韓東 王鵬新 張悅 田惠仁 周西嘉
摘要:干旱是影響農(nóng)業(yè)生產(chǎn)的主要氣候因素。傳統(tǒng)的農(nóng)業(yè)干旱監(jiān)測主要是基于氣象和水文數(shù)據(jù),雖然能提供監(jiān)測點上較為精確的干旱監(jiān)測結果,但是在監(jiān)測面上的農(nóng)業(yè)干旱時,仍存在一定的局限。遙感技術的快速發(fā)展,尤其是目前在軌的衛(wèi)星傳感器感測的電磁波段涵蓋了可見光、近紅外、熱紅外和微波等波段,為區(qū)域尺度農(nóng)業(yè)干旱監(jiān)測提供了新的手段。充分利用衛(wèi)星遙感數(shù)據(jù)獲得的豐富地表信息進行農(nóng)業(yè)干旱監(jiān)測和預測具有重要的研究意義。本文從遙感指數(shù)方法、土壤含水量方法和作物需水量方法三個方面闡述了基于衛(wèi)星遙感的農(nóng)業(yè)干旱監(jiān)測研究進展。農(nóng)業(yè)干旱預測是在干旱監(jiān)測的基礎上進行時間軸的預測,本文在總結干旱監(jiān)測進展的基礎上,進一步簡述了以干旱指數(shù)方法和作物生長模型方法為主的農(nóng)業(yè)干旱預測研究進展。
關鍵詞:衛(wèi)星;遙感;農(nóng)業(yè)干旱;作物生長模型;監(jiān)測;預測
中圖分類號:S423;TP79文獻標志碼:A文章編號:202104-SA002
引用格式:韓東,王鵬新,張悅,田惠仁,周西嘉.農(nóng)業(yè)干旱衛(wèi)星遙感監(jiān)測與預測研究進展[J].智慧農(nóng)業(yè)(中英文),2021,3(2):1-14.
HAN Dong, WANG Pengxin, ZHANG Yue, TIAN Huiren, ZHOU Xijia. Progress of agricultural. drought monitoring and forecasting using satellite remote sensing[J]. Smart Agriculture, 2021, 3(2): 1-14. (in Chinese with English abstract)
1引言
干旱是一種頻繁發(fā)生的自然災害現(xiàn)象。Wilhite和Glanze[1]將干旱劃分為氣象干旱、水文干旱、農(nóng)業(yè)干旱和社會經(jīng)濟干旱四種類別。氣象干旱通常表示某一特定時間段內(nèi)降水量偏離正常降水量的程度。水文干旱表示干旱期對地表或地下水文的影響。農(nóng)業(yè)干旱是指某一特定時期內(nèi)土壤水分不能滿足農(nóng)作物正常生長的需求,其表達的是干旱現(xiàn)象對農(nóng)業(yè)生產(chǎn)的影響程度。社會經(jīng)濟干旱指某一特定時期內(nèi),由于自然界水分供應不足而對人類社會經(jīng)濟生活產(chǎn)生影響的現(xiàn)象,通常與氣象干旱、水文干旱和農(nóng)業(yè)干旱有關。
農(nóng)業(yè)干旱作為常見的農(nóng)業(yè)災害之一,主要表現(xiàn)為土壤水分無法正常供給作物生長所需的水分,影響作物對水分的吸收利用,進而影響作物產(chǎn)量[2]。干旱脅迫對于作物的各生理參數(shù)具有不同程度的影響,主要表現(xiàn)為抑制作物根系對水分和養(yǎng)分的吸收,抑制光合作用、蒸騰作用等生理功能,嚴重時甚至導致作物枯萎死亡[3]。目前已經(jīng)有大量方法來監(jiān)測和表征農(nóng)業(yè)干旱,主要包括基于地面站點測量氣象、水文數(shù)據(jù)的方法和基于遙感數(shù)據(jù)的方法[4]。地面站點測量的數(shù)據(jù),比如降水和溫度等,在局部或區(qū)域尺度上的適用性主要取決于地面站點的布設密度和空間分布,這就使得數(shù)據(jù)結果在區(qū)域尺度的應用受到限制[5]。隨著衛(wèi)星傳感器的不斷發(fā)展,獲取大范圍、長時間序列地表監(jiān)測信息的能力逐漸提高。衛(wèi)星傳感器可以感知土壤、植被、溫度等地表信息,將這些信息納入到干旱評估過程是大范圍農(nóng)業(yè)干旱監(jiān)測的有效方法[6,7]。
隨著現(xiàn)代社會經(jīng)濟的快速發(fā)展,農(nóng)業(yè)管理部門不僅要求對農(nóng)業(yè)干旱等災害做出精準的實時監(jiān)測,還要求能實現(xiàn)農(nóng)業(yè)干旱的動態(tài)預警。在這些方面,衛(wèi)星遙感可以提供良好的數(shù)據(jù)源。本文從基于遙感指數(shù)、土壤含水量和作物需水量方法三方面闡述農(nóng)業(yè)干旱遙感監(jiān)測研究進展,并在此基礎上簡述基于遙感干旱指數(shù)和基于作物生長模型方法的農(nóng)業(yè)干旱預測研究進展。
2農(nóng)業(yè)干旱遙感監(jiān)測研究進展
在區(qū)域尺度,基于遙感技術的農(nóng)業(yè)干旱監(jiān)測方法能綜合考慮植被和土壤層信息,因此被廣泛應用于農(nóng)業(yè)干旱監(jiān)測研究[3]。植物生長所需的水分來源于土壤,土壤水分的虧缺直接影響植物的生長發(fā)育,因此通過遙感手段監(jiān)測土壤含水量也能間接感知干旱發(fā)生程度[9,10]。對于農(nóng)業(yè)而言,作物在生長季前的地表覆蓋主要是裸土,而在生長季內(nèi)則主要是植被,此時植被覆蓋下的土壤水分監(jiān)測精度受植被層的影響較大,監(jiān)測難度較高。因此,多源遙感數(shù)據(jù)結合被廣泛應用于區(qū)域尺度的土壤含水量和農(nóng)業(yè)干旱監(jiān)測[11]。此外,農(nóng)業(yè)干旱程度還取決于作物對干旱脅迫的反應狀況。這是由于不同作物對干旱脅迫的響應程度不同,且同一作物在不同生育期對干旱脅迫的響應也具有差異化的表現(xiàn)[12,13]。因此,將作物需水作為出發(fā)點,以作物生長模型為載體,結合遙感對地觀測技術,對作物生育期的水分脅迫進行客觀、詳細地描述是農(nóng)業(yè)干旱監(jiān)測的重要研究方向。
2.1基于遙感指數(shù)的農(nóng)業(yè)干旱監(jiān)測
在基于遙感的農(nóng)業(yè)干旱監(jiān)測方法中,大量研究利用可見光和熱紅外波段反演的遙感干旱指數(shù)估測區(qū)域尺度干旱程度,例如,植被干旱指數(shù)中的植被狀態(tài)指數(shù)(Vegetation Condition Index,VCI)[14]和溫度干旱指數(shù)中的溫度條件指數(shù)(Temperature Condition Index, TCI)[15]。在此基礎上,研究人員基于遙感反演的歸一化植被指數(shù)(Normalized Difference Vegetation Index,NDVI)和地表溫度(Land Surface Temperature,LST)的散點圖呈三角形區(qū)域分布的特征,提出了綜合植被-溫度干旱指數(shù)中條件植被溫度指數(shù)(Vegetation Temperature Condition Index,VTCI)的干旱監(jiān)測方法[16],并將其廣泛地應用于區(qū)域性的農(nóng)業(yè)干旱監(jiān)測[17,18]。此外,介于可見光和熱紅外之間的短波紅外波段對作物葉片含水量更敏感,因此通過短波紅外波段建立的水分干旱指數(shù)也被用來監(jiān)測作物的干旱脅迫狀態(tài),進而評估農(nóng)業(yè)干旱程度,如歸一化差異水分指數(shù)(Normalized Difference Water Index,NDWI)[19],短波紅外水分脅迫指數(shù)(Shortwave Infrared Water Stress Index,SIWSI)[20]。在更長波長的電磁波譜段,基于微波遙感建立的微波干旱指數(shù)也被用來進行農(nóng)業(yè)干旱監(jiān)測。例如,Esch等[21]基于ERS-SAR數(shù)據(jù),提出了農(nóng)作物覆蓋區(qū)域的土壤含水量監(jiān)測指數(shù)(Soil Moisture Index,SMI),研究得出SMI能較好地估計近地表土壤含水量的變化情況。表1列出了主要的遙感干旱監(jiān)測指數(shù)。
2.2基于土壤含水量的農(nóng)業(yè)干旱監(jiān)測
2.2.1可見光-熱紅外數(shù)據(jù)反演土壤含水量
農(nóng)業(yè)干旱遙感監(jiān)測的重要途徑之一是間接監(jiān)測土壤含水量。在作物季前,地表覆蓋主要是裸土。在監(jiān)測土壤含水量時,較早的研究主要分析相關遙感干旱指數(shù)與實測土壤含水量間的相關性,進而建立兩者之間的線性關系。例如,Ghulam等[24]根據(jù)植被在紅-近紅波段的光譜空間特征提出了垂直干旱指數(shù)(Perpendicular Drought Index,PDI),并發(fā)現(xiàn)該指數(shù)與地表實測0~20cm土壤深度的土壤水分具有較好的相關性,能有效監(jiān)測裸土區(qū)域的干旱狀況。此外,熱紅外遙感也被廣泛應用于土壤含水量的監(jiān)測。熱紅外遙感反演土壤含水量的一個重要方法是基于土壤熱慣量法。大量實驗表明,土壤熱慣量與土壤水分的變化密切相關。土壤熱慣量越大,其溫度的變化幅度越小。通過監(jiān)測土壤溫度在某個時間段內(nèi)的變化程度可以定量地推導出熱慣量與土壤含水量之間的關系。對于同一土壤介質(zhì),土壤熱慣量的表達式為:
(25)
其中,P為土壤熱慣量,J/(m2·s·K);k為土壤導熱系數(shù),J/(m·s·K);ρ為土壤密度,kg/m;c為土壤比熱容,J/(kg·K)。通過建立土壤熱慣量與土壤水分間的回歸模型,可以間接有效地監(jiān)測較小區(qū)域內(nèi)的土壤水分狀況。熱慣量法適用于裸土區(qū)的土壤含水量反演,但在植被覆蓋區(qū)域的反演精度較差。劉振華和趙英時[36]在植被覆蓋區(qū)域將地表潛熱通量和顯熱通量引入到熱慣量模型中,從而使熱慣量方法的適用范圍由裸地擴展到植被覆蓋區(qū)域,并提高了土壤水分的監(jiān)測精度。吳黎等[37]提出一種改進的表觀熱慣量計算模型,并基于該模型計算了不同植被覆蓋和不同實驗區(qū)土壤含水量下的熱慣量值,結果表明該模型在植被覆蓋度較低的情況下(NDVI≤0.35)具有較高的土壤含水量監(jiān)測精度。
在作物生長季內(nèi),植被覆蓋程度較高,此時基于可見光-熱紅外遙感數(shù)據(jù)反演的遙感干旱指數(shù)與土壤水分具有密切關系。相關研究通過分析干旱指數(shù)與實測土壤含水量間的相關性,進而建立兩者之間的回歸模型,并將其應用于作物覆蓋下的土壤含水量的反演。例如,Sun等[38]分析了中國關中平原土壤表層水分與VTCI之間的相關性,結果表明,在冬小麥生長季,旬尺度的VTCI與0~10cm和0~20cm土壤深度土壤水分之間均存在顯著的線性相關性。
雖然VTCI與土壤水分間的較強相關性可以使得VTCI直接用來反演土壤含水量,但VTCI作為綜合植被-溫度干旱指數(shù)的一種,要想較為準確地確定VTCI在某一應用區(qū)域的冷熱邊界卻并不容易。楊永民等[39]使用ASTER數(shù)據(jù)對比了基于TVDI方法和基于蒸散比/潛在蒸散比方法的土壤含水量反演結果,認為雖然TVDI方法簡單易用且不需要額外的氣象數(shù)據(jù),但是區(qū)域干、濕的變化會導致其特征空間干、濕邊界的確定出現(xiàn)誤差,進而為土壤水分估算引入誤差,而基于蒸散比/潛在蒸散比的土壤水分估算方法會在一定程度改善TVDI方法估算的經(jīng)驗性,提高土壤水分的反演精度。為了降低在構建植被-溫度綜合干旱指數(shù)時,冷、熱邊界確定困難帶來的土壤含水量反演難題,研究人員從另一角度出發(fā),不再考慮地表溫度(熱紅外數(shù)據(jù))對遙感干旱監(jiān)測的影響,直接基于可見光和近紅外遙感數(shù)據(jù)來反演土壤含水量。例如,Ghulam等[40]在PDI的基礎上考慮了植被層的影響并引入植被覆蓋度,提出了改進型的垂直干旱指數(shù)(Modified Perpendicular Drought Index,MPDI),該指數(shù)在植被生長中后期與地表實測0~20cm土壤深度的土壤含水量具有較高的相關性,可有效監(jiān)測作物生長季內(nèi)的干旱程度。
2.2.2微波數(shù)據(jù)反演土壤含水量
相比可見光和熱紅外譜段,合成孔徑雷達衛(wèi)星發(fā)射的電磁波對地表水分的變化更為敏感。在作物季前,通過分析雷達后向散射信息與地面實測土壤含水量之間的相關性,可基于經(jīng)驗模型的方法反演裸土區(qū)土壤含水量,在此基礎上,基于機器學習方法的土壤含水量估計被廣泛研究[41]。此外,一些基于半經(jīng)驗模型和物理模型的方法也被廣泛研究并用于反演裸土區(qū)的土壤含水量,并取得了較為理想的結果[42,43]。然而,在作物生長季內(nèi),監(jiān)測有植被覆蓋的農(nóng)作物種植區(qū)域的土壤含水量時,由于雷達信號受到土壤層和作物冠層的混合影響,增加了土壤含水量的監(jiān)測難度。為解決這一問題,研究人員通過引入雷達散射模型,消除植被層對雷達信號的干擾,進而提高土壤含水量的反演精度。以MIMICS模型(Michigan Microwave Canopy Scattering Model)為代表的植被散射模型被廣泛應用于土壤含水量的反演研究,并取得了較為理想的結果[44,45]。然而,由于MIMICS模型對植被層的刻畫較為細致,導致其模型結構較為復雜,一般只適用于高大植被覆蓋下的土壤含水量反演[46]。對于農(nóng)作物來說,其植被層一般較為低矮,內(nèi)部結構簡單,因此可基于簡化的植被貢獻模型來進行土壤含水量反演。在此基礎上,水云模型(Water Cloud Model,WCM)被提出,它簡化了植被層與土壤層之間復雜的雷達散射效應,并假定植被層是均勻介質(zhì)且定義了一個參數(shù)來描述植被層的特征[47]。因此,該模型適用于農(nóng)作物覆蓋下的土壤含水量反演。目前,在基于水云模型反演農(nóng)作物覆蓋下的土壤含水量研究中,以雷達衛(wèi)星數(shù)據(jù)與光學衛(wèi)星數(shù)據(jù)相結合的研究居多[48,49]。其中,光學衛(wèi)星數(shù)據(jù)一般被用來反演水云模型中描述植被層的特征參數(shù)。考慮到光學遙感數(shù)據(jù)易受云霧天氣的影響,導致其在作物生育期內(nèi)的獲取具有不穩(wěn)定性。Han等[50]將WCM與一個簡單的產(chǎn)量估計(Simple Algorithm For Yield estimate,SAFY)模型結合,利用Sentinel-2衛(wèi)星數(shù)據(jù)反演的葉面積指數(shù)(Leaf Area Index,LAI)作為SAFY模型的狀態(tài)變量并模擬了冬小麥生育期內(nèi)水云模型中植被描述參數(shù)(LAI)的日變化,有效地解決了由于光學數(shù)據(jù)可用性差造成的水云模型中植被描述參數(shù)不能準確描述雷達衛(wèi)星過境時地表植被狀態(tài)的問題,從而提高了土壤含水量反演的精度。表2列出了微波遙感反演土壤含水量的主要模型。
土壤水分的變化會影響土壤介電常數(shù)的大小,使得微波比輻射率隨之發(fā)生變化,從而導致被動微波傳感器記錄的地表亮溫發(fā)生變化。因此,可通過被動微波遙感記錄地表亮溫來監(jiān)測土壤的熱輻射,從而間接監(jiān)測土壤含水量。即首先通過輻射傳輸模型建立亮溫與土壤介電常數(shù)間的關系,然后通過介電混合模型建立土壤介電常數(shù)與土壤含水量間的關系,最后即可實現(xiàn)土壤含水量的反演[56]。Mo等[57]提出的t-w模型是大多數(shù)輻射傳輸模型的基礎,該模型是零階輻射傳輸模型,包含兩個參數(shù),一個是植被光學厚度(Vegetation Optical. Depth,VOD),另一個是單次散射反照率。針對植被覆蓋下的土壤含水量反演,一個關鍵的環(huán)節(jié)是有效估計VOD參數(shù)。研究發(fā)現(xiàn),通過建立植被含水量與VOD間的經(jīng)驗線性關系可以估計VOD[58]。因此,利用可見光-近紅外遙感數(shù)據(jù)反演與植被含水量相關的植被指數(shù)可以間接實現(xiàn)VOD的估計。此外,也有研究嘗試綜合多因子數(shù)據(jù)建立亮溫數(shù)據(jù)與土壤水分間的回歸模型反演地表土壤含水量[59]。隨著被動微波傳感器的快速發(fā)展和廣泛應用,使用多頻率、多角度的雙極化亮溫數(shù)據(jù)可降低以往利用單傳感器數(shù)據(jù)反演土壤含水量時模型的不確定性。被動微波傳感器相比主動微波傳感器普遍具有較大的幅寬,因此適合于全球或區(qū)域尺度的土壤水分監(jiān)測。表3列出了主要的基于微波遙感反演的土壤水分含量產(chǎn)品。其中,基于被動微波遙感數(shù)據(jù)的土壤水分含量產(chǎn)品較多,在將被動微波遙感數(shù)據(jù)與主動微波遙感數(shù)據(jù)相結合后,其反演的土壤水分含量產(chǎn)品的空間分辨率可由約30km提高到約3km。
2.3基于作物需水量的農(nóng)業(yè)干旱監(jiān)測
2.3.1基于作物冠層含水量的農(nóng)業(yè)干旱監(jiān)測
通過監(jiān)測作物對水分需求的變化情況可有效反映當前農(nóng)業(yè)干旱程度。監(jiān)測作物水分需求的直接手段是監(jiān)測作物冠層含水量。最簡單的方法是通過建立相關光譜指數(shù)與植被冠層含水量之間的回歸模型進行植被含水量的反演[67]。例如,Gao[68]基于NDWI建立了植被冠層水分遙感估測模型并取得了較好的結果。隨著機器學習算法的廣泛應用,研究人員通過直接建立多個光譜波段與植被含水量間的回歸模型可充分挖掘遙感影像的多維光譜特征,有效提高植被含水量的反演精度。作物受到水分脅迫會使其冠層溫度發(fā)生變化,因此結合可見光-熱紅外數(shù)據(jù)監(jiān)測作物的冠層含水量變化情況對評估農(nóng)業(yè)干旱程度具有重要意義。Gerhards等[69]對基于多光譜/高光譜熱紅外遙感的作物水分脅迫監(jiān)測進行了詳細總結。此外,微波數(shù)據(jù)也被逐漸應用到植被含水量的反演研究中。一些研究嘗試結合光學與微波遙感數(shù)據(jù)進行作物冠層含水量的反演[70]。與基于統(tǒng)計模型的冠層含水量反演方法相比,基于輻射傳輸模型的作物水分監(jiān)測方法更具機理性。以往研究中較為常用的輻射傳輸模型有PROSPECT(The Leaf Optical. Properties Spectra)葉片模型、SAIL(Scattering by Arbitrarily Inclined Leaves)冠層模型和PROSPECT+SAIL耦合的葉片-冠層模型。吳伶等[71]通過耦合葉片輻射傳輸模型——PROSPECT模型和冠層輻射傳輸模型——SAIL模型,并以植被指數(shù)NDWI作為優(yōu)化比較對象來反演植被含水量,有效地提高了植被含水量的反演精度。隨著高光譜衛(wèi)星數(shù)據(jù)的廣泛應用,研究人員將輻射傳輸模型與高光譜數(shù)據(jù)相結合,進一步提高了作物含水量的反演精度[72,73]。
2.3.2基于作物生長模型的農(nóng)業(yè)干旱監(jiān)測
作物在生長季內(nèi)受到干旱脅迫的影響會使其生理功能受到抑制,導致其生理參數(shù)發(fā)生變化,這是遙感技術監(jiān)測農(nóng)業(yè)干旱的理論依據(jù)。作物生長發(fā)育受到的水分脅迫程度在相對較大的區(qū)域是變化的,而且作物在不同生育階段受到不同程度的干旱脅迫其生長狀況也會不同[12]。在作物相同生育時期,不同程度干旱脅迫下作物生長狀況也存在差異。在進行區(qū)域尺度農(nóng)業(yè)干旱監(jiān)測時,遙感數(shù)據(jù)反演的干旱指數(shù)受時空條件的限制,一般具有較低的時間分辨率,然而作物在水分虧缺發(fā)生后,較短時間內(nèi)就會出現(xiàn)缺水生理反應。因此,這些指數(shù)可能無法反映作物在關鍵生長階段缺水導致的產(chǎn)量損失。而基于作物需水量和土壤供水日步長的干旱指數(shù)能更為詳細地描述干旱脅迫對作物生長的影響。在評估水分虧缺對植物生長的影響時,作物生長模型可以捕獲作物生理學和水耗竭之間的響應,能夠考慮到干旱脅迫對葉片生長過程的影響。因此,相比單獨使用遙感干旱指數(shù),作物生長模型與遙感數(shù)據(jù)相結合進行農(nóng)業(yè)干旱脅迫監(jiān)測將更具優(yōu)勢。一般來說,基于過程的作物生長模型將作物特性、土壤特性和環(huán)境條件對作物生長和產(chǎn)量形成的影響結合起來進行作物生長模擬,可以在不同區(qū)域、不同生長季節(jié)進行廣泛應用[74]。例如,相比四個主要生育期的VTCI,旬尺度的VTCI與土壤含水量之間具有強的相關性且能更全面、準確地反映作物主要生育期內(nèi)的干旱變化[25]。因此,旬尺度的VTCI可以作為反演土壤水分的變量,并通過間接法將其與作物生長模型相結合,即首先建立VTCI與模型某一中間變量(如土壤含水量)之間的線性關系,然后借助該中間變量進行模型運轉。例如,Xie等[75]利用Landsat衛(wèi)星數(shù)據(jù)反演的旬尺度VTCI線性估計土壤含水量,將估算的土壤含水量和Landsat衛(wèi)星數(shù)據(jù)反演的LAI作為CERES- Wheat模型同化過程中的觀測值,并在冬小麥主要生育期同化LAI和土壤含水量,提高了干旱脅迫的模擬精度。雖然間接法初步解決了將遙感干旱監(jiān)測結果與作物生長模型結合的問題,然而間接法容易引入外部誤差,且在考慮遙感干旱監(jiān)測結果與作物生長模型間存在的時間尺度異化問題時還較為粗略。
大量研究基于數(shù)據(jù)同化算法,結合遙感數(shù)據(jù)與作物生長模型進行農(nóng)業(yè)干旱監(jiān)測,提高了農(nóng)業(yè)干旱的動態(tài)監(jiān)測水平[74]。雖然一些經(jīng)典作物生長模型(如DSSAT模型和WOFOST模型)在結構方面具有很強的機理性,均能較為細致地描述作物生長過程,并能對影響作物生長的相關脅迫因子的變化進行動態(tài)模擬[76]。但是在用衛(wèi)星遙感數(shù)據(jù)結合作物生長模型解決區(qū)域尺度的干旱監(jiān)測問題時,這些模型的應用受到一定程度的限制。主要是因為這類復雜模型需要較多的輸入數(shù)據(jù),包括詳細的農(nóng)業(yè)氣象數(shù)據(jù)、品種數(shù)據(jù)和田間管理數(shù)據(jù)等。而這些參數(shù)在區(qū)域尺度是很難獲取的,這就導致模型很難與遙感數(shù)據(jù)結合進行大范圍的農(nóng)業(yè)干旱監(jiān)測。
相比較為復雜的作物模型,近年來,一些能夠模擬作物干旱脅迫狀態(tài)的結構較為簡單的作物生長模型被提出,使得區(qū)域尺度的農(nóng)業(yè)干旱遙感監(jiān)測變得可能,其中比較具有代表性的有AquaCrop模型[77]和SAFY-WB模型[78]。目前,這兩種模型得到了研究人員越來越多的關注,在區(qū)域尺度研究,特別是遙感數(shù)據(jù)同化方面,具有很好的應用前景。AquaCrop模型是一個水分驅動模型,可以準確地描述不同水分脅迫條件下主要草本作物的產(chǎn)量與需水量之間的關系。該模型通過模擬作物蒸騰作用,并使用標準化作物水分生產(chǎn)力(Normalized Crop Water Productivity,NCWP)將日蒸騰量轉化為作物的日生長量。NCWP的引入使得AquaCrop模型可以在不同地點和生長季節(jié)進行應用。SAFY-WB模型由SAFY模型[79]與FAO水平衡模型[80]結合構成。在原始SAFY模型中用來衡量作物受到的干旱脅迫程度的參數(shù)因子為一固定值,但實際上作物在生長季內(nèi)受到的干旱脅迫與農(nóng)業(yè)干旱程度密切相關。在作物生長季內(nèi),農(nóng)業(yè)干旱經(jīng)常是變化的。因此,SAFY-WB模型通過引入動態(tài)的水分脅迫系數(shù),能有效模擬作物在生長過程中的干旱脅迫動態(tài)變化情況。此外,Silvestro等[81]分析了AquaCrop模型和SAFY-WB模型在干旱脅迫條件下模擬冬小麥產(chǎn)量的敏感性,發(fā)現(xiàn)在區(qū)域尺度下,AquaCrop模型在不同缺水環(huán)境下的作物生長模擬表現(xiàn)要強于SAFY-WB模型,但相比SAFY-WB模型,其對參數(shù)校準的要求更嚴格。
3農(nóng)業(yè)干旱預測研究進展
基于衛(wèi)星遙感預測農(nóng)業(yè)干旱有兩種方法,一種方法是在干旱監(jiān)測的基礎上,通過干旱時空預測模型對未來時間段內(nèi)的農(nóng)業(yè)干旱狀況進行模擬;另一種方法是在作物生長模型的基礎上,改進其水分脅迫模塊,構建作物干旱監(jiān)測模型,將遙感觀測作為同化干旱脅迫的中間變量,并結合短、中、長期氣象數(shù)據(jù)進行農(nóng)業(yè)干旱預測。
3.1基于干旱指數(shù)的農(nóng)業(yè)干旱預測
基于遙感干旱指數(shù)預測農(nóng)業(yè)干旱具有重要研究價值。這類研究主要是以時間序列遙感干旱指數(shù)作為輸入數(shù)據(jù)并基于時序分析等方法預測未來時間段內(nèi)的干旱變化。例如,韓萍等[82]運用求和自回歸移動平均(Auto Regressive Integrated Moving Average,ARIMA)模型對VTCI時空序列進行分析建模并開展冬小麥生長季內(nèi)干旱分析預測,結果表明基于該模型的1~2步預測可以較好地預測區(qū)域干旱變化情況。李俐等[83]應用ARIMA模型和季節(jié)性求和自回歸移動平均(Seasonal. Auto Regressive Integrated Moving Average,SARIMA)模型對夏玉米生長季內(nèi)的VTCI進行建模預測,結果表明ARIMA模型具有比SARIMA模型更高的VTCI預測精度,且基于ARIMA模型的VTCI 1~3步預測在多個年份間具有較穩(wěn)定的精度表現(xiàn)。
歷史干旱數(shù)據(jù)呈現(xiàn)大數(shù)據(jù)特征,人工智能算法可有效挖掘歷史年份的數(shù)據(jù)特征,進一步提高干旱預測精度。近年來,研究人員開始基于神經(jīng)網(wǎng)絡和深度學習的方法進行干旱預測,取得了較好的結果[84,85]。隨著遙感技術的快速發(fā)展,遙感反演的面上干旱指數(shù)本身已具備空間大數(shù)據(jù)特征,此外隨著多衛(wèi)星傳感器的組合使用,遙感干旱指數(shù)也具有越來越高的時間維度。目前,針對遙感干旱指數(shù),基于機器學習算法的農(nóng)業(yè)干旱預測研究還較少,這將是未來的一個研究熱點。
3.2基于作物生長模型的農(nóng)業(yè)干旱預測
農(nóng)業(yè)干旱預測的落腳點為預測干旱對作物長勢的影響程度。從這一角度考慮,基于作物生長模型的農(nóng)業(yè)干旱預測方法具有重要研究價值。作物生長模型依靠氣象數(shù)據(jù)進行驅動,通過引入未來一段時間內(nèi)的氣象預報數(shù)據(jù),可以有效模擬作物在未來時間段的生長狀態(tài)并預報作物的干旱脅迫狀態(tài)。此外,將作物生長模擬與農(nóng)業(yè)干旱監(jiān)測相結合,對作物生長模型進行改進,以實現(xiàn)對農(nóng)業(yè)干旱的監(jiān)測與預警[86]。例如,吳熠婷等[87]利用天氣發(fā)生器LarsWG5.5模擬未來時間段內(nèi)的氣象數(shù)據(jù)并將其輸入到校準后的作物生長模型,進而預測氣候變化條件下冬小麥產(chǎn)量并評估減產(chǎn)風險。遙感觀測可以及時反映地表的瞬時狀態(tài),有效監(jiān)測農(nóng)業(yè)干旱程度。對于上述干旱預測模型而言,引入相關遙感觀測量可以進一步提高模型的預測能力。因此,將氣象、水利和農(nóng)業(yè)農(nóng)村部門提供的中長期氣象數(shù)據(jù)作為這些模型的輸入數(shù)據(jù),運用數(shù)據(jù)同化技術,耦合遙感觀測量(如土壤水分)與模型模擬值,可以有效提高模型的農(nóng)業(yè)干旱預測能力[88,89]。例如,王治海等[90]基于改進后的ARID CROP模型,利用AMSR-E傳感器獲取的區(qū)域農(nóng)田水分信息作為模型中間變量,從而預測農(nóng)業(yè)干旱的動態(tài)變化,結果表明將遙感觀測信息引入改進后的作物生長模型能有效提高冬小麥生長發(fā)育的預測能力和區(qū)域農(nóng)業(yè)干旱的預測精度。
4仍需解決的問題與展望
雖然,目前基于衛(wèi)星遙感數(shù)據(jù)的農(nóng)業(yè)干旱監(jiān)測預測研究在很多方面已經(jīng)取得了大量突破,然而,相關領域的研究還存在一些需要解決的問題。
(1)多個遙感干旱監(jiān)測指數(shù)的結合可提高單個干旱指數(shù)的監(jiān)測誤差。目前這種綜合考慮作物不同生育期干旱指數(shù)的合成指數(shù)并不多,也缺乏與作物機理相關的指數(shù)。因此,需要結合多種指數(shù),從機理上考慮干旱脅迫的影響因素,開發(fā)綜合指數(shù)提高干旱監(jiān)測精度。
(2)基于中等空間分辨率遙感數(shù)據(jù)反演的遙感干旱監(jiān)測指數(shù)已被廣泛研究,且被證實能有效表征區(qū)域的干旱程度。近年來,隨著高等空間分辨率遙感衛(wèi)星的投入使用,在獲取時間序列更高空間分辨率遙感干旱監(jiān)測指數(shù)時,研究人員探索融合中等和高等空間分辨率遙感數(shù)據(jù),以得到降尺度后的時間序列遙感干旱監(jiān)測指數(shù)。同時也嘗試基于多個同質(zhì)遙感數(shù)據(jù)源,對其包含的數(shù)據(jù)進行整合,以得到時間序列遙感干旱監(jiān)測指數(shù)。然而,目前融合異質(zhì)遙感數(shù)據(jù)的干旱監(jiān)測指數(shù)研究目前總體還偏少,相關方法還比較初步。未來可嘗試探索利用深度學習技術深入挖掘異質(zhì)遙感數(shù)據(jù)間的特征信息,構建相關干旱監(jiān)測指數(shù),并建立農(nóng)業(yè)干旱動態(tài)監(jiān)測系統(tǒng),推動智慧農(nóng)業(yè)發(fā)展。
(3)融合多源遙感數(shù)據(jù)的干旱監(jiān)測方法能較好地表征植被層和土壤層的綜合干旱狀態(tài),進而獲得較為理想的農(nóng)業(yè)干旱監(jiān)測結果。在此背景下,不同數(shù)據(jù)源間的時空一致性問題需要進一步考慮。例如,在以往的研究中對基于水云模型并結合光學和雷達影像反演土壤含水量已經(jīng)進行了詳細的研究和分析,但是在一個相對較大的研究區(qū)域,特別是我國南方地區(qū),光學影像在大多數(shù)時間內(nèi)都易受多云天氣的影響,數(shù)據(jù)源的可用性并不是很高,限制了遙感數(shù)據(jù)的應用。在這種情況下,及時有效地確定水云模型的植被描述參數(shù)非常重要,這就需要研究人員考慮不同類型遙感數(shù)據(jù)源獲取日期間的差異對干旱監(jiān)測結果的影響,目前針對這方面的研究還不是很多。未來可針對這一問題探索使用同系統(tǒng)遙感數(shù)據(jù)源,例如Sentinel系列衛(wèi)星搭載了涵蓋微波-可見光-熱紅外的有效載荷,其各個數(shù)據(jù)源具備相近的空間和時間分辨率,各個數(shù)據(jù)源間的協(xié)同性很高,且具備全球范圍的觀測能力。因此,可探索利用深度學習方法獲取更廣泛數(shù)據(jù)源之間的互補信息,以得到高等空間和時間分辨率的遙感干旱監(jiān)測結果。
(4)農(nóng)業(yè)干旱作為重要的環(huán)境因子對作物生長具有重要影響,在利用數(shù)據(jù)同化方法將其與作物生長模型結合進行作物產(chǎn)量估測時,現(xiàn)有研究大多采取間接法。為減少外部誤差的引入,同時考慮遙感干旱監(jiān)測結果與作物生長模型間存在的時間尺度異化問題,未來可探索采取直接法將干旱監(jiān)測結果與作物生長模型相結合,通過引入深度學習方法,同步時間序列干旱監(jiān)測結果與作物生長模型的時間尺度,針對作物在各個生育時期的生理特點,更為科學地考慮土壤層和植被層水分脅迫對作物生長的階段化和差異化的影響,從而提高利用作物生長模型的干旱動態(tài)監(jiān)測精度。
5總結
衛(wèi)星遙感技術的快速發(fā)展使得針對衛(wèi)星數(shù)據(jù)的農(nóng)業(yè)干旱監(jiān)測研究不斷深入,同時也促使基于衛(wèi)星遙感數(shù)據(jù)的農(nóng)業(yè)干旱監(jiān)測逐步市場化。本文以衛(wèi)星遙感的農(nóng)業(yè)干旱監(jiān)測為目標,重點對其研究進展進行了闡述,并在此基礎上簡述了衛(wèi)星遙感的農(nóng)業(yè)干旱預測研究進展。目前,國產(chǎn)遙感衛(wèi)星數(shù)據(jù)已呈現(xiàn)大數(shù)據(jù)特征,基于人工智能的信息自動獲取和解譯技術將使得遙感數(shù)據(jù)能夠更加廣泛有效地應用于農(nóng)業(yè)干旱監(jiān)測領域。同時,針對海量遙感數(shù)據(jù)源,將深度學習技術和作物生長模型有機結合起來,基于數(shù)據(jù)同化思想,深入探索衛(wèi)星遙感在農(nóng)業(yè)干旱動態(tài)遙感監(jiān)測方面的潛力,可進一步推動智慧農(nóng)業(yè)的發(fā)展。
參考文獻:
[1]WILHITE D A, GLANTZ M H. Understanding: The drought phenomenon: The role of definitions[J]. Water International, 1985, 10(3): 111-120.
[2]劉戰(zhàn)東,張凱,米兆榮,等.不同土壤容重條件下水分虧缺對作物生長和水分利用的影響[J].水土保持學報,2019, 33(2): 117-122.
LIU Z, ZHANG K, MI Z. Effects of water deficit on crop growth and water use under different soil bulk densities[J]. Journal. of Soil and Water Conservation, 2019,33(2): 117-122.
[3]潘曉迪,張穎,邵萌,等.作物根系結構對干旱脅迫的適應性研究進展[J].中國農(nóng)業(yè)科技導報,2017, 19(2): 51-58.
PAN X, ZHANG Y, SHAO M, et al. Research progress on adaptive responses of crop root structure to drought stress[J]. Journal. of Agricultural. Science and Technology, 2017, 19(2): 51-58.
[4]劉高鳴,謝傳節(jié),何天樂,等.基于多源數(shù)據(jù)的農(nóng)業(yè)干旱監(jiān)測模型構建[J].地球信息科學學報,2019, 21 (11): 1811-1822.
LIU G, XIE C, HE T, et al. Agricultural. drought monitoring model constructing based on multi-source data[J]. Journal. of Geo-Information Science, 2019, 21 (11): 1811-1822.
[5]馬奕,白磊,李倩,等.區(qū)域氣候模式在中國西北地區(qū)氣溫和降水長時間序列模擬的誤差分析[J].冰川凍土,2016, 38(1): 77-88.
MA Y, BAI L, LI Q, et al. The error analysis of the long term air temperature and precipitation in Northwest China simulated by WRF model[J]. Journal. of Glaciology and Geocryology, 2016, 38(1): 77-88.
[6]路中,雷國平,馬泉來,等.基于重構的Landsat8時間序列數(shù)據(jù)和溫度植被指數(shù)的區(qū)域旱情監(jiān)測[J].水土保持研究,2018, 25(5): 371-377, 384.
LU Z, LEI G, MA Q, et al. Regional. drought monitoring based on reconstructed Landsat 8 data and temperature vegetation index[J]. Research of Soil and Water Conservation, 2018, 25(5): 371-377, 384.
[7]姚遠,陳曦,錢靜.遙感數(shù)據(jù)在農(nóng)業(yè)旱情監(jiān)測中的應用研究進展[J].光譜學與光譜分析,2019, 39(4): 15-22.
YAO Y, CHEN X, QIAN J. Advance in agricultural. drought monitoring using remote sensing data[J]. Spectroscopy and Spectral. Analysis, 2019, 39(4): 15-22.
[8]張有智,解文歡,吳黎,等.農(nóng)業(yè)干旱災害研究進展[J].中國農(nóng)業(yè)資源與區(qū)劃,2020, 41(9): 182-188.
ZHANG Y, XIE W, WU L, et al. Research progress on agriculture drought disaster[J]. Chinese Journal. of Agricultural. Resources and Regional. Planning, 2020, 41 (9): 182-188.
[9]趙春江.農(nóng)業(yè)遙感研究與應用進展[J].農(nóng)業(yè)機械學報,2014, 45(12): 277-293.
ZHAO C. Advances of research and application in remote sensing for agriculture [J]. Transactions of the CSAM, 2014, 45(12): 277-293.
[10]張偉楊,錢希旸,李銀銀,等.土壤干旱對小麥生理性狀和產(chǎn)量的影響[J].麥類作物學報,2016, 36(4):491-500.
ZHANG W, QIAN X, LI Y, et al. Effect of soil drought on the physiological. traits and grain yield of wheat[J]. Journal. of Triticeae Crops, 2016, 36(4): 491-500.
[11]陳仲新,任建強,唐華俊,等.農(nóng)業(yè)遙感研究應用進展與展望[J].遙感學報,2016, 20(5): 748-767.
CHEN Z, REN J, TANG H, et al. Progress and perspectives on agricultural. remote sensing research and applications in China[J]. Journal. of Remote Sensing, 2016,20(5): 748-767.
[12]姚寧,宋利兵,劉健,等.不同生長階段水分脅迫對旱區(qū)冬小麥生長發(fā)育和產(chǎn)量的影響[J].中國農(nóng)業(yè)科學,2015,48(12):2379-2389.
YAO N, SONG L, LIU J, et al. Effects of water stress at different growth stages on the development and yields of winter wheat in arid region[J]. Scientia Agricultura Sinica, 2015, 48(12): 2379-2389.
[13]張建平,何永坤,王靖,等.不同發(fā)育期干旱對玉米籽粒形成與產(chǎn)量的影響模擬[J].中國農(nóng)業(yè)氣象,2015,36(1): 43-49.
ZHANG J, HE Y, WANG J, et al. Impact simulation of drought at different growth stages on grain formation and yield of maize[J]. Chinese Journal. of Agrometeorology, 2015, 36(1):43-49.
[14] LIU W T, KOGAN F N. Monitoring regional. drought using the vegetation condition index[J]. International. Journal. of Remote Sensing, 1996, 17(14): 2761-2782.
[15] KOGAN F N. Global. drought watch from space[J]. Bulletin of the American Meteorological. Society,1997, 78(4): 621-636.
[16]王鵬新,龔健雅,李小文.條件植被溫度指數(shù)及其在干旱監(jiān)測中的應用[J].武漢大學學報(信息科學版),2001,26(5):412-418.
WANG P, GONG J, LI X. Vegetation-temperature condition index and its application for drought monitoring[J], Geomatics and Information Science of Wuhan University, 2001, 26(5): 412-418.
[17] TIAN M, WANG P, KHAN J. Drought forecasting with vegetation temperature condition index using ARIMA models in the Guanzhong Plain[J]. Remote Sensing, 2016, 8(9): 690-708.
[18] ZHOU X, WANG P, TANSEY K, et al. Drought monitoring using the Sentinel-3-based multiyear vegetation temperature condition index in the Guanzhong Plain,China[J]. IEEE Journal. of Selected Topics in Applied Earth Observations and Remote Sensing, 2020, 13: 129-142.
[19] GAO B. NDWI—A normalized difference water index for remote sensing of vegetation liquid water from space[J]. Remote Sensing of Environment, 1995, 58 (3): 257-266.
[20] FENSHOLT R, SANDHOLT I. Derivation of a shortwave infrared water stress index from MODIS near-and shortwave infrared data in a semiarid environment[J]. Remote Sensing of Environment, 2003, 87(1): 111-121.
[21] ESCH S, KORRES W, REICHENAU T, et al. Soil moisture index from ERS-SAR and its application to the analysis of spatial. patterns in agricultural. areas[J]. Journal. of Applied Remote Sensing, 2018, 12(2): ID 022206.
[22] HUETE A, DID AN K, MIURA T, et al. Overview of the radiometric and biophysical. performance of the MODIS vegetation indices[J]. Remote Sensing of Environment, 2002, 83(1-2): 195-213.
[23]陳維英,肖乾廣,盛永偉.距平植被指數(shù)在1992年特大干旱監(jiān)測中的應用[J].環(huán)境遙感,1994, 9(2): 106-112.
CHEN W, XIAO Q, SHENG Y. Application of the anomaly vegetation index to monitoring heavy drought in 1992[J]. Remote Sensing of Environment (China), 1994, 9(2): 106-112.
[24] GHULAM A, QIN Q, ZHAN Z. Designing of the perpendicular drought index[J]. Environmental. Geology, 2007, 52(6): 1045-1052.
[25] MCVICAR T R, JUPP D L B. Using covariates to spatially interpolate moisture availability in the Murray- Darling Basin: A novel use of remotely sensed data[J]. Remote Sensing of Environment, 2002, 79(2): 199-212.
[26] KOGAN F. Droughts of the late 1980s in the United States as derived from NOAA polar-orbiting satellite data[J]. Bulletin of the American Meteorological. Society, 1995, 76(5): 655-668.
[27] SANDHOLT I, RASMUSSEN K, ANDERSEN J. A simple interpretation of the surface temperature/vegetation index space for assessment of surface moisture sta- tus[J]. Remote Sensing of Environment, 2002, 79(2): 213-224.
[28] MORAN M S, CLARKE T R, INOUE Y, et al. Estimating crop water deficit using the relation between surface-air temperature and spectral. vegetation index[J]. Remote Sensing of Environment, 1994, 49(3): 246-263.
[29] SU Z, YACOB A, WEN J, et al. Assessing relative soil moisture with remote sensing data: Theory, experimental. validation, and application to drought monitoring over the North China Plain[J]. Physics and Chemistry of the Earth, 2003, 28(1-3): 89-101.
[30] HUNT E R, ROCK B N. Detection of changes in leaf water content using near-and middle-infrared reflectances[J]. Remote Sensing of Environment, 1989, 30(1):43-54.
[31] ZARCO-TEJADA P J, USTIN S L. Modeling canopy water content for carbon estimates from MODIS data at land EOS validation sites[C]// International. Geoscience and Remote Sensing Symposium. Piscataway, New York, USA: IEEE, 2001: 342-344.
[32] WANG L, QU J. NMDI: A normalized multi-band drought index for monitoring soil and vegetation moisture with satellite remote sensing[J], Geophysical. Research Letters, 2007, 34(20): ID 20405.
[33] ZHANG N, HONG Y, QIN Q, et al. VSDI: A visible and shortwave infrared drought index for monitoring soil and vegetation moisture based on optical. remote sensing[J]. International. Journal. of Remote Sensing,2013,34(13-14):4585-4609.
[34] SHI J, JACKSON T, TAO J, et al. Microwave vegetation indices for short vegetation covers from satellitepassive microwave sensor AMSR-E[J]. Remote Sensing of Environment, 2008, 112(12): 4285-4300.
[35]王永前,施建成,劉志紅,等.微波植被指數(shù)在干旱監(jiān)測中的應用[J].遙感學報,2014, 18(4): 843-867.
WANG Y, SHI J, LIU Z, et al. Application of microwave vegetation index in drought monitoring[J]. Journal. of Remote Sensing, 2014, 18(4): 843-867.
[36]劉振華,趙英時.遙感熱慣量反演表層土壤水的方法研究[J].中國科學D輯:地球科學,2006, 36(6): 552-558.
LIU Z, ZHAO Y. The research on monitoring of land surface soil moisture by the thermal. inertia method in remote sensing[J]. Science in China Series D: Earth Sciences, 2006, 36(6): 552-558.
[37]吳黎,張有智,解文歡,等.改進的表觀熱慣量法反演土壤含水量[J].國土資源遙感,2013, 25(1): 44-49.
WU L, ZHANG Y, XIE W, et al. The inversion of soil water content by the improved apparent thermal. inertia[J]. Remote Sensing for Land and Resources,2013,25(1): 44-49.
[38] SUN W, WANG P, ZHANG S, et al. Using the vegetation temperature condition index for time series drought occurrence monitoring in the Guanzhong Plain, PR China[J]. International. Journal. of Remote Sensing, 2008, 29(17-18):5133-5144.
[39]楊永民,邱建秀,蘇紅波,等.基于熱紅外的四種土壤含水量估算方法對比[J].紅外與毫米波學報,2018,37(4): 459-467.
YANG Y, QIU J, SU H, et al. Estimation of surface soil moisture based on thermal. remote sensing: Intercomparison of four methods[J]. Journal. of Infrared and Millimeter Waves, 2018, 37(4): 459-467.
[40] GHULAM A, QIN Q, TEYIP T, et al. Modified perpendicular drought index (MPDI): A real-time drought monitoring method[J]. ISPRS Journal. of Photogrammetry and Remote Sensing, 2007, 62(2): 150-164.
[41] ALI I, GREIFENEDER F, STAMENKOVIC J, et al. Review of machine learning approaches for biomass and soil moisture retrievals from remote sensing data[J]. Remote Sensing, 2015, 7(12): 221-236.
[42] OH Y, SARABANDI K, ULABY F T. An empirical. model and an inversion technique for radar scattering from bare soil surfaces[J]. Transactions on Geoscience and Remote Sensing, 1992, 30(2): 370-381.
[43] FUNG A K, LI Z, CHEN K S. Backscattering from a randomly rough dielectric surface[J]. IEEE Transactions on Geoscience and Remote Sensing, 1992, 30(2): 356-369.
[44] ULABY F T, MCDONALD K, SARABANDI K, et al. Michigan microwave canopy scattering models (MIMICS) [C]// International. Geoscience and Remote Sensing Symposium (IGARSS). Piscataway, New York,USA: IEEE. ID 1009.
[45] SONG X, MA J, LI X, et al. First results of estimating surface soil moisture in the vegetated areas using ASAR and hyperion data: The Chinese Heihe River Basin case study[J]. Remote Sensing, 2014, 6(12): 12055-12069.
[46] MENG Q, XIE Q, WANG C, et al. A fusion approach of the improved Dubois model and best canopy water retrieval. models to retrieve soil moisture through all maize growth stages from Radarsat-2 and Landsat-8 data[J]. Environmental. Earth Sciences, 2016, 75(20): 1377-1391.
[47] ATTEMA E P W, ULABY F T. Vegetation modeled as a water cloud[J]. Radio Science, 1978, 13(2): 357-364.
[48] NOTARNICOLA C, ANGIULLI M, POSA F. Use of radar and optical. remotely sensed data for soil moisture retrieval. over vegetated areas[J]. IEEE Transactions on Geoscience and Remote Sensing, 2006, 44(4): 925-935.
[49] HAJJ MEL, BAGHDADI N, ZRIBI M, et al. Synergic use of Sentinel-1 and Sentinel-2 Images for operational. soil moisture mapping at high spatial. resolution over agricultural. areas[J]. Remote Sensing, 2017, 9(12): ID 1292.
[50] HAN D, WANG P, TANSEY K, et al. Linking an agro- meteorological. model and a water cloud model for estimating soil water content over wheat fields[J]. Computers and Electronics in Agriculture, 2020, 179: ID 105833.
[51] DUBOIS P C, ZYL J, ENGMAN T. Measuring soil moisture with imaging radars[J]. IEEE Transactions on Geoscience and Remote Sensing, 1995, 33(4): 915-926.
[52] STOGRYN A. Electromagnetic scattering from rough,finitely conducting surfaces[J]. Radio Science, 1967, 2(4):415-428.
[53] ULABY F T, MOORE R K, FUNG A K. Microwave remote sensing: Active and passive. Volume 2 - Radar remote sensing and surface scattering and emission theory[M]. Reading, MA: Addison Wesley, 1982.
[54] RICE S O. Reflection of electromagnetic waves from slightly random surfaces[J]. Communications on Pure and Applied Mathematics, 1951, 4(2-3): 351-378.
[55] VORONOVICH A G. Small-slope approximation in wave scattering from rough surfaces[J]. Waves in Random Media, 1985,6(2): 151-167.
[56] KARTHIKEYAN L, PAN M, WANDERS N, et al. Four decades of microwave satellite soil moisture observations: Part 1. A review of retrieval. algorithms[J].Advances in Water Resources, 2017, 109: 106-120.
[57] MO T, CHOUDHURY B J, SCHMUGGE T J, et al. A model for microwave emission from vegetation-covered fields[J]. Journal. of Geophysical. Research Oceans, 1982, 87(C13): 11229-11237.
[58] JACKSON T J, SCHMUGGE T J, et al. Vegetation effects on the microwave emission of soils[J]. Remote Sensing of Environment, 1991, 36(3): 203-212.
[59] RODRIGUEZ-FERNANDEZ N J, AIRES F, RICHAUME P, et al. Soil moisture retrieval. using neural. networks: Application to SMOS[J]. IEEE Transactions on Geoscience and Remote Sensing, 2015, 53(11): 5991- 6007.
[60] KOIKE T, NAKAMURA Y, KAIHOTSU I, et al. Development of an advanced microwave scanning radiometer (AMSR-E) algorithm for soil moisture and vegetation water content[J]. Proceedings of Hydraulic Engineering, 2004, 48: 217-222.
[61] KASAHARA M, IMAOKA K, KACHI M, et al. Status of AMSR2 on GCOM-W1[C]// Proc. International. Society for Optical. Engineering, Edinburgh, USA: SPIE,2012: ID 853307.
[62] PARINUSSA R M, WANG G, HOLMES T R H, et al. Global. surface soil moisture from the microwave radiation imager onboard the Fengyun-3B satellite [J]. International. Journal. of Remote Sensing, 2014, 35(19/20): 7007-7029.
[63] WAGNER W, HAHN S, KIDD R, et al. The ASCAT soil moisture product: A review of its specifications, validation results, and emerging applications[J]. MeteorologischeZeitschrift, 2013, 22: 5-33.
[64] FERNANDEZ-MORAN R, AL-YAARI A, MIALON A, et al. SMOS-IC: An alternative SMOS soil moisture and vegetation optical. depth product [J]. Remote Sensing, 2017, 9(5): ID 457.
[65] ENTEKHABI D, NJOKU E G, NEILL P E O, et al. The Soil Moisture Active Passive (SMAP) mission[J]. Proceedings of the IEEE, 2010, 98(5): 704-716.
[66] NJOKU E G, CROW W, YUEH S, et al. The SMAP mission combined active-passive soil moisture product at 9 km and 3 km spatial. resolutions [J]. Remote Sensing of Environment: An Interdisciplinary Journal, 2018,211: 204-217.
[67] SWATHANDRAN S, ASLAM M. Assessing the role of SWIR band in detecting agricultural. crop stress: A case study of Raichur district, Karnataka, India[J]. Environmental. Monitoring and Assessment, 2019, 191 (7): 442.
[68] GAO B C. NDWI-A normalized difference water index for remote sensing of vegetation liquid water fromspace[J]. Remote Sensing of Environment, 1996, 58 (3): 257-266.
[69] GERHARDS M, SCHLER F M, MALLICK K, et al. Challenges and future perspectives of multi-/hyperspectral. thermal. infrared remote sensing for crop waterstress detection: A review [J]. Remote Sensing, 2019, 11 (10): ID 1240.
[70] HAN D, LIU S, DU Y, et al. Crop water content of winter wheat revealed with Sentinel-1 and Sentinel-2 imag- ery[J]. Sensors, 2019, 19(18): ID 4013.
[71]吳伶,劉湘南,周博天,等.利用PROSPECT+SAIL模型反演植物生化參數(shù)的植被指數(shù)優(yōu)化模擬[J].應用生態(tài)學報,2012, 23(12): 3250-3256..
WU L, LIU X, ZHOU B, et al. Simulation of vegetation indices optimizing under retrieval. of vegetation biochemical. parameters based on PROSPECT+SAIL model[J]. Chinese Journal. of Applied Ecology, 2012, 23(12):3250-3256.
[72] DAS B, SAHOO R N, PARGAL. S, et al. Comparison of different uni- and multi-variate techniques for monitoring leaf water status as an indicator of water-deficit stress in wheat through spectroscopy [J]. Biosystems Engineering, 2017, 160: 69-83.
[73] CLEVERS J, KOOISTRA L, SCHAEPMAN M E. Estimating canopy water content using hyperspectral. remote sensing data[J]. International. Journal. of Applied Earth Observation and Geoinformation, 2010, 12(2): 119-125.
[74] JIN X, KUMAR L, LI Z, et al. A review of data assimilation of remote sensing and crop models [J]. European Journal. of Agronomy, 2018, 92: 141-152.
[75] XIE Y, WANG P, BAI X, et al. Assimilation of the leaf area index and vegetation temperature condition index for winter wheat yield estimation using Landsat imagery and the CERES-Wheat model[J]. Agricultural. and Forest Meteorology, 2017, 246: 194-206.
[76] HOLZWORTH D P, SNOW V, JANSSEN S, et al. Agricultural. production systems modelling and software: Current status and future prospects[J]. Environmental. Modelling & Software, 2015: 276-286.
[77] STEDUTO P, HSIAO T C, RAES D, et al. AquaCrop—The FAO crop model to simulate yield response to water: I. concepts and underlying principles[J]. Agronomy Journal, 2009, 101(3): 426-437.
[78] DUCHEMIN B, FIEUZAL. R, RIVERA M, et al. Impact of sowing date on yield and water use efficiency of wheat analyzed through spatial. modeling and FOR- MOSAT-2 imagesfJ]. Remote Sensing, 2015, 7(5): 5951-5979.
[79] DUCHEMIN B, MAISONGRANDE P, BOULET G,et al. A simple algorithm for yield estimates: Evaluation for semi-arid irrigated winter wheat monitored with green leaf area index[J]. Environmental. Modelling and Software, 2008, 23(7): 876-892.
[80] ALLEN R, PEREIRA L, RAES D. Crop evapotranspiration: Guidelines for computing crop water requirements, FAO irrigation and drainage paper 56[M]. Rome: FAO, 1998.
[81] SILVESTRO P C, PIGNATTI S, YANG H, et al. Sensitivity analysis of the Aquacrop and SAFYE crop models for the assessment of water limited winter wheat yield in regional. scale applications[J]. PLoS One,2017, 12(11): ID e0187485.
[82]韓萍,王鵬新,張樹譽,等.基于條件植被溫度指數(shù)的干旱預測研究[J].武漢大學學報(信息科學版),2010,35(10): 1202-1206.
HAN P, WANG P, ZHANG S, et al. Drought forecasting with vegetation temperature condition index[J]. Geomatics and Information Science of Wuhan University, 2010,35(10): 1202-1206.
[83]李俐,許連香,王鵬新,等.基于條件植被溫度指數(shù)的夏玉米生長季干旱預測研究[J].農(nóng)業(yè)機械學報,2020,51(1): 146-154.
LI L, XU L, WANG P, et al. Drought forecasting during maize growing season based on vegetation temperature condition index[J]. Transactions of the CSAM,2020,51(1): 146-154.
[84]張建海,張棋,許德合,等.ARIMA-LSTM組合模型在基于SPI干旱預測中的應用一以青海省為例[J].干旱區(qū)地理,2020, 43(4): 1004-1013.
ZHANG J, ZHANG Q, XU D, et al. Application of a combined ARIMA-LSTM model based on SPI for the forecast of drought: A case study in Qinghai province[J]. Arid Land Geography, 2020, 43(4): 1004- 1013.
[85]于洋,遲道才,陳濤濤,等.基于EMD的BP神經(jīng)網(wǎng)絡在凌河流域旱災預測中的應用[J].沈陽農(nóng)業(yè)大學學報,2014, 45(1): 68-72.
YU Y, CHI D, CHEN T, et al. Drought forecast application of BP prediction models based on EMD in Ling River Basin[J]. Journal. of Shenyang Agricultural. University, 2014, 45(1): 68-72.
[86]郭建平.農(nóng)業(yè)氣象災害監(jiān)測預測技術研究進展[J].應用氣象學報,2016,27(5): 620-630.
GUO J. Research progress on agricultural. meteorological. disaster monitoring and forecasting[J]. Journal. of Applied Meteorological. Science, 2016, 27(5): 620-630.
[87]吳熠婷,江琪,孟遠,等.氣候變化條件下陜西省不同氣候區(qū)灌溉對冬小麥減產(chǎn)風險影響評估[J].江蘇農(nóng)業(yè)科學,2021, 49(10): 213-222.
WU Y, JIANG Q, MENG Y, et al. Assessment of impact of irrigation in different climatic regions of Shaanxi province on risk of winter wheat yield reduction under climate change[J]. Jiangsu Agricultural. Sciences, 2021,49(10): 213-222.
[88]劉維,侯英雨,吳門新.基于WOFOST模型的2014年河南省干旱對夏玉米產(chǎn)量的預估[J].中國農(nóng)學通報,2016, 32(36): 146-151.
LIU W, HOU Y, WU M. Estimate of drought effect on summer maize yield based on WOFOST model in Henan Province in 2014[J]. Chinese Agricultural. Science Bulletin, 2016, 32(36): 146-151.
[89]楊霏云,鄭秋紅,李文科,等.基于WOFOST模型的遼寧省春玉米干旱災損風險評估[J].干旱地區(qū)農(nóng)業(yè)研究,2020, 38(6):218-225.
YANG F, ZHENG Q, LI W, et al. Risk assessment of drought damage of spring maize in Liaoning province based on WOFOST model[J]. Agricultural. Research In The Arid Areas, 2020, 38(6): 218-225.
[90]王治海,劉建棟,劉玲,等.基于遙感信息的區(qū)域農(nóng)業(yè)干旱模擬技術研究[J].水土保持通報,2013, 33(5): 96-100.
WANG Z, LIU J, LIU L, et al. Regional. agro-drought simulation based on remote sensing technology[J]. Bulletin of Soil and Water Conservation, 2013, 33(5): 96-100.
Progress of Agricultural. Drought Monitoring and Forecasting Using Satellite Remote Sensing
HAN Dong, WANG Pengxin, ZHANG Yue, TIAN Huiren, ZHOU Xijia
(College of Information and Electrical. Engineering. China Agricultural. University, Beijing 100083, China)
Abstract: Agricultural. drought is a major factor that affects agricultural. production. Traditional. agricultural. drought monitoring is mainly based on meteorological. and hydrological. data, and although it can provide more accurate drought monitoring results at the point level, there are still limitations in monitoring agricultural. drought at the regional. scale. The rapid development of remote sensing technology has provided a new mean of monitoring agricultural. droughts at the regional. scale, especially since the electromagnetic wavelengths sensed by satellite sensors in orbit now cover visible, near-infrared, thermal. infrared and microwave wavelengths. It is important to make full use of the rich surface information obtained from satellite remote sensing data for agricultural. drought monitoring and forecasting. This paper described the research progress of agricultural. drought monitoring based on satellite remote sensing from three aspects: remote sensing index-based method, soil water content method and crop water demand method. The research progress of agricultural. drought monitoring based on remote sensing index-based method was elaborated from five aspects: vegetation drought index, temperature drought index, integrated vegetation and temperature drought index, water drought index and microwave drought index; the research progress of agricultural. drought monitoring based on soil water content method was elaborated from two aspects: soil water content retrieval. based on visible to thermal. infrared data and soil water content retrieval. based on microwave data; the research progress of agricultural. drought monitoring based on crop water demand method was elaborated from two aspects: agricultural. drought monitoring based on crop canopy water content retrieval. method and crop growth model method. Agricultural. drought forecasting is a timeline prediction based on drought monitoring. Based on the summary of the progress of drought monitoring, the research progress of agricultural. drought forecasting by the drought index method and the crop growth model method was further briefly described. The existing agricultural. drought monitoring methods based on satellite remote sensing were summarized, and its shortcomings were sorted out, and some prospects were put forward. In the future, different remote sensing data sources can be used to combine deep learning methods with crop growth models and based on data assimilation methods to further explore the potential. of satellite remote sensing data in the monitoring of agricultural. drought dynamics, which can further promote the development of smart agriculture.
Key words: satellite; remote sensing; agricultural. drought; crop growth model; monitor; forecast
(登陸www.smartag.net.cn免費獲取電子版全文)
作者簡介:韓東(1994—),男,博士研究生,研究方向為農(nóng)業(yè)定量遙感。E-mail:hd5877@cau.edu.cn。
*通訊作者:王鵬新(1965—),男,博士,教授,研究方向為定量遙感及農(nóng)業(yè)應用。電話:010-62737622。E-mail:wangpx@cau.edu.cn。