瓦拉米地區臺灣山羌之活動習性

Activity and habit of Formosan muntjac at Walami area

陳怡君(Chen, Yi Jun)

指導教授 ： 王穎

國立臺灣師範大學/理學院/生物學系/博士(2000年)

摘要

自1992年1月至1997年4月在玉山國家公園瓦拉米地區針對台灣山羌(Muntiacus reevesi micrurus)的活動模式、吠叫行為、個體出沒及其與環境之關係進行研究。對3隻雄性及1隻雌性山羌進行無線電追蹤之結果顯示，山羌在一日當中有48%的時間處於活動狀態，每日的活動期以8次最常見，平均每天7.7次；活動期的時間在0.5至7.5小時間，平均為1.5小時，日間(6:00-18:00)的活動期時間長度明顯大於夜間(1.6及1.3小時，t-test，p<0.05)，日間的不活動期長度亦明顯比夜間短(1.4及1.8小時，t-test，p<0.05)。活動時間比例最高的時間為16時，活動比例最低時為20時，活動模式呈現晨昏活動偏日行性的型式，清晨及黃昏的活動時間比例最高時間在日出後1小時及日落前1小時內出現。各季節之每日活動時間比例由春季最高，依序向夏、秋及冬季遞減，雨天時山羌的活動時間比例較低，其主要影響時間在夜晚；包括霧天、雨天及霧雨天在內的雨日時間，山羌活動較集中於清晨及黃昏，每日的活動時間比例比其他天氣時低。
就山羌之吠叫及目擊調查結果顯示，白天調查時間(8hr)在步道上記錄之山羌吠叫頻度平均為10.48次/天，在夏季的記錄最低；而在步道上目擊山羌之頻度為0.91隻次/天，最高為8隻次/天，在夏季之目擊率較低。白天各小時記錄的吠叫及目擊頻度有由上午向黃昏時段漸增之現象，與由無線電追蹤所得之活動模式相似；雨日時間的目擊與吠叫頻度都較低。步道上平均每日目擊山羌頻度有逐年增加之現象，吠叫頻度之增加趨勢則較小。清晨在穿越線上記錄之山羌吠叫頻度平均為4.25次/hr，目擊山羌頻度則平均為0.46隻次/hr，影響目擊頻度的因素有森林底層1m以下植被隱蔽度及林相，植被隱蔽度越高則目擊頻度越低(r2=0.47，p<0.05)，排遺數量及吠叫頻度也有隨隱蔽度越高則降低的趨勢，但未有顯著差異。天然闊葉林中有較高的目擊與吠叫頻度，排遺數量也較多，針闊葉混淆林中之記錄則為次高，而人工針葉林中的目擊與吠叫頻度及排遺數量最低。
分析1994年至1997年間21個月之訪客資料顯示，該期間平均每天目擊山羌1.1±1.4隻次，無訪客時平均每天目擊山羌1.5±1.6隻次，有訪客時平均每天目擊0.8±1.1隻，而當日訪客人數在15人以下時(含15人)目擊山羌數量平均為1.0隻次/天，訪客人數在15人以上時則為0.4隻次/天，其間有顯著差異(ANOVA，p<0.05)。調查期間平均每天記錄10.3±5.8次吠叫回合，訪客有無對當日記錄叫聲次數並無影響。近距叫聲(發聲山羌與步道距離<50m)佔所有叫聲記錄之15%，在無訪客時平均單日內近距叫聲所佔比例為21%，有訪客時則為11%，有明顯差別(t-test，p<0.05)。

關鍵字

台灣山羌 ； 活動模式 ； 吠叫 ； 目擊數量 ； 訪客

並列摘要

From January 1992 to April 1997 the activity pattern, barking behavior, and impact of visitors on muntjac (Muntiacus reevesi micrurus) were studied at Walami area in Yushan National Park, Taiwan. The radio tracking results from 4 muntjac (3 males, 1female) showed that muntjac was active about 48% of the 24 hr time budget. Eight active periods were most commonly found within a day with an average about 7.7 periods. Each active period lasted from 0.5 to 7.5 hr with an average 1.5 hr. The average length of the active period during the daytime (6:00-18:00) was significantly longer than that of the nighttime (1.6 vs. 1.3hr，t-test，p<0.05), while the length of the inactive period was shorter (1.4 vs. 1.8hr，t-test，p<0.05). Within a day the highest proportion of the active hour was recorded in 16:00 hr; the lowest was in 20:00 hr. Though there was more daytime activity than that of nighttime, the activity pattern was concerned to be crepuscular with two peak activity one found one hour after dawn and another found one hour before dusk. The daily activity varied among seasons; spring was the highest followed by summer, fall, and winter. Weather also affected muntjac activity. Among different rainy days (foggy, rainy, and rain within fog) low activity was found especially during the nighttime and the active period was concentrated at dawn and dusk.
Daytime trail survey (8hr/day) showed that muntjac barking was on the average 10.48 bouts per day and sighting was 0.91 muntjac per day with the highest record of 8 muntjac a day. Both barking and sighting were recorded increasingly from the morning to the evening and were coincided with the activity pattern obtained from radio tracking. They were also recorded less frequently during the rainy weather. Summer had both the lowest records of barking and sighting. During the study period both sighting and barking increased through years, while the latter was less prominent than the former. Morning survey from transect lines showed that the average barking was 4.25 bouts/hr and sighting was 0.46 muntjac/hr. Vegetation type and vertical cover (<1m) of the understory were related to sighting frequency. There was a negative correlation between cover and sighting (r2=0.47, p<0.05). Similar trends could also be found between cover and pellet count and between cover and barking, but no significant difference. Among different vegetation types natural broad-leaved forest had the highest counts of sighting, barking, and pellet; followed by mixed forest and then pine plantation.
Impact of visitors on muntjac for 21 months from 1994 to 1997 were monitored. The average daily sighting was 1.5±1.6 muntjac when there were no visitors and 0.8±1.1 when there were visitors with overall sighting 1.1±1.4 muntjac. Party size also affected sighting. It was 1.0 muntjac per day when the visitor number was below 15(including 15), it became 0.4 when the visitor number was above 15. There was a significant difference among visitor numbers on muntjac sighting (ANOVA, p<0.05). Visitors had no significant effect on overall average daily muntjac barking (10.3±5.8 bouts). However, when the short distanced (<50m) barking which was composed of 15% of the total counts was compared with visitor numbers, it was 21% of the total count when there was no visitor and 11% when there were visitors.

並列關鍵字

Formosan muntjac ； activity pattern ； barking ； sighting ； visitor

參考文獻

Pei, K. 1996b. Recent data on the use of auto-trigger camera to record activity patterns of wild animals in Taiwan. Proceedings of the international Union of Game Biologists; XXII Congress.

林良恭．1981．臺灣陸生哺乳動物研究．東海大學生物研究所，384pp．

Aulak, W. and J. Babinska-Werka. 1990a. Estimation of roe deer density based on the abundance and rate of disappearance of their faeces from the forest. Acta Theriogica. 35(1-2):111-120.

Aulak, W. and J. Babinska-Werka. 1990b.Use of agricultural habitats by roe deer inhabiting a small forest area . Acta Theriologica. 35(1-2):121-127.

Aulak, W. and J. Babinska-Werka. 1990c. Preference of different habitat and age classes of forest by roe deer. Acta Theriologica. 35(3-4):289-298.