多測站氣象合成模式驗證：單測站統計特性驗證

驗證完空間統計特性後，本研究進一步檢視，多測站氣象合成模式於個別測站的模擬情 形與統計特性維持狀況，其中包含一階馬可夫鍊參數（降雨事件）、雨量機率分布的統計檢定 與溫度相關氣象參數平均值與標準差統計檢定。圖 6-6 為各測站與各月份一階馬可夫鍊參數 200 年模擬資料與 10 年觀測資料的線性迴歸結果，從左到右分別為 P（W）、P（W|D）與 P

（W|W），其中𝑟²皆大於 0.99，而迴歸斜率也維持在 45 度線上，表現良好。

圖 6-6、一階馬可夫鍊參數模擬驗證結果

從左到右分別為 P（W）、P（W|D）與 P（W|W）的模擬驗證結果；驗證結果資料點為 七個測站用各月模擬與觀測資料計算出的馬可夫鍊參數的對應點。

在日雨量模擬上，本研究先以 Kolmogorov-Smirnov 統計測試法（KS Test）決定分佈，結 果發現，在指數分布、伽玛分佈、韋伯分佈與指數常態分佈中，韋伯分佈與指數常態分佈於 整體表現較好，故選用韋伯分佈作為本研究的雨量統計分佈。未來應用上應針對各測站與各 月的降雨資料，選用各自適合的雨量分佈，並將極端降雨事件分開產生，以降低參數估計時 所帶來的偏誤，本研究採 MLE 的方式估計參數，相較於動差法（Moment Method），有低估 平均值之偏誤（Bias），原因出自極端降雨事件提高整體平均值，但因本研究不探討極端降雨 事件所帶來的風險，故採 MLE 估計參數，使分佈曲線能夠最大化，擬合觀測資料點，降低極 端降雨事件對於參數估計之影響。產製資料後，本研究以無母數 Kruskal Wallis

統計檢定（K-W Test）檢測模擬過程中，其統計特性在 95%信心水準下（α = 0.05）統計分布的維持性，結 果顯示，在 200 年的模擬資料下，各測站於各月份皆通過檢定。本研究選用之韋伯分佈的 KS Test 與 K-W Test 呈現於表 6-1。

多測站氣象合成 200 年之氣溫相關氣象變數的雙尾 T 檢定與雙尾 F 檢定結果呈現於表 6-2 與表 6-3，於 95%信心水準（α = 0.05）下，在平均值部分，只有 C0C590 測站於 9 月月 均溫與三峽代表測站日最低溫之月平均於 10、11 月被拒絕，其餘部分皆為通過。然而，在標 準差部分，僅日最高溫之月標準差能有效被維持（C0C590 測站於 9 月被拒絕），日均溫之月 標準差與日最低溫之月標準差無法被有效維持，其中又以日最低溫之月標準差被拒絕總數較 多，如以月份檢視，則 3 到 5 月與 9 月到 12 月為主要被拒絕月份，而被拒絕之參數皆為高 估。這個部分可能為空間模擬中低估空間相關性或殘差計算方式所導致，期待未來研究進一 步探討與修正。

表 6-1、多測站氣象合成模式雨量分佈挑選 KS 檢定與模擬結果測試 K-W 檢定結果 韋伯分布 KS 檢定 （α = 0.05） K-W 檢定 （α = 0.05）

月份 C0C540 C0C590 C0C630 C0C650 C1C510

石門 三峽 C0C540 C0C590 C0C630 C0C650 C1C510 石門

三峽 1 Pass Reject Pass Pass Reject Reject Pass Pass Pass Pass Pass Pass Pass Pass 2 Reject Reject Reject Reject Reject Reject Pass Pass Pass Pass Pass Pass Pass Pass 3 Pass Pass Pass Pass Pass Reject Pass Pass Pass Pass Pass Pass Pass Pass 4 Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass 5 Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass 6 Reject Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass 7 Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass 8 Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass 9 Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass 10 Pass Pass Pass Pass Pass Reject Pass Pass Pass Pass Pass Pass Pass Pass 11 Pass Pass Reject Reject Pass Reject Pass Pass Pass Pass Pass Pass Pass Pass 12 Pass Pass Reject Pass Reject Reject Pass Pass Pass Pass Pass Pass Pass Pass 拒絕

總數 2 2 3 2 3 6 0 0 0 0 0 0 0 0

表 6-2、多測站氣象合成模式氣溫月平均值雙尾 T 檢定結果 月份

雙尾 T 檢定（α = 0.05）

月均溫 日最高溫之月平均 日最低溫之月平均

C0C540 C0C590 C0C630 C0C650 C1C510 石門 三峽 C0C540 C0C590 C0C630 C0C650 C1C510 石門 三峽 C0C540 C0C590 C0C630 C0C650 C1C510 石門 三峽

1 Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass 2 Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass 3 Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass 4 Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass 5 Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass 6 Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass 7 Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass 8 Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass 9 Pass Reject Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass 10 Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Reject 11 Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Reject 12 Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass 拒絕

總數 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2

表 6-3、多測站氣象合成模式氣溫月標準差雙尾 F 檢定結果 月份

雙尾 F 檢定（α = 0.05）

日均溫之月標準差 日最高溫之月標準差 日最低溫之月標準差

C0C540 C0C590 C0C630 C0C650 C1C510 石門 三峽 C0C540 C0C590 C0C630 C0C650 C1C510 石門 三峽 C0C540 C0C590 C0C630 C0C650 C1C510 石門 三峽

1

Pass Pass Pass Pass Pass Pass Reject Pass Pass Pass Pass Pass Pass Pass Pass Pass Reject Pass Pass Pass Reject

2

Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Reject Reject Reject Reject Reject Pass

3

Reject Reject Reject Reject Reject Pass Reject Pass Pass Pass Pass Pass Pass Pass Reject Reject Reject Reject Reject Reject Reject

4

Reject Reject Reject Reject Reject Reject Reject Pass Pass Pass Pass Pass Pass Pass Reject Reject Reject Reject Reject Reject Reject

5

Reject Reject Reject Reject Reject Reject Reject Pass Pass Pass Pass Pass Pass Pass Reject Reject Reject Reject Reject Reject Reject

6

Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Reject Reject Reject Reject Reject Reject Reject

7

Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass

8

Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass Pass

9

Reject Reject Reject Reject Reject Pass Reject Pass Reject Pass Pass Pass Pass Pass Reject Reject Reject Reject Reject Pass Reject

10

Reject Reject Pass Pass Pass Pass Reject Pass Pass Pass Pass Pass Pass Pass Pass Pass Reject Reject Pass Pass Reject

11

Reject Reject Reject Reject Reject Pass Reject Pass Pass Pass Pass Pass Pass Pass Reject Reject Reject Reject Reject Pass Reject

12

Reject Reject Reject Reject Reject Reject Reject Pass Pass Pass Pass Pass Pass Pass Reject Reject Reject Reject Reject Reject Reject

拒絕

總數

^{7 7 6 6 6 3 8 0 1 0 0 0 0 0 7 8 10 9 8 6 9}

6.2 水資源模組

在文檔中 發展氣候、水資源和糧食跨領域整合模式與結合氣候智慧調適演算法之應用－以桃園為例 (頁 97-102)