|Abstract: ||蝦類為臺灣東北部宜蘭灣內深海散射層之主要組成物種，本研究於2010~2015年間以海研二號之Simrad EK60雙頻(38與120 kHz)科學聲探系統在宜蘭灣海域進行9個航次之定點與測線觀測，以聲探能量差分(ΔSV)和加總(+SV)方式，並配合105個IKMT網次之生物採樣，進行蝦類體長與密度配對分析，以聲探流體散射模式(fluid sphere scattering model)與統計方法建立蝦類之雙頻聲探判釋法則，藉以擷取出不同體型蝦類之日週垂直洄游路徑，解明不同體型蝦類之日週垂直分布模式，最後以泛加成模式(Generalized Additive Models, GAM)來比較不同體型蝦類之分布密度與水文環境間之變動關係。結果如下： (一)以105個IKMT網次採集之DSL生物組成包括蝦類、魚類、水母及浮游動物。如以聲探幾何散射體觀之(含蝦類、魚類及水母)，蝦類比例約為97.76%，當中主要之蝦類有日本玻璃蝦(Pasiphaea japonica)、正櫻蝦(Sergia lucens)、晶瑩腹刺蝦(Systellaspis pellucida)、細角刺蝦(Oplophorus gracilirostris)及長角鬥劍赤蝦(Mataoenaeopsis provocatoria longirostris)，其中以日本玻璃蝦之數量最多，平均約佔組成比例之92.84%。 (二)透過群集分析法及38與120 kHz之聲探能量差(ΔSV)與能量加總(+SV)方式，建立了蝦類之雙頻聲探判別式；其中能量差分(ΔSV)之範圍為1.21 ~ 10.18 dB，能量加總(+SV)判別閾值為-136.38 dB。進一步計算出蝦類聲探等價球半徑(a)及標物反射強度(TS)與蝦類體長(TL，眼窩後緣至尾端)之線性迴歸關係，其關係式分別為：a = 0.1499 TL - 0.839 (R2 = 0.7155，df = 7，p < 0.05)；TS38 = 55.91 Log(TL) - 171.23 (R2=0.79，df=7，p < 0.05)；TS120= 25.65 Log(TL) - 118.08 (R2=0.78，df=7，p < 0.05)。另由蝦類體長之雙峰頻度分布型態與群集分析界定出大體型(44≤TLB≤51 mm)蝦與小體型(31≤TLS<44 mm)蝦之ΔSV值範圍分別為1.21≤ΔSVB≤2.58與2.58<ΔSVS≤10.18 dB，兩者之等價球半徑a分別為6≤aB≤8 mm及3≤aS<6 mm。 (三)本海域蝦類有日週垂直移動現象，日夜間分布分別在100~300 m及0~70 m左右，但似隨季節與體型差異而不同，夏季日間宜蘭灣海域經常觀測到上下兩層深海散射層(DSL)；其一為日間分布在水深約100~150 m之上層DSL，以小體型蝦類為主要聲探反射體；另一個為日間分布在約150~300 m之下層DSL，以大體型蝦類為主要反射體。而在冬季期間，上層DSL在日間之分布深度有加深之情形，並且會與下層DSL會合形成同一層DSL。 (四)GAM模式分析大小體型蝦類棲地環境結果顯示，日間大體型蝦最適棲息溫度為15.5 ℃，小體型蝦為16.2 ℃。夜間大體型蝦之分布與溶氧量最有關係，小體型蝦之分布則受溫度、鹽度、光照強度及葉綠素濃度之變動而左右。|
The Sergia 1ucens and Pasiphaea japonica are the main species of the sergestid shrimp fisheries in the I-Lan Bay of northeastern Taiwan. In this study, a multidisciplinary observation was carried out on board “Ocean Research II” in I-Lan Bay in northeastern Taiwan, recording acoustic, biological, and hydrographic data. The 120 and 38 kHz acoustic volume backscattering strength (SV) data of Simrad EK60 echosounder were simultaneously collected with the 105 biological net samples by using an IKMT. The Generalized Additive Models (GAM) was used to evaluate the relationship between abundance of shrimp and oceanic environmental factors. The results were summarized as below: (1) In the I-Lan Bay, the species of DSL including shrimp, fish, Jellyfish and zooplankton caught by 105 IKMT net. The pelagic shrimps were dominant. Based on the mean percentage composition of the species, Pasiphaea japonica occupied 92.84% of the catch. (2) The Bi-frequency discrimination method was used to differentiate the organisms in the DSL, and this was undertaken at 38 and 120 kHz. The SV strength values of -74.91 dB (SV38) and -61.52 dB (SV120) were the threshold levels to remove the ambient noise. The +SV values of -136.38 dB (Sum of SV38 and SV120) was the threshold level to remove the data without shrimp. The ∆SV between 1.21 and 10.18 dB indicated shrimp. The two-frequency algorithm with a highpass fluid sphere scattering model was used to measure the equivalent spherical radius (a) and the target strength (TS). For the pelagic shrimps, total lengths (TL) and the acousitic radius a values were positively linear (a = 0.1499 TL - 0.839, R2 = 0.7155, df = 7, p < 0.05). The TS and TL relationships were TS38 = 55.91 Log(TL) - 171.23 (R2=0.79，df=7，p < 0.05) and TS120= 25.65 Log(TL) - 118.08 (R2=0.78，df=7，p < 0.05) at 38 kHz and 120 kHz, respectively. Based on cluster Analysis and hypothesis test, the range of ΔSV was decided about 1.21 ~ 2.58 dB for the bigger size shrimp, and 2.58 ~ 10.18 dB for the smaller size shrimp; the range of radius a (mm) were estimated about 6~8 mm (the bigger size shrimp) and 3~6 mm (the smaller size shrimp). (3) In summertime, the echogram exhibits two distinct kinds of DSL. The two layers here referred to as upper DSL (UD) and lower DSL (LD). The UD in daytime distributed at the depths of 100~150 m, and the LD in daytime distributed at the depths of 150~300 m, respectively. The result showed the smaller size shrimp usually distributed in the UD, and the bigger size shrimp usually distributed in the LD. But in wintertime, the UD and LD always combined together. In summer, the bigger size shrimp started to ascend at dusk at a speed of 2.72 m/min, the smaller size shrimp started to ascend at a speed of 2.51 m/min; the bigger size shrimp started to descend at dawn with a speed of 1.20 m/min, the smaller size shrimp started to descend with a speed of 0.24 m/min. In winter, the bigger and smaller size shrimp always move together, the bigger size shrimp started to ascend at dusk at a speed of 0.89 m/min, the smaller size shrimp started to ascend at a speed of 0.87 m/min; the bigger size shrimp started to descend at dawn with a speed of 1.46 m/min, the smaller size shrimp started to descend with a speed of 1.39 m/min. (4) About the relationship between abundance of shrimp and oceanic environmental factors showed temperature influenced abundance of the bigger size shrimp in daytime (Optimum temperature was 15.5 ℃), but dissolved oxygen influenced abundance of the bigger size shrimp in nighttime (Optimum dissolved oxygen was 5.8 ppm). Temperature and salinity influenced abundance of the smaller size shrimp in daytime (Optimum temperature and salinity were 16.2 ℃ and 34.53 , respectively), but temperature, salinity, illuminance and Chlorophyll-a influenced abundance of the smaller size shrimp in nighttime (Optimum temperature, salinity, illuminance and Chlorophyll-a were 24~25℃, 34, 0.0001~0.005 μmol/m2/s and 0.19 mg/m3, respectively).