application of this statistic model not only provides help in search and rescue in the accidents by giving estimates of the deviated distance (or area), but also reduces risk of mountaineering by verifying the schedule with simulation in advance.