郭剛剛,樊 偉,張勝茂,鄭巧玲,王曉璇
(1.上海海洋大學(xué)海洋科學(xué)學(xué)院,上海 201306;2.中國水產(chǎn)科學(xué)研究院東海水產(chǎn)研究所,農(nóng)業(yè)部東海與遠(yuǎn)洋漁業(yè)資源開發(fā)利用重點實驗室,上海 200090)
船位監(jiān)控系統(tǒng)數(shù)據(jù)挖掘與應(yīng)用研究進(jìn)展
郭剛剛1,2,樊 偉2,張勝茂2,鄭巧玲1,2,王曉璇2
(1.上海海洋大學(xué)海洋科學(xué)學(xué)院,上海 201306;2.中國水產(chǎn)科學(xué)研究院東海水產(chǎn)研究所,農(nóng)業(yè)部東海與遠(yuǎn)洋漁業(yè)資源開發(fā)利用重點實驗室,上海 200090)
船位監(jiān)控系統(tǒng)(vessel monitoring system,VMS)作為一種漁船監(jiān)控手段,同時也為漁業(yè)科學(xué)研究提供了一種新的數(shù)據(jù)來源。VMS數(shù)據(jù)記錄了漁船實時的船位、航速、航向等動態(tài)信息,已被廣泛應(yīng)用于海洋漁業(yè)的諸多領(lǐng)域。本文結(jié)合國內(nèi)外研究現(xiàn)狀,對VMS數(shù)據(jù)分析挖掘方法進(jìn)行了歸納和總結(jié),對采用VMS數(shù)據(jù)進(jìn)行捕撈努力量估算、漁民行為特點和漁場分析、捕撈活動對海洋生態(tài)環(huán)境影響等方面的研究進(jìn)展進(jìn)行了綜述,并在此基礎(chǔ)上分析了VMS數(shù)據(jù)在海洋漁業(yè)上的應(yīng)用前景及其存在的問題,對今后我國采用VMS數(shù)據(jù)進(jìn)行相關(guān)研究提出了建議。
VMS數(shù)據(jù);捕撈努力量;漁場;生態(tài)環(huán)境影響評估
漁業(yè)數(shù)據(jù)是進(jìn)行漁業(yè)科學(xué)研究的基礎(chǔ),數(shù)據(jù)的精度直接影響到研究結(jié)果的準(zhǔn)確性。目前,原始的漁業(yè)數(shù)據(jù)主要來自于漁船作業(yè)時記錄的漁撈日志,但漁撈日志往往只記錄漁船處于捕撈狀態(tài)時的情況,漁船的實時位置、航速、航向、航行軌跡等信息則無從得知,且隨著漁撈日志數(shù)據(jù)的層層上報,錯報、誤報的情況也時有發(fā)生,數(shù)據(jù)的完整性和準(zhǔn)確性均有待提高[1]。船位監(jiān)控系統(tǒng)(vessel monitoring system,VMS)是一種集全球衛(wèi)星定位、電子地圖、電子海圖、計算機(jī)網(wǎng)絡(luò)通訊和數(shù)據(jù)庫技術(shù)于一體的綜合應(yīng)用系統(tǒng)[2]。其主要功能是實時獲取并存儲船舶的船位和運(yùn)行狀態(tài)信息,并將這些信息通過網(wǎng)絡(luò)通訊傳送給岸上監(jiān)控中心,實現(xiàn)船舶與岸上監(jiān)控中心之間信息的交互[3]。VMS可以實時獲取漁船的船位、航速、航向等漁船動態(tài)信息,從而彌補(bǔ)了漁撈日志數(shù)據(jù)在這些信息方面的不足。VMS的技術(shù)核心在于衛(wèi)星定位和網(wǎng)絡(luò)通訊,衛(wèi)星定位廣泛采用美國的全球定位系統(tǒng)(global position system,GPS);在網(wǎng)絡(luò)通訊方面,遠(yuǎn)洋漁船主要使用Inmarsat-C和ARGOS系統(tǒng),近海和內(nèi)陸漁船主要使用船舶自動識別系統(tǒng)(automatic identification system,AIS);我國自主研發(fā)的北斗衛(wèi)星船位監(jiān)控系統(tǒng)集衛(wèi)星定位與網(wǎng)絡(luò)通訊功能于一體,目前也已投入使用。
VMS設(shè)計之初是為了對船舶位置進(jìn)行實時監(jiān)測,以確保當(dāng)船舶出現(xiàn)意外狀況時可以提供及時的救助。20世紀(jì)末期,由于世界主要傳統(tǒng)經(jīng)濟(jì)漁業(yè)資源的衰退,國際社會要求加強(qiáng)漁業(yè)資源養(yǎng)護(hù)與管理的呼聲越來越高,各海洋國家對其所轄海域內(nèi)漁業(yè)資源的管理也不斷加強(qiáng)。傳統(tǒng)的漁船監(jiān)控和管理主要依靠海上巡邏和登臨檢查,在管理上有一定的局限性。1988年,葡萄牙開發(fā)了世界上首個漁船監(jiān)控系統(tǒng)MONICAP,該系統(tǒng)可以將漁船的實時船位、航向、航速等數(shù)據(jù)自動傳送到岸上監(jiān)控中心,從而使監(jiān)控中心能實時掌握和監(jiān)督漁船的作業(yè)動態(tài),大大提高了漁船監(jiān)控的效率和力度,隨后美國、澳大利亞、新西蘭等漁業(yè)國家也都相繼研發(fā)了本國的VMS對漁船進(jìn)行監(jiān)管[4]。1996年,歐盟要求歐洲所有長度大于24 m的漁船強(qiáng)制安裝VMS,到2012年,VMS的強(qiáng)制安裝范圍擴(kuò)大到了所有長度大于12 m的漁船[5]。截至目前,基本上世界上所有的漁業(yè)國家均采用VMS作為監(jiān)控手段來管理和養(yǎng)護(hù)所轄海域的漁業(yè)資源。
隨著安裝有VMS的漁船數(shù)量以及VMS數(shù)據(jù)積累時間的不斷增長,VMS數(shù)據(jù)在海洋漁業(yè)上的應(yīng)用領(lǐng)域也不斷拓展。我國于本世紀(jì)初引進(jìn)VMS。2006年,隨著我國自主研發(fā)的北斗衛(wèi)星船位監(jiān)控系統(tǒng)在南沙正式投入使用,我國漁船監(jiān)控系統(tǒng)的建設(shè)進(jìn)入快速發(fā)展時期[6]。本文根據(jù)相關(guān)國內(nèi)外文獻(xiàn)資料,概述了VMS數(shù)據(jù)的分析挖掘方法,以及其在海洋漁業(yè)相關(guān)領(lǐng)域應(yīng)用的研究進(jìn)展,分析了VMS數(shù)據(jù)的應(yīng)用前景和存在的問題,并提出了相應(yīng)的建議,以期能為我國船位監(jiān)控系統(tǒng)數(shù)據(jù)的挖掘和應(yīng)用提供參考。
VMS數(shù)據(jù)包含有漁船船位、發(fā)報時間、航向、航速等漁船動態(tài)信息,數(shù)據(jù)的定位精度多為10 m,但不同通訊系統(tǒng)的VMS數(shù)據(jù)間回報頻率有很大差異。遠(yuǎn)洋船位監(jiān)控系統(tǒng)Inmarsat-C和ARGOS的數(shù)據(jù)回報頻率較低,每隔約4 h發(fā)送一次;AIS數(shù)據(jù)回報頻率與航速呈正比,航行時回報頻率在12 s以內(nèi);我國北斗衛(wèi)星船位監(jiān)控系統(tǒng)數(shù)據(jù)回報頻率為3 min。VMS數(shù)據(jù)雖然包含了豐富的漁船動態(tài)信息,但這些信息未能直接體現(xiàn)漁船的狀態(tài),即漁船是否處于捕撈狀態(tài)。且VMS的兩個點數(shù)據(jù)之間有一定的時間間隔,單純的點數(shù)據(jù)分析并不能反映出漁船的真實航行軌跡。因此,目前對于VMS數(shù)據(jù)的分析和挖掘研究主要集中在兩個方面:一是漁船狀態(tài)判別,二是漁船軌跡重構(gòu)。
1.1 漁船狀態(tài)判別
由于VMS并沒有傳遞具體的漁船狀態(tài)信息,因此對于漁船狀態(tài)的識別和劃分成為VMS數(shù)據(jù)挖掘中不可避免的一個步驟,且識別的準(zhǔn)確度直接關(guān)系到VMS數(shù)據(jù)的后續(xù)應(yīng)用。目前,國內(nèi)外研究中采用VMS數(shù)據(jù)分析漁船狀態(tài)的方法可以概括為2種:(1)通過分析漁船船速的變化判斷漁船狀態(tài);(2)通過分析船速、航向等特征數(shù)據(jù)組成向量判斷漁船狀態(tài)。
以往的研究多是通過對漁船速度設(shè)定閾值來實現(xiàn)對漁船狀態(tài)的劃分,這種方法在拖網(wǎng)漁船狀態(tài) 的 判 別 中 應(yīng) 用 較 為 廣 泛[7-10]。但BERTRAND等[11]分析指出,簡單的通過速度閾值來判斷漁船狀態(tài)會使處于捕撈狀態(tài)的船位點數(shù)量被高估。LEE等[12]對不同研究中捕撈狀態(tài)的速度識別閾值進(jìn)行了統(tǒng)計,發(fā)現(xiàn)沒有任何一種速度閾值適用于所有的漁船狀態(tài)識別。且不同作業(yè)方式的漁船在速度、航向等特征數(shù)據(jù)的表現(xiàn)方式上也有著明顯的差異,從而導(dǎo)致這種方法很難適用于所有類型的漁船,不具備很好的可推廣性。
綜合考慮船速和航向特征數(shù)據(jù)與漁船狀態(tài)間的非線性關(guān)系,通過構(gòu)建各類分析模型來識別漁船的狀態(tài)是當(dāng)前漁船狀態(tài)判別研究的重點。其中,應(yīng)用較為廣泛的是狀態(tài)空間模型(statespace model),狀態(tài)空間模型在處理離散的數(shù)據(jù)方面有獨(dú)特的優(yōu)勢,多被用于描述動物種群的動態(tài)變化以及在特定環(huán)境中重新估算標(biāo)記動物在不同狀態(tài)下的真實活動軌跡[13-17]。WALKER等[18]首先將狀態(tài)空間模型引入金槍魚圍網(wǎng)漁船的狀態(tài)判別研究中,漁船的狀態(tài)(找魚,捕撈,停泊,航行)由以航向和航速為參數(shù)的隨機(jī)多項式來表示,并通過隱馬爾可夫轉(zhuǎn)移概率矩陣進(jìn)行判別,最后采用貝葉斯框架對模型進(jìn)行簡化,驗證結(jié)果表明狀態(tài)空間模型對漁船狀態(tài)的識別有較高的準(zhǔn)確性和可推廣性。此外,JOO等[19]利用人工神經(jīng)網(wǎng)絡(luò)的方法來降低對于捕撈位置判斷的錯誤率,并通過敏感性試驗對參數(shù)和訓(xùn)練函數(shù)進(jìn)行優(yōu)化,使得對于捕撈位置的判斷達(dá)到了較高的正確率。對于高時空分辨率的北斗數(shù)據(jù),張勝茂等[20]提出了用統(tǒng)計學(xué)的方法來分析拖網(wǎng)漁船狀態(tài),通過對較長時間的航速和航向數(shù)據(jù)進(jìn)行統(tǒng)計分析,了解漁船航速和航向差數(shù)據(jù)的分布特征,進(jìn)而獲取捕撈狀態(tài)下漁船航速和航向差的閾值來提取處于捕撈狀態(tài)的船位點,最后采用過濾窗修正提高判定的準(zhǔn)確率。
1.2 漁船軌跡重構(gòu)
VMS數(shù)據(jù)是一系列離散的點數(shù)據(jù),且不同通訊系統(tǒng)的VMS數(shù)據(jù)回報頻率不一[21]。許多學(xué)者嘗試用這些離散的點數(shù)據(jù)來分析漁業(yè)活動[22-26],但單純的點數(shù)據(jù)分析很難反映漁船的真實航行軌跡。如何通過這些離散的點數(shù)據(jù)來準(zhǔn)確的重構(gòu)漁船活動軌跡是VMS數(shù)據(jù)在海洋漁業(yè)上應(yīng)用的關(guān)鍵。漁船軌跡重構(gòu)的方法主要是插值,通過插值的方法能夠較好的還原漁船的真實活動軌跡[27]。在陸地動物行為學(xué)研究領(lǐng)域,多采用插值的方法研究動物的活動軌跡[28-31],目前對于漁船軌跡的研究也大多是借鑒了這些插值方法。
最簡單的插值算法是線性插值(straight linear interpolation)[32-33],這種方法的優(yōu)點是簡單快速,而且對于連續(xù)和不連續(xù)的數(shù)據(jù)都可以處理,但線性插值的結(jié)果可能與漁船的實際軌跡存在較大偏差,特別是對于低回報頻率的VMS數(shù)據(jù),很可能導(dǎo)致漁船實際活動軌跡的長度被低估,而且線性插值沒有考慮漁船的航向和速度對其軌跡的影響,所以軌跡重構(gòu)的效能較低。SKAAR等[34]研究發(fā)現(xiàn),當(dāng)數(shù)據(jù)回報頻率為2 h時,采用線性插值方法重構(gòu)的漁船軌跡誤差在3 km以上。
為了得到更為準(zhǔn)確的漁船航行軌跡,許多復(fù)雜的插值方法逐漸被引入,其中最具代表性的是樣條插值(spline interpolation)。樣條插值是利用最小表面曲率的數(shù)學(xué)表達(dá)式來模擬生成能通過所有輸入點的光滑曲線。樣條插值兼顧了計算方法的快捷性和數(shù)據(jù)結(jié)構(gòu)的復(fù)雜性,而且綜合考慮了航向和航速對漁船軌跡的影響,實現(xiàn)了漁船軌跡與VMS數(shù)據(jù)最大程度的結(jié)合。樣條函數(shù)種類繁多,每種樣條函數(shù)有各自的優(yōu)缺點和適用范圍,尋找適合VMS數(shù)據(jù)的樣條函數(shù)插值方法是漁船軌跡重構(gòu)的關(guān)鍵。HINTZEN等[35]首次使用三次赫爾米特樣條插值(cubic hermite spline)的方法來對漁船軌跡進(jìn)行重構(gòu),三次赫爾米特樣條函數(shù)使用時間、位置和切向量來構(gòu)建多項式計算插值點,整個過程分為兩個步驟:(1)計算控制點的切向;(2)計算插值點的位置,其中不同切向量的計算方式會產(chǎn)生不同的軌跡曲線。HINTZEN等[35]分別用兩個多項式來描述經(jīng)度和緯度兩個方向的插值,兩個多項式的切向量通過航向和速度計算得到。這種插值方法重構(gòu)的漁船軌跡誤差較小,且對不同類型的漁船軌跡均有較好的擬合效果。RUSSO等[36]引入了Catmull-Rom插值來重構(gòu)漁船軌跡,并在Catmull-Rom方法的基礎(chǔ)上對切向量的計算公式進(jìn)行了改進(jìn),其切向量通過相鄰兩個點的平均變化率計算得到。
2.1 捕撈努力量估算
捕撈努力量是指在單位時間內(nèi)某種捕撈方式投入捕撈生產(chǎn)的作業(yè)單位數(shù)量,它是漁業(yè)資源學(xué)研究中的一個重要參數(shù)[37]。通過對VMS數(shù)據(jù)的分析挖掘,根據(jù)處于捕撈狀態(tài)的漁船點位數(shù)據(jù),結(jié)合漁船的數(shù)量、噸位、發(fā)動機(jī)功率、作業(yè)方式等信息,可以宏觀、實時的把握水域內(nèi)捕撈努力量的時空分布狀況。目前,國內(nèi)外學(xué)者采用VMS數(shù)據(jù)進(jìn)行捕撈努力量估算的方法主要是運(yùn)用空間分析技術(shù)中的點密度分析方法來估算捕撈努力量的時空分布。
點密度分析是指將漁船的作業(yè)時間分配給各個處于捕撈狀態(tài)的船位點,然后通過密度分析,計算一定分辨率大小的地理網(wǎng)格內(nèi)船位點的數(shù)量來表示捕撈努力量。如張勝茂等[38]根據(jù)北斗衛(wèi)星船位數(shù)據(jù),計算了0.1°×0.1°經(jīng)緯網(wǎng)格內(nèi)累計的捕撈時間,結(jié)合拖網(wǎng)漁船的功率,估算出象山港拖網(wǎng)船捕撈努力量的分布情況。CORO等[39]計算了0.5°×0.5°分辨率下加拿大近海的月捕撈努力量的分布情況,并探討了基于VMS數(shù)據(jù)的捕撈努力量估算業(yè)務(wù)化應(yīng)用的可行性。MILLS等[40]探討了在3 km×3 km的高空間分辨率下,采用VMS數(shù)據(jù)分析在北海作業(yè)的英國拖網(wǎng)漁船捕撈努力量分布的可行性。HINZ等[41]通過鄰域分析,計算了以1 km2大小的網(wǎng)格為中心、以3 km為搜索半徑的區(qū)域內(nèi)所有處于捕撈狀態(tài)船位點的均值,作為該1 km2網(wǎng)格內(nèi)的捕撈努力量。點密度分析可以很好地反映大時空尺度下捕撈努力量的變化趨勢,但分辨率的大小對評估結(jié)果影響很大。分辨率過大容易導(dǎo)致網(wǎng)格內(nèi)捕撈努力量被高估,從而影響評估結(jié)果的精度,因此,選擇合適的空間分辨率是采用點密度分析方法來估算捕撈努力量的關(guān)鍵所在[42]。
2.2 漁民行為特點及漁場分析
為獲取最大的經(jīng)濟(jì)收益,漁民往往尋求目標(biāo)魚類集群的、適宜于捕撈的海域進(jìn)行捕撈作業(yè),這些捕撈努力量密集分布的海域可以定義為漁場[43]。因此,捕撈努力量的時空分布可以反映漁民的行為特點、漁場位置及其動態(tài)變遷。
分析漁民的行為特點是目前VMS應(yīng)用研究的熱點之一。如BERTRAND等[44]采用VMS數(shù)據(jù)重構(gòu)了秘魯鳀(Engraulis ringens)圍網(wǎng)漁船的作業(yè)軌跡,結(jié)合相關(guān)聲學(xué)調(diào)查所獲取的魚群空間分布情況,分析了漁民行為特點與魚群空間分布的關(guān)系。FONSECA等[45]將葡萄牙近海拖網(wǎng)漁船的VMS數(shù)據(jù)與上岸漁獲數(shù)據(jù)相匹配,并引入了計量經(jīng)濟(jì)學(xué)中的離散選擇模型(discrete choice model)來模擬漁船船位動態(tài)變化與漁場變遷間的關(guān)系。JOO等[46]采用VMS數(shù)據(jù)重構(gòu)了秘魯鳀圍網(wǎng)漁船的軌跡,通過聚類分析將漁船狀態(tài)劃分為找魚、捕撈以及航行3種,并分析了不同狀態(tài)下漁民捕撈策略的選擇以及變化情況。
在漁場分析方面,JENNINGS等[47]將VMS數(shù)據(jù)與漁船上岸漁獲數(shù)據(jù)相結(jié)合,綜合考慮漁船的船型、漁具、主捕魚種等因素,將英國西南部海域內(nèi)的漁場定義為:整個捕撈海域內(nèi),捕撈努力量超過總量的10%且所占海域不超過總捕撈海域面積50%的區(qū)域。RUSSO等[48]采用格里菲斯時空自相關(guān)指數(shù)(Griffith’s spatio-temporal index,GSTI)模型分析了基于VMS數(shù)據(jù)的地中海沿岸意大利拖網(wǎng)漁船捕撈努力量時空分布狀況,研究發(fā)現(xiàn),當(dāng)GSTI>0時漁船處于漁場捕撈狀態(tài),此時漁船所處水域即定義為地中海沿岸意大利拖網(wǎng)漁船的作業(yè)漁場。鄒建偉等[49]根據(jù)南海外海廣西燈光罩網(wǎng)漁船的北斗船位監(jiān)控數(shù)據(jù),計算了各漁區(qū)內(nèi)船位點數(shù)量占同期南海外??偞稽c數(shù)量的比例,并按各漁區(qū)內(nèi)漁船生產(chǎn)集中程度將作業(yè)區(qū)域劃分為作業(yè)高密集區(qū)、密集區(qū)、低密集區(qū)和生產(chǎn)外圍區(qū)4類,高密集區(qū)和密集區(qū)的捕撈努力量占總量的2/3以上,為廣西燈光罩網(wǎng)漁船在南海外海的主要漁場。
2.3 捕撈活動對生態(tài)環(huán)境影響分析
漁業(yè)活動對生態(tài)環(huán)境的影響主要表現(xiàn)在兩個方面:對海洋環(huán)境的影響以及對海洋生物資源的影響[50]。評估漁業(yè)活動對海洋環(huán)境影響的一個重要指標(biāo)是捕撈強(qiáng)度,采用VMS數(shù)據(jù)計算單位時間、單位面積水域內(nèi)投入作業(yè)的捕撈努力量,可以得到高時空分辨率的捕撈強(qiáng)度分布狀況。目前,采用VMS數(shù)據(jù)評估漁業(yè)活動對海洋環(huán)境影響的研究多集中于對海洋底棲環(huán)境影響較大的拖網(wǎng)漁業(yè)上,如LAMBERT等[5]估算了英國馬恩島歐洲扇貝(Pecten maximus)底拖網(wǎng)漁船捕撈強(qiáng)度的分布狀況,量化分析了拖網(wǎng)捕撈對海洋底棲環(huán)境的影響。GERRITSEN等[51]網(wǎng)格化計算了愛爾蘭凱爾特海底拖網(wǎng)漁船的捕撈強(qiáng)度,探討了高時空分辨下底拖網(wǎng)捕撈對海洋環(huán)境的影響。HINZ等[41]計算了愛爾蘭坎布里亞海域挪威龍蝦(Nephrops norvegicus)拖網(wǎng)漁船的拖網(wǎng)次數(shù)和拖拽范圍,并與實地海底采樣相結(jié)合,分析了拖網(wǎng)作業(yè)對海洋底棲環(huán)境、資源豐度以及生物多樣性的影響。
VMS數(shù)據(jù)與漁撈日志數(shù)據(jù)、港口上岸漁獲數(shù)據(jù)以及GPS數(shù)據(jù)等相結(jié)合,還可用于分析捕撈活動對海洋生物資源的影響。DENG等[52]將VMS數(shù)據(jù)與漁撈日志數(shù)據(jù)相結(jié)合,分析了拖網(wǎng)捕撈對澳大利亞北部對蝦資源種群損耗的影響。VOTIER等[53]重構(gòu)了英格蘭西南部海域漁船的航行軌跡,并將其與安裝有GPS裝置的塘鵝(Morus bassanus)群飛行軌跡進(jìn)行時空匹配來研究漁船丟棄的漁獲物與塘鵝覓食行為之間的關(guān)系,軌跡匹配以及塘鵝胃含物的穩(wěn)定同位素分析均表明,漁船丟棄的漁獲物是塘鵝食物的重要來源。SANTOS等[54]將在印度洋作業(yè)的葡萄牙延繩釣漁船VMS數(shù)據(jù)、觀察員記錄的漁獲物采樣數(shù)據(jù)以及上岸漁獲物采樣數(shù)據(jù)相結(jié)合,分析了不同捕撈強(qiáng)度下,其主要兼捕種類大青鯊(Prionaceglauca)和尖吻鯖鯊(Isurus oxyrinchus)的釣獲率以及個體大小的時空分布情況。
VMS數(shù)據(jù)以其獨(dú)有的實時性、準(zhǔn)確性、宏觀性等優(yōu)勢,在漁撈日志驗證、漁業(yè)資源評估以及水產(chǎn)品溯源等方面也有著良好的應(yīng)用前景。根據(jù)對VMS數(shù)據(jù)進(jìn)行分析挖掘所獲取的漁船狀態(tài)及軌跡信息,可以驗證漁撈日志中記錄的捕撈作業(yè)位置、卸貨港口信息的準(zhǔn)確性[55];VMS數(shù)據(jù)加入漁業(yè)資源評估模型中,可以更好地反映漁業(yè)資源的時空分布特征,使評估結(jié)果更具真實性[56];VMS數(shù)據(jù)與漁撈日志數(shù)據(jù)、漁獲物銷售數(shù)據(jù)相結(jié)合,可將從捕撈到銷售的整個水產(chǎn)品產(chǎn)業(yè)鏈連接起來,實現(xiàn)對水產(chǎn)品的溯源[57]。另外,根據(jù)北斗導(dǎo)航系統(tǒng)建設(shè)總體規(guī)劃,2020年左右,將建成覆蓋全球的北斗衛(wèi)星導(dǎo)航系統(tǒng)。北斗數(shù)據(jù)具有極高的時空精度,北斗大數(shù)據(jù)的分析和挖掘?qū)⒃跐O業(yè)安全、應(yīng)急救援、環(huán)境監(jiān)測、信息化服務(wù)等方面極大地推動我國海洋事業(yè)的發(fā)展。然而,VMS從上世紀(jì)末出現(xiàn)到如今只有短短二三十年的時間,對VMS數(shù)據(jù)進(jìn)行挖掘和拓展應(yīng)用的時間則更短,諸多方面有待提高。
(1)VMS數(shù)據(jù)應(yīng)用的基礎(chǔ)是通過對漁船船位、航速、航向等信息進(jìn)行挖掘來獲取漁船的捕撈狀態(tài)及航行軌跡,然而由于漁船大小、作業(yè)方式、作業(yè)時間、主捕魚種等因素的不同,甚至海洋環(huán)境、水深、天氣情況等方面的差異也都會導(dǎo)致漁船船位、航速以及航向的變化,因此,設(shè)計更為合適的VMS數(shù)據(jù)挖掘方法與模型仍是今后研究的重點。
(2)船載VMS的通訊系統(tǒng)有Inmarsat-C、ARGOS、AIS、北斗導(dǎo)航系統(tǒng)等,不同系統(tǒng)的VMS數(shù)據(jù)回報頻率標(biāo)準(zhǔn)不一,數(shù)據(jù)間難以實現(xiàn)匹配和兼容,已經(jīng)成為限制VMS數(shù)據(jù)應(yīng)用的一個重要問題,加強(qiáng)國際間在VMS研發(fā)方面的交流與合作,制定統(tǒng)一的VMS國際標(biāo)準(zhǔn)十分有必要。
(3)目前,VMS數(shù)據(jù)在海洋漁業(yè)上的應(yīng)用還處在試驗性的數(shù)據(jù)挖掘階段,今后可以更多的考慮將其與漁撈日志自動采集技術(shù)、無線射頻識別、電子代碼、物聯(lián)網(wǎng)等信息技術(shù)相結(jié)合,在漁業(yè)信息動態(tài)采集、漁海況自動分析、水產(chǎn)品自動溯源等方面進(jìn)行拓展研究和應(yīng)用。
[1]WATSON R,PAULY D.Systematic distortions in world fisheries catch trends[J].Nature,2001(414):534-536.
[2]季 民,靳奉祥,李云嶺,等.遠(yuǎn)洋漁船動態(tài)監(jiān)控系統(tǒng)研究[J].測繪科學(xué),2005,30(5):92-94.
JIM,JIN F X,LIY L,et al.The study of dynamic monitor system of deep sea fishing vessels[J].Science of Surveying and Mapping,2005,30(5):92-94.
[3]張壽桂,彭國均.海監(jiān)船舶導(dǎo)航與監(jiān)控管理信息系統(tǒng)[J].上海海事大學(xué)學(xué)報,2006,27(3):31-35.
ZHANG S G,PENG G J.Information system of navigation andmanagementofmonitor and control for sea supervision ships[J].Journal of Shanghai Maritime University,2006,27(3):31-35.
[4]曹世娟,黃碩琳,郭文路.我國漁業(yè)管理運(yùn)用漁船監(jiān)控系統(tǒng)的探討[J].上海海洋大學(xué)學(xué)報,2002,11(1):89-93.
CAO S J,HUANG S L,GUOW L.Discussion on adopting the vessel monitoring system in Chinese fishery management[J].Journal of Shanghai Fisheries University,2002,11(1):89-93.
[5]LAMBERT G I,JENNINGS S,HIDDINK JG,etal.Implications of using alternative methods of vessel monitoring system(VMS)data analysis to describe fishing activities and impacts[J].Ices Journal of Marine Science,2012,69(4):682-693.
[6]居 禮.北斗衛(wèi)星導(dǎo)航系統(tǒng)在海洋漁業(yè)的應(yīng)用[J].衛(wèi)星與網(wǎng)絡(luò),2013(3):14-22.
JU L.The application of Beidou satellite navigation system in marine fisheries[J].Satellite&Network,2013(3):14-22.
[7]WITTM J,GODLEY B J.A step towards seascape scale conservation:Using vessel monitoring systems(VMS)to map fishing activity[J].Plos One,2007,2(10):e1111.
[8]JONSEN ID,MYERSR A,JAMESM C.Identifying leatherback turtle foraging behavior from satellite telemetry using a switching state-space model[J].Marine Ecology Progress,2007,337(12):255-264.
[9]MURAWSKI S A,WIGLEY S E,F(xiàn)OGARTY M J,et al.Effort distribution and catch patterns adjacent to temperate MPAs[J].Ices Journal of Marine Science,2005,62(6):1150-1167.
[10]FOCK H O.Fisheries in the context of marine spatial planning:Defining principal areas for fisheries in the German EEZ[J].Marine Policy,2008,32(4):728-739.
[11]BERTRAND S,BURGOS JM,GERLOTTO F,et al.Lévy trajectories of Peruvian purse-seiners as an indicator of the spatial distribution of anchovy(Engraulis ringens)[J].Ices Journal of Marine Science,2005,62(3):477-482.
[12]LEE J,SOUTH A B,JENNINGSS,et al.Developing reliable,repeatable and accessible methods to provide high-resolution estimates of fishing effort distributions from vessel monitoring system(VMS)data[J].Ices Journal of Marine Science,2010,67(6):1260-1271.
[13]OVASKAINEN O,HANSKI I.Spatially structured metapopulation models:Global and local assessment of metapopulation capacity[J].Theoretical Population Biology,2001,60(4):281-302.
[14]PATTERSON T A,THOMAS L,WILCOX C,et al.State-space models of individual animal movement[J].Trends in Ecology&Evolution,2008,23(2):87-94.
[15]ROYER F,F(xiàn)ROMENTIN JM,GASPAR P.A statespace model to derive bluefin tuna movement and habitat from archival tags[J].Oikos,2005,109(3):473-484.
[16]JONSEN I D,F(xiàn)LEMMING J M.Meta-analysis of animal movement using state-space models[J].Ecology,2003,84(11):3055-3063.
[17]VERMARD Y,RIVOT E,MAHEVAS S,et al.Identifying fishing trip behavior and estimating fishing effort from VMS data using bayesian hidden markov models[J].Ecological Modelling,2010,221(17):1757-1769.
[18]WALKER E,BEZN.A pioneer validation of a statespace model of vessel trajectories(VMS)with observers’data[J].Ecological Modelling,2010,221(17):2008-2017.
[19]JOO R,BERTRAND S,CHAIGNEAU A,et al.Optimization of an artificial neural network for identifying fishing set positions from VMS data:An example from the Peruvian anchovy purse seine fishery[J].Ecological Modelling,2011,222(4):1048-1059.
[20]張勝茂,楊勝龍,戴 陽,等.北斗船位數(shù)據(jù)提取拖網(wǎng)捕撈努力量算法研究[J].水產(chǎn)學(xué)報,2014,38(8):1190-1199.
ZHANG SM,YANG SL,DAIY,et al.Algorithm of fishing effort extraction in trawling based on Beidou vessel monitoring system data[J].Journal of Fisheries of China,2014,38(8):1190-1199.
[21]CHANG S K.Application of a vessel monitoring system to advance sustainable fisheries management-Benefits received in Taiwan[J].Marine Policy,2011,35(2):116-121.
[22]RIJNSDORP A D,BUYS A M,STORBECK F,et al.Micro-scale distribution of beam trawl effort in the southern North Sea between 1993 and 1996 in relation to the trawling frequency of the sea bed and the impact on benthic organisms[J].Ices Journal of Marine Science,1996,55(3):403-419.
[23]DINMORE T A,DUPLISEA D E,RACKHAM B D,et al.Impact of a large-scale area closure on patterns of fishing disturbance and the consequences for benthic communities[J].Ices Journal of Marine Science,2003,60(2):371-380.
[24]HIDDINK J,JENNINGSS,KAISER M J.Indicators of the ecological impact of bottom-trawl disturbance on seabed communities[J].Ecosystems,2006,9(7):1190-1199.
[25]HIDDINK J,JENNINGS S,KAISER M J,et al.Cumulative impacts of seabed trawl disturbance on benthic biomass,production,and species richness in different habitats[J].Canadian Journal of Fisheries&Aquatic Sciences,2006,63(4):721-736.
[26]PIET G J,QUIRIJNS F J,ROBINSON L,et al.Potential pressure indicators for fishing,and their data requirements[J].Ices Journal of Marine Science,2007,64(1):110-121.
[27]WANG Y,WANG Y B,ZHENG J.Analyses of trawling track and fishing activity based on the data of vessel monitoring system(VMS):A case study of the single otter trawl vessels in the Zhoushan fishing ground[J].Journal of Ocean University of China,2015,14(1):89-96.
[28]FLEMMING J M,MYERS R A,JONSEN I D.Robust state-space modeling of animal movement data[J].Ecology,2005,86(11):2874-2885.
[29]RYAN P G,PETERSAN S L,PETERS G,et al.GPS tracking a marine predator:the effects of precision,resolution and sampling rate on foraging tracks of African Penguins[J].Marine Biology,2004,145(2):215-223.
[30]TREMBLAY Y,SHAFFER S A,F(xiàn)OWLER S L,et al.Interpolation of animal tracking data in a fluid environment[J].Journal of Experimental Biology,2006,209(1):128-140.
[31]HEDGER R D,MARTIN F,DODSON J J,et al.The optimized interpolation of fish positions and speeds in an array of fixed acoustic receivers[J].Ices Journal of Marine Science,2008,65(7):1248-1259.
[32]EASTWOOD PD,MILLSCM,ALDRIDGE JN,et al.Human activities in UK offshore waters:An assessment of direct,physical pressure on the seabed[J].Ices Journal of Marine Science,2007,64(3):453-463.
[33]STELZENMULLER V,ROGERSS I,MILLSC M.Spatio-temporal patterns of fishing pressure on UK marine landscapes,and their implications for spatial planning and management[J].Ices Journal of Marine Science,2008,65(6):1081-1091.
[34]SKAAR K L,JORGENSEN T,ULVESTAD B K H,et al.Accuracy of VMS data from Norwegian demersal stern trawlers for estimating trawled areas in the Barents Sea[J].Ices Journal of Marine Science,2011,68(8):1615-1620.
[35]HINTZEN N T,PIET G J,BRUNEL T.Improved estimation of trawling tracks using cubic hermite spline interpolation of position registration data[J].Fisheries Research,2010,101(1):108-115.
[36]RUSSO T,PARISI A,CATAUDELLA S.New insights in interpolating fishing tracks from VMS data for different métiers[J].Fisheries Research,2011,108(1):184-194.
[37]陳新軍.漁業(yè)資源與漁場學(xué)[M].北京:海洋出版社,2004.
CHEN X J.Fisheries biology and fishing ground[M].Beijing:Ocean Press,2004.
[38]張勝茂,崔雪森,伍玉梅,等.基于北斗衛(wèi)星船位數(shù)據(jù)分析象山拖網(wǎng)捕撈時空特征[J].農(nóng)業(yè)工程學(xué)報,2015,31(7):151-156.
ZHANG SM,CUIX S,WU Y M,et al.Analyzing space-time characteristics of Xiangshan trawling based on Beidou Vessel Monitoring System data[J].Transactions of the Chinese Society of Agricultural Engineering,2015,31(7):151-156.
[39]CORO G,F(xiàn)ORTUNATI L,PAGANO P.Deriving fishing monthly effort and caught species from vessel trajectories[C]//OCEANS.The challenges of the Northern Dimension.Bergen:MTS/IEEE,2013:1-5.
[40]MILLSCM,TOWNSEND SE,JENNINGSS,et al.Estimating high resolution trawl fishing effort from satellite-based vessel monitoring system data[J].Ices Journal of Marine Science,2007,64(2):248-255.
[41]HINZ H,PRIETO V,KAISER M J.Trawl disturbance on benthic communities:Chronic effects and experimental predictions[J].Ecological Applications a Publication of the Ecological Society of America,2009,19(3):761-773.
[42]CHANG S K,YUAN T L.Deriving high-resolution spatiotemporal fishing effort of large-scale longline fishery from vessel monitoring system(VMS)data and validated by observer data[J].Canadian Journal of Fisheries&Aquatic Sciences,2014,71(9):1363-1370.
[43]MARCHAL P,BO A,CAILLART B,et al.Impact of technological creep on fishing effort and fishing mortality,for a selection of European fleets[J].Ices Journal of Marine Science,2007,64(1):192-209.
[44]BERTRAND S,DIAZ E,LENGAIGNE M.Patterns in the spatial distribution of Peruvian anchovy(Engraulis ringens)revealed by spatially explicit fishing data[J].Progress in Oceanography,2008,79(2):379-389.
[45]FONSECA T,CAMPOSA,AFONSO-DIASM,etal.Trawling for cephalopods off the Portuguese coast fleet dynamics and landings composition[J].Fisheries Research,2008,92(2):180-188.
[46]JOO R,SALCEDO O,GUTIERREZM,et al.Defining fishing spatial strategies from VMS data:Insights from the world’s largest monospecific fishery[J].Fisheries Research,2015,164(4):223-230.
[47]JENNINGSS,LEE J.Defining fishing grounds with vessel monitoring system data[J].Ices Journal of Marine Science,2012,69(1):51-63.
[48]RUSSO T,PARISI A,CATAUDELLA S.Spatial indicators of fishing pressure:Preliminary analyses and possible developments[J].Ecological Indicators,2013,26(1):141-153.
[49]鄒建偉,陳立峰,林蔣進(jìn),等.南海外海燈光罩網(wǎng)主要漁場分布及變動研究—基于廣西漁船的生產(chǎn)監(jiān)測統(tǒng)計[J].南方水產(chǎn)科學(xué),2014,10(4):78-84.
ZOU JW,CHEN L F,LIN J J,et al.Analysis on variation&distribution of center fishing ground for light falling-net in offshore of the South China Sea:Based on statistics of fishery surveillance to Guangxi fishing vessels[J].South China Fisheries Science,2014,10(4):78-84.
[50]WATSON R A,CHEUNGW L,ANTICAMARA JA,et al.Global marine yield halved as fishing intensity redoubles[J].Fish&Fisheries,2013,14(4):493-503.
[51]GERRITSEN H D,MINTO C,LORDAN C.How much of the seabed is impacted by mobile fishing gear?Absolute estimates from vessel monitoring system(VMS)point data[J].Ices Journal of Marine Science,2013,70(3):523-531.
[52]DENG R,DICHMONT C,MILTON D,et al.Can vessel monitoring system data also be used to study trawling intensity and population depletion?The example of Australia’s northern prawn fishery.[J].Canadian Journal of Fisheries&Aquatic Sciences,2005,62(12):611-622.
[53]VOTIER S C,BEARHOP S,WITT M J,et al.Individual responses of seabirds to commercial fisheries revealed using GPS tracking,stable isotopes and vessel monitoring systems[J].Journal of Applied Ecology,2010,47(2):487-497.
[54]SANTOSM N,LINOPG,F(xiàn)ERNANDZEC J,et al.Preliminary observations on the by-catch of elasmobranchs caught by the Portuguese longline fishery in the Indian Ocean:Biology,ecology and fishery[C]//IOTC.The 7th Session of the IOTCWorking Party on Ecosystem and Bycatch.Lankanfinolhu:Republic of Maldives,2011:1-14.
[55]PALMER M C,WIGLEY SE.Using positional data from vessel monitoring systems to validate the logbook-reported area fished and the stock allocation of commercial fisheries Landings[J].North American Journal of Fisheries Management,2009,29(4):928-942.
[56]MULLOWNEY D R,DAWE E G.Development of performance indices for the Newfoundland and Labrador snow crab(Chionoecetes opilio)fishery using data from a vessel monitoring system[J].Fisheries Research,2009,100(3):248-254.
[57]張勝茂,唐峰華,張 衡,等.基于北斗船位數(shù)據(jù)的拖網(wǎng)捕撈追溯方法研究[J].南方水產(chǎn)科學(xué),2014,10(3):15-23.
ZHANG S M,TANG F H,ZHANG H,et al.Research on trawling tracing based on Beidou vessel monitoring system data[J].South China Fisheries Science,2014,10(3):15-23.
Advances in mining and application of vessel monitoring system data
GUO Gang-gang1,2,F(xiàn)AN Wei2,ZHANG Sheng-mao2,ZHENG Qiao-ling1,2,WANG Xiao-xuan2
(1.College of Marine Sciences,Shanghai Ocean University,Shanghai 201306,China;2.Key Lab of East China Sea&Oceanic Fishery Resources Exploitation and Utilization,Ministry of Agriculture,East China Sea Fisheries Research Institute,Chinese Academy of Fishery Sciences,Shanghai 200090,China)
Vessel monitoring system,as a fishing vessel monitoring means,provides a new data source for fisheries scientific research.VMS data records the dynamic information of real-time position,speed and heading of fishing vessels,making up for the deficiency of log book data in these aspects,and it has been widely used in marine fisheries.To further understand its advances in analyzing and mining of VMS data,we summarized the methods and models of fishing vessels states recognition and fishing vessels trajectory reconstruction,based on related literatures by researchers at home and abroad.On this basis,according to VMS data in fishing states,combining with the tonnage,engine power,fishing gear,such as information of fishing vessels,it could be used to estimate the spatial-temporal distribution of fishing effort;the spatialtemporal distribution of fishing effort could be used to analyze the behavior characteristics of fishermen,the location of fishing ground and dynamic changes of fishing ground;the fishing intensity was calculated by fishing effort in unit time and unit area,the spatial-temporal distribution of fishing intensity could be used to analyze the impact of fishing activity on the marine environment and marine biological resources.The research of the applications of VMS data in fishing effort estimation,the analysis of fishermen behavior characteristics and fishing grounds,and the impact of fishing activities on marine eco-environment have made great progresses.The technology of analyzing and mining of VMS data has been relatively mature.What’s more,the application prospect of VMS data in marine fisheries is also very wide,but there still has many challenges in mining and application of VMS data,the research emphasis of mining and application of VMS data in the future will include:(1)the methods and models in analyzing and mining of VMS data need to be further perfected;(2)strengthening the international exchanges and cooperation to implement the unity of the recording frequency of VMS data between different communication systems;(3)combining the VMS data with the information technology just like the logbook automatic acquisition technology,radio frequency identification technology,electronic product code technology and internet of things technology,to explore the application of VMS data in the automatic acquisition of fisheries information,the automatic analysis of fisheries information and marine environment and the automatic tracing of aquatic products.
VMS data;fishing effort;fishing ground;eco-environmental impact assessment
S 972.9
A
1004-2490(2016)02-0217-09
2015-07-02
上海市科學(xué)技術(shù)委員會長三角科技聯(lián)合攻關(guān)領(lǐng)域項目(15595811000);中央級公益性科研院所基本科研業(yè)務(wù)費(fèi)專項資金項目(東海水產(chǎn)研究所2014T13)
郭剛剛(1991-),男,碩士研究生。E-mail:gzguogang@126.com
王曉璇(1983-),女,助理研究員。Tel:021-65682395,E-mail:wxx1012@163.com