Timeline|時間軸

440,000 Years Ago (Late Pleistocene)

The Shanjiao Fault began to activate, with the overlying block of the Xinzhuang Fault shifting upwards and subsequently sliding downwards, forming the initial triangular shape of the Taipei Basin. The ancient Xindian Creek and the ancient Keelung River also shifted their courses to the northwest, draining into the northern part of the Taiwan Strait, leading to the cessation of accumulation on the Linkou Fan.


440000年前(更新世晚期)

山腳斷層開始活動,新莊斷層向上逆衝的上盤轉為下滑,形成雛形的三角形臺北盆地。古新店溪與古基隆河同時也轉向西北注入臺灣海峽北部,林口扇洲停止堆積。

440000

Between 44,000 to 180,000 Years Ago

In the northwestern corner of the basin, the areas from Wugu and Luzhou to the Guandu Plain had sunk near sea level. This region, influenced by the ancient Xindian Creek and the ancient Keelung River, formed a river-lake environment that accumulated layers of sandy and muddy sediment mixed with gravel.


44~180000年間

盆地西北角的五股、蘆洲到關渡平原一帶已經沉降到海平面附近,有古新店溪與古基隆河組成的河湖環境,累積了含有礫石的砂泥沉積層。

180000

180,000 Years Ago
The depositional environment of the ancient Tamsui River underwent a significant transformation. At that time, a large-scale lahar from the Datun Volcanic Group occurred, blocking the river systems’ outlets to the sea and resulting in the formation of the Taipei barrier lake.

160,000 Years Ago
The ancient Tamsui River broke through the Guandu Pass, causing the lake waters to recede and its area to shrink. The basin gradually reverted to its original river-lake sedimentary environment.


180000年前
古淡水河的堆積環境曾經一度發生重大的變遷,當時大屯火山群發生大規模的火山泥流,堵塞盆地河系的出海口,造成臺北堰塞湖。

160000年前
古淡水河切穿關渡隘口,湖水流逝、面積縮小,盆地逐漸恢復原來的河湖沉積環境。

170000

100,000 Years Ago

The area returned to its original river-lake environment and size.


100000年前

恢復到原來河湖環境與大小。

100000

100,000 Years Ago Onwards

Following the onset of the last ice age, the Taipei Basin underwent another major environmental transformation. At that time, the global sea level dropped by up to 120 meters. The base level of the ancient Tamsui River consequently lowered, intensifying headward erosion. The ancient Sanxia Creek captured the flow of the ancient Dahan River, which originally flowed toward the Taoyuan Plateau. Near Shimen, it turned northeast into the Taipei Basin, joining the ancient Xindian Creek and thus forming today’s Tamsui River system. During this period, the rivers’ transport capacity and maximum load-bearing capacity significantly increased, leading to the formation of alluvial fans and sedimentary layers dominated by large boulders within the basin.


100000年前開始

繼上次冰期開始,臺北盆地又經歷一次大規模的環境變遷。當時全球海水面下降可達120公尺,古淡水河基準面隨之下降,溯源侵蝕加強,古三峽溪襲奪原先流往桃園台地的古大漢溪,在石門附近轉向東北進入臺北盆地,匯入古新店溪而造就了今日的淡水河系。當時河流搬運力(capacity)與最大負載力(competence)都大增,在盆地中形成沖積扇與巨礫為主的沉積層。

90000

18,000 Years Ago

Since the start of the current interglacial period, rapidly rising sea levels flooded into the Taipei Basin, creating a vast inland bay. The tides could only ebb and flow through the narrow Guandu Pass.


18000年前

在現今的間冰期開始以來,快速上升的海水湧入臺北盆地,形成一個廣闊的內陸海灣,潮水出入僅能依靠狹隘的關渡口。

18000

8,000 Years Ago

The lake environment reached its peak, with the Song 4 sedimentary layer exhibiting its widest spatial distribution. Subsequently, the lake waters gradually receded from the Taipei Basin, reverting to a terrestrial river-lake sedimentary environment. The Song 5 and Song 6 layers, primarily composed of sediments from major rivers, are distributed within a depth of approximately 20 meters and are key sub-layers affecting soil liquefaction.


8000年前

湖泊環境到達最高點,此時沉積之松四層之空間分布最為寬廣;爾後湖水逐漸消退出臺北盆地,重現陸相河湖沉積環境,沉積之松五層及松六層,主要河流所造成之沉積物,分布深度則約在深度20m之內,為影響土壤液化最關鍵之次層。

8000

1654 – No Lakes in the Taipei Basin

Through the map “Sketch of Tamsui and its Nearby Villages and Keelung Island” drawn by the Dutch at the time, we can see that there were no lakes in the Taipei Basin then. The map also shows settlements along the banks of the Keelung and Tamsui Rivers, indicating that there were already inhabitants active in the basin area.


1654 台北盆地沒有湖泊存在

透過當時荷蘭人所繪製的〈淡水與其附近村社暨雞籠島略圖〉,我們可以得知當時的台北盆地沒有湖泊存在,且地圖中的基隆河與淡水河沿岸已有聚落,顯示當地已經有住民在盆地活動。

1654 淡水與其附近村社暨雞籠島略圖

1694 – The Great Kangxi Earthquake

“In April of the year Jiaxu (1694), the earth shook incessantly. The aboriginal people were terrified and moved away en masse. Suddenly, a vast inundation occurred, and it has been less than three years since then.” – This was recorded by Yu Yonghe in “Small Sea Travel Diaries” in 1697, as told to him by his interpreter, Zhang Da. This lake submerged at least the present-day areas of Beitou, Luzhou, Wugu, Shezi Island, and the Guandu Plain in the northwestern part of the Taipei Basin, covering several tens of square kilometers. For many years, scholars have believed that this “Kangxi Taipei Lake” was formed by the collapse of the Shanjiao Fault.


1694年 發生康熙大地震

「甲戌(1694年)四月,地動不休,番人恐怖,相率徙去,俄陷為巨浸,距今不三年耳」 –郁永河1697年記於《裨海紀遊》,由同行的通事張大告訴他。
這座湖,至少淹沒今日的北投、蘆洲、五股、社子島、關渡平原等台北盆地西北部,大小約莫數十平方公里,過去多年來許多學者認為這座「康熙台北湖」是山腳斷層陷落所造成。

 

 

1697 – Yu Yonghe’s Visit to Taiwan

After experiencing the turbulent waves of the Taiwan Strait and traveling north from Tainan in a jolting cart, Yu Yonghe finally arrived in Tamsui in May. Accompanied by the interpreter Zhang Da, he entered the Taipei Basin by boat. He wrote, “I, along with Mr. Gu and servants, boarded a sea vessel from the Tamsui Harbor. Ahead, between two mountains, lies the ‘Ganda Gate,’ a very narrow waterway. Upon entering the gate, the water suddenly widened, spreading into a vast lake, endless and boundless.” The ‘Ganda Gate’ he referred to is today’s Guandu. Once past the narrow Guandu, Yu Yonghe was greeted by a wide expanse of a vast lake, with no visible edges.


1697年 郁永河來台

郁永河在經歷過台灣海峽的驚濤駭浪、從台南一路「乘笨車」顛簸北上後,5月終於抵達淡水,並與一名通事張大乘船進入台北盆地。他寫到「余與顧君暨僕役平頭共乘海舶,由淡水港入。前望兩山夾峙處,曰甘答門,水道甚隘,入門,水忽廣,漶為大湖,渺無涯涘。」甘答門即為今日關渡,一過狹窄的關渡,一個碧波萬頃的廣闊湖面出現在郁永河眼前,看不到邊際。

1699-1704 – The “Kangxi Map of Taiwan” was completed during this period

The Kangxi Map of Taiwan does not show the existence of Taipei Lake. If the 1694 earthquake had indeed created Taipei Lake, it would have been implausible for such a large lake to have been filled in within just 5 to 10 years. Moreover, according to data from the Central Geological Survey, there is no geological evidence supporting the existence of Taipei Lake.


1699~1704 《康熙台灣輿圖》完成繪製於這段時間

康熙台灣輿圖裡沒有台北湖的存在,如果1694年的地震真造成台北湖,僅5~10年的時光不可能就把這麼大的湖給填平。而且,根據中央地質調查所的資料,並未找到台北湖的地質證據。

康熙台灣輿圖

1717, the 56th year of Emperor Kangxi’s reign, “General Map of Mountains and Rivers of Zhuluo County” 

In this map, the main river is the Tamsui River, which flows from Xiu Lang She (near the current Xiu Lang Bridge), passing through Dalangbeng She (now in the Datong District, Dalongdong area) and finally passing by Ling Shan Gong (now Guandu Temple) before reaching its estuary.

The map also shows a large lake with sea-going ships in the middle, indicating the existence of Taipei Lake at that time, large enough to accommodate large vessels.


1717年,康熙五十六年,《諸羅縣誌山川總圖》

在其圖中,主要的河道為淡水河,一路從秀朗社(現秀朗大橋一帶)沿著淡水河流域經過大浪泵社(現大同區大龍峒一帶)最後經過靈山宮(現關渡宮)最後抵達出海口。
也可以看到湖泊中央有海船在內,可見當時台北湖的存在,且規模是能夠讓大船駛入的。

諸羅縣誌

The right image is drawn using 2020 Google Earth data to closely approximate the Taipei Lake basin as depicted in the 1717 “Zhuluo County Annals.” Although the scale in the “Zhuluo County Annals” slightly differs from today’s actual proportions, the geographical locations in the map correspond with present-day realities.


右圖為使用2020Google Earth的圖資所繪製出最接近1717年《諸羅縣誌》的台北湖流域。雖然《諸羅縣誌》的比例與現今的真實比例略有不同,但是圖中的地理位置卻與現今是吻合的。

Google erth

1723-1734 – “Yongzheng Map of Taiwan”

During the Yongzheng era, the map of Taiwan depicted the Taipei Basin as being completely submerged in water. These ancient maps are considered definitive evidence of the existence of the Kangxi Taipei Lake.


1723~1734,《雍正台灣輿圖》

雍正年間繪製的台灣輿圖,台北盆地更完全浸在水中,這些古地圖都被認為是康熙台北湖存在過的確切證據。

雍正台灣輿圖

Conclusion:

Ancient Taipei Lake did indeed exist. However, the only historical witness to the Kangxi Taipei Lake during Yu Yonghe’s time was Yu himself, with no other corroborating documentary evidence. Considering that his visit occurred in June of the solar calendar, Professor Zheng Shinan hypothesized that Taipei Lake might have been a result of accumulated, undrained floodwaters. Two weeks before Yu Yonghe reached Taipei, he was stranded for a week in Niumashe (now Qingshui, Taichung) due to heavy rains, describing “overcast and heavy rain, making travel impossible; in the afternoon the rain stopped, and the sound of the roaring sea was heard, like the angry tides of Qiantang, incessant through the night.” Three days of heavy rain and towering sea waves likely indicated the impact of a plum rain front or a typhoon. This prolonged rainfall could have led to the formation of the Taipei Lake that Yu later observed.

I lean towards the belief that both a heavy rainfall and an earthquake might have occurred simultaneously. According to Yu Yonghe’s account, there was indeed heavy rainfall for three days. However, there might also have been an earthquake in Taipei, perhaps not a magnitude 7, which could explain why there were no records of it in the surrounding areas. The earthquake, combined with heavy rain, could have caused severe soil liquefaction. Driftwood and sediment accumulation at the river mouth might have contributed to the situation Yu observed.

As for the absence of the lake in the Kangxi Map of Taiwan, I suspect it could be due to inaccurate records. The cartographers might not have actually reached Taipei themselves, relying instead on hearsay and imagination. This is because the later General Map of Mountains and Rivers and the Yongzheng Map of Taiwan both depict the lake. I am more inclined to believe this explanation.


結論:

古台北湖是的的確確的存在過的,但郁永河時期的康熙台北湖其唯一的歷史見證人卻只有郁永河,之外沒有任何其他記錄文獻佐證。
由於當時值陽曆六月,鄭世楠教授推測台北湖可能是洪水蓄積不退。郁永河到台北前兩個禮拜,曾因大雨被困在牛罵社(台中清水)一個禮拜不得前進,記述「陰霾,大雨,不得行;午後雨止,聞海吼聲,如錢塘怒潮,至夜不息」,大雨連下三日,海浪滔天,很可能是梅雨鋒面或颱風侵襲,這波雨勢帶來的積水不退,或許就是後來所見的台北湖。
我則是認為可能兩者同時發生,也就是根據郁永河記載的,真的下了大雨三日,但台北也真的有地震,只是可能不是七級地震,所以其他周邊地方沒有記載,但因為地震,又大雨,所以土壤液化嚴重,然後出海口有漂流木以及泥沙淤積等等而形成郁永河看到的情況。
然,康熙台灣輿圖沒有湖到原因我想是因為記載不實,繪圖人員其實並未真正抵達台北,只憑口耳相傳與想像,因為後來的山川總圖與雍正台灣輿圖都有畫到,我比較是這樣相信的。