日常近紅外光譜檢測在奶牛營養(yǎng)管理中的應用

ApplicationofRoutineNIRTestinginDairyNutritionalManagement

日常近紅外光譜檢測

在奶牛營養(yǎng)管理中的應用RalphWard拉夫·沃爾德DanniYe葉丹妮CumberlandValleyAnalyticalServices

美國CVAS飼料分析1CumberlandValleyAnalyticalServices,

Maryland,USA

CVAS飼料分析，美國馬里蘭州

Forageandfeedtestinglaboratory

牧草和飼料分析實驗室Largest“chemistry”foragelaboratoryinU.S.

全美最大的牧草化學分析實驗室Support18NIRlaboratoriesintheUSandothercountriesthroughouttheworld

支持美國及世界其他國家的18個近紅外光譜分析服務In2014weperformedanalysisonover250,000forageandfeedsamples.2014年，本公司為超過25萬個牧草和飼料樣品提供了分析服務2DairyProductivity乳業(yè)生產(chǎn)Whatdrivesdairyproductivity?

什么因素推動了乳業(yè)生產(chǎn)率Foragequalityandconsistency

粗飼料的質(zhì)量及穩(wěn)定性Cowcomfort奶牛舒適度Feedbunkmanagement飼槽管理Milkingmanagement擠奶管理Reproduction

繁殖Transitioncowmanagement

過渡期奶牛管理Costcontrol成本控制DairyProductivity乳業(yè)生產(chǎn)Whatdrivesdairyproductivity?

什么因素推動了乳業(yè)生產(chǎn)率Foragequalityandconsistency

粗飼料的質(zhì)量及穩(wěn)定性Cowcomfort奶牛舒適度Feedbunkmanagement飼槽管理Milkingmanagement擠奶管理Reproduction

繁殖Transitioncowmanagement

過渡期奶牛管理Costcontrol成本控制NIRTestingRoutineNIRtestinghasbeenakeymanagementtoolallowingforadvancementinefficiencyandprofitabilityoftheU.S.dairyindustry日常近紅外光譜（NIR）檢測是美國乳業(yè)高效運作及盈利的重要管理工具5NowmoreavailableinChinaLowcost,routineanalysis<300RMBFastturnaround:Resultsprovidedthesamedayordayafterreceiptinthelaboratory近紅外的檢測方法？目前具備：低成本，日常檢測，<300元/樣品快速獲得檢測報告：樣品到達實驗室的當日或者次日出報告6NIRTesting

近紅外光譜分析ForageEvaluation–WhatisImportant?

牧草分析：什么是重要的？Takingthesample采樣Themostimportantpartoftheprocessofevaluationofhayoranyforage!采樣是任何干草或者其他牧草檢測的最重要的部分！Propertechniqueisrequiredifalotofhaycomposedofmanybalesistoberepresentedbythesample.草捆包中含大量干草時，需要適當?shù)牟蓸蛹夹g才能采到具有代表性的樣品Manytonsaretoberepresentedbyperhapsa1kgsamplesenttothelab.1千克的樣品通常從上噸重的牧草中采集送至實驗室分析789Howfrequentlyshouldwetest?

飼料分析應當多頻繁？Weneedtounderstandthequalityvariationinfeedsonanongoingbasis.我們必須了解飼料質(zhì)量的日常動態(tài)變化Frequencyoftestingdependsonperceived“rateofchange”inforagequality檢測頻率決定于觀察到的粗飼料變化速度Highrateofchange,testmorefrequently變化快->須檢測頻繁Highvolumeofuse,testmorefrequently

用量大->須檢測頻繁Testfrequentlyenoughtorecognizechange必須檢測足夠頻繁以幫助確認飼料的變化11Howfrequentlyshouldwetest?

飼料分析應當多頻繁？InU.S.manysmallandlargedairyfarmstestforagesweekly,ormoreoften在美國，很多小型或大型牛場，每周甚至更頻繁，寄送粗飼料樣品進行檢測12Howfrequentlyshouldwetest?

飼料分析應當多頻繁？Forexample,ifwearetestinganalfalfahaybyNIR,thecalibrationmustincludesamplesofalfalfahay.例如：我們用NIR檢測苜蓿干草時，NIR的校準數(shù)據(jù)庫就必須已包含苜蓿干草的樣品Ifthesamplespectrahasspectrasimilartowhatisinthedatabase,NIRwillpredictresultscorrectly.如果通過NIR掃描后的樣品的光譜，與數(shù)據(jù)庫中的某個光譜接近，則NIR預測的結(jié)果準確；WhatfeedscanwetestbyNIR?

NIR能檢測哪些飼料？Ifthesamplespectraisnotwellrepresentedinthecalibrationdatabase,theNIRmaynotpredictresultscorrectly.如果樣品的光譜在校準數(shù)據(jù)庫中沒有接近的光譜，則NIR預測結(jié)果可能不夠準確。WhatfeedscanwetestbyNIR?

NIR能檢測哪些飼料？ExampleofSampleSpectranotMatchingCalibrationSpectra

舉例：樣品光譜與校準數(shù)據(jù)庫的光譜不吻合NIRPredictionChallenges

NIR分析面臨的挑戰(zhàn)Whatshouldthelaboratorydowhenaforageorfeedsampledoesnothaveacalibrationthatpredictswell?當某種牧草或飼料樣品的NIR預測結(jié)果不準確時，作為實驗室，該如何做？ThelaboratoryshouldrecognizewhenasampledoesnotpredictwellbyNIRandreportthistotheclient.首先，實驗室發(fā)現(xiàn)NIR結(jié)果不準確時應當告知客戶Thelaboratorymaychoosetoreportresultsas“potentiallynotaccurate”andtheverifyandreportbychemistrymethods.實驗室應當在檢測報告中注明“NIR結(jié)果可能不準確”，并通過化學分析進行核對NIRPredictionChallenges

NIR分析面臨的挑戰(zhàn)ThelaboratorywillusetheresultsofchemistryanalysisonafeedthatisnotpredictedbyNIRto“improve”theNIRpredictionortoworktobuildanewpredictionequation.實驗室將使用某種飼料的傳統(tǒng)化學分析結(jié)果來改善NIR的預測準確度，或者建立新的校準數(shù)據(jù)庫NIRPredictionChallenges

NIR分析面臨的挑戰(zhàn)NIREvaluationofMinerals

礦物質(zhì)的NIR分析NIRtestingisbasedonthereflectanceofnearinfra-redlightfromorganicbondsinfeedmaterials.NIR是基于近紅外光在飼料中的有機化學鍵的反射的分析方法。Protein(aminoacids)willpredictwellbecausetheyarebasedonorganicbonds.因此，NIR對蛋白質(zhì)（氨基酸）的預測比較準確Minerals(Ca,P,Mg,K)willnotpredictwellbecausethereislimitedorganicbonding.而礦物質(zhì)，比如鈣磷鎂鉀，就無法檢測準確，因為缺乏有機化學鍵Ifunderstandingthemineralcontentinaforageorfeedisimportant,theusershouldrequestthatthemineralanalysisbeperformedbychemistrymethods,notbyNIRanalysis.如果礦物質(zhì)含量對某種牧草或飼料成分很重要，應當要求實驗室通過傳統(tǒng)化學方法檢測，而不是NIRDietarycation-aniondifference(DCAD)foruseinestablishingdietsfordrycowsshouldnotbedeterminedbyNIR!同理，干奶期奶牛日糧中的陰陽離子平衡也不應該用NIR檢測。NIREvaluationofMinerals

礦物質(zhì)的NIR分析ApplicationofNIRTesting

NIR分析的應用Definefeedvariability定義“飼料變化”Recognizewhenchangeoccursinfeedstocks當飼料存貨開始變化時，及時確認Understandfeedqualitychallengesandpromotechange理解飼料品質(zhì)面臨的挑戰(zhàn)并優(yōu)化管理以提高飼料品質(zhì)Benchmarkchangeovertime長期關注飼料變化，建立參照基準Monitorapparentwholecowstarchdigestibility監(jiān)測玉米淀粉的表觀消化率Qualitycontroloffeedmixinganddeliverprocess飼料混合及投料過程的質(zhì)量監(jiān)控Provideinputsforrationmodeling為配方模型提供信息InterpretationofResults

檢測報告的解讀Theremaybeashcontaminationinmanyforagesources.灰分受污染是很多牧草的重要因素Ashcontaminationfromsoilmayreducethequalityofthefermentationinsilages土壤造成的灰分污染會降低青貯飼料的發(fā)酵品質(zhì)Ashcontaminationmayreducepalatabilityofafeed灰分受污染會降低飼料的適口性Ashcontaminationshouldbeaqualityfactortoconsiderwhenpurchasingforage購買牧草時，灰分受污染應當是一個重要考慮因素Ifyouhaveanashvalueof12.4%inalfalfahay,doesthisindicatethatthereisaproblemwithashcontamination?如果您的苜蓿干草的灰分含量為12.4%，是否看得出灰分被污染？Weneedtohaveanunderstandingoftheaverageandrangeofash(oranynutrient)inafeedcategoryinordertounderstandhowtousetheinformation.要想正確解讀報告中的數(shù)據(jù)，首先需要了解灰分（或者任何其他營養(yǎng)成分）在某類飼料類型中的平均值及分布范圍。InterpretationofResults

檢測報告的解讀DistributionofAsh,WesternStatesAlfalfaHay(Chemistry,CVAS2011)

美國西部苜蓿干草的粗灰分分布圖(化學分析,CVAS2011)N樣品數(shù)=1,520Ave.平均值=10.76St.Dev.標準差=1.29發(fā)樣品比例26RecognizingChangeinForageQuality

確認飼料質(zhì)量的變化Starchcontentofcornsilagesamplesanalyzedovertime:玉米青貯隨時間所測的淀粉含量如下：May2 5月2日

34.4%June16月1日 31.7%June30th6月30日

26.5%

RecognizingChangeinForageQuality

確認飼料質(zhì)量的變化Starchcontentofcornsilagesamplesanalyzedovertime:玉米青貯隨時間所測的淀粉含量如下：June10th

6月10日 32.4%June16th

6月16日 31.7%June24th

6月24日 33.6%June30th6月30日

26.5% July5th

7月5日 31.8%July12th

7月12日 34.1%

Average(blue)藍色部分平均值33.2%Theresultofonesampleanalysisdoesnotnecessarilyrepresentthetrueanalysisoftheforagematerialbeingsampled.某一個樣品的檢測結(jié)果不一定就代表所采集的牧草的真正結(jié)果Musttakefrequentsamplesovertimeanduseanaveragevalueforrationmodelingpurposes.必須長期采樣，在配方模型中使用平均值RecognizingChangeinForageQuality

確認飼料質(zhì)量的變化FeedQualityEvaluationsbyNIR

NIR分析對飼料品質(zhì)的評估FermentationQuality發(fā)酵品質(zhì)FiberDigestibility纖維消化率StarchDigestibility淀粉消化率ParticleSize顆粒度FermentationQuality發(fā)酵品質(zhì)Extentoffermentation發(fā)酵程度Levelofratiooflacticacidtoaceticacid乳酸乙酸的濃比Presence

ofbutyricacidinhaysilage半干青貯中是否有丁酸的產(chǎn)生Levelofammonia發(fā)酵過程中的氨濃度FeedQualityEvaluationsbyNUR

NIR分析對飼料品質(zhì)的評估發(fā)樣品比例NH3CrudeProteinEquivalent%DMinTMR(CVAS2015)

全混合日糧中氨濃度（轉(zhuǎn)換為粗蛋白，干物質(zhì)基礎）(CVAS2015)N樣本數(shù)=1204Ave.平均值=1.1St.Dev.標準差=0.5發(fā)樣品比例N樣本數(shù)=6285Ave.平均值=40.8Std.Dev.標準差=4.12Distributionof30hrInVitroDigestibilityinWesternAlfalfaHay(CVAS2014)

美國西部苜蓿干草30小時體外消化率分布圖(CVAS,2014)CornSilageProcessingScore

(CSPS)

(CVASData,2006to2014)

玉米青貯加工評分（CVAS數(shù)據(jù)，2006-2014）收割年份CropYear樣品數(shù)Number平均值％Average最佳級別％PercentOptimum最低級別％Percent

Poor20069752.88.243.3200727252.39.237.9200825054.65.234.8200924451.16.148.0201037351.45.943.4201172655.512.333.1201287160.814.819.92013265864.626.222.12014463462.225.810.4AnExampleofBenchmarking

CSPSonadairyfarm

玉米青貯加工評分（CSPS）舉例CSPSbyYearCSPS按年2010 55.4%2011 57.1%2012 61.8%2013 62.5%2014 56.4%ProductionbyYear產(chǎn)量按年2010 29.5kg2011 30.9kg2012 31.2kg2013 33.8kg2014 30.8kgQualityEvaluationofCornSilage

玉米青貯的質(zhì)量評估CornSilage#1玉米青貯1DM干物質(zhì) 29.2%CP 粗蛋白 8.7%ADF 酸洗 24.8%NDF 中洗 42.6%Starch淀粉

31.5%

CornSilage#2玉米青貯2DM 干物質(zhì) 31.6%CP 粗蛋白 7.4%ADF 酸洗 26.2%NDF 中洗 44.6%Starch淀粉 29.5%CornSilage#1玉米青貯1Lactic乳酸 3.4% Acetic乙酸 4.7%NDFD30 54.5%IVSD768.2%CSPS加工評分48.9%CornSilage#2玉米青貯2Lactic乳酸 5.3%Acetic乙酸 2.7%NDFD30 58.5%IVSD7 76.2%CSPS加工評分

67.1%

QualityEvaluationofCornSilage

玉米青貯的質(zhì)量評估NDFD30：30hNDF消化率IVSD7：7h體外淀粉消化率CornSilage#1玉米青貯1DM干物質(zhì) 29.2%CP粗蛋白 8.7%ADF 酸洗 24.8%NDF 中洗 42.6%Starch淀粉 31.5%Lactic乳酸 3.4%Acetic乙酸 4.7%NDFD30 54.5%IVSD7 68.2%CSPS 加工評分48.9%CornSilage#2玉米青貯2DM干物質(zhì) 31.6%CP粗蛋白 7.4%ADF 酸洗 26.2%NDF 中洗 44.6%Starch淀粉 29.5%Lactic乳酸 5.3%Acetic乙酸 2.7%NDFD30 58.5%IVSD7 76.2%CSPS 加工評分67.1%QualityEvaluationofCornSilage

玉米青貯的質(zhì)量評估NDFD30：30hNDF消化率IVSD7：7h體外淀粉消化率發(fā)樣品比例DistributionofStarchValuesinCattleFeces(CVAS2012)

牛糞便淀粉含量分布圖

(CVAS2012)N樣本數(shù)=2,267Ave.平均值=6.74St.Dev.標準差=8.94MonitoringStarchDigestibility

淀粉消化率Opportunitytomonitortotaltractstarchdigestibility助于監(jiān)測全消化道淀粉消化率PullasampleofTMRandevaluateforstarchcontent

采集TMR樣，檢測淀粉含量TakeasampleofassociatedfecalmaterialandevaluatebyNIRforstarch

采集對應的糞便樣品，用NIR檢測淀粉含量Canestimatetotaltractstarchdigestibilityfromthisinformation

通過計算可得全消化道淀粉消化率ApparentStarchDigestibilityInLactatingDairyCattle(CVAS,2014)

泌乳奶牛的淀粉表觀消化率(CVAS,2014)N樣本數(shù)=375Ave.平均值=93.6SD標準差=3.83RationDevelopmentandModeling

日糧配方NewerrationmodelingprogramssuchastheCornellModel,AMTS,orNDSrequiremanyinputs.新的日糧模型軟件比如康奈爾模型，AMTS或NDS都需要大量的數(shù)據(jù)基礎NIRcanprovidetheseinputsrapidlyatverylowcostcomparedtochemistryanalysis.NIR能快速提供所需數(shù)據(jù)，相比傳統(tǒng)化學分析，成本低很多DM干物質(zhì)CP粗蛋白NH3氨NDF-CP中洗不溶蛋白

NDFom中洗（有機物基礎）Ash灰分FattyAcids脂肪酸NDFD30HR中洗30h消化率NDFD120HR中洗120h消化率NDFD240HR中洗240h消化率FermentationAcids發(fā)酵酸AnalysisbyNIR:基礎NIR分析：300RMB

300元／樣1-2daysinlaboratory

1至2天出報告

Analysisbytraditionalchemistrymethods:傳統(tǒng)化學分析：>900RMB

>900元／樣>4daysinlaboratory

>4天出報告>12daysforlongtermdigestibilitydata

>12天出120h或者240h消化率數(shù)據(jù)RationDevelopmentandModeling

日糧配方NIRisanexcellenttooltoprovideinformationformanagementdecisionsregardingfeedqualityandfeedingmanagement.NIR分析是飼料品質(zhì)及飼喂管理決策的極好的工具。However,NIRmustbeusedandinterpretedproperlyinorderprovidethemostvalue.然而，NIR分析必須正確使用及解讀，以最大化其用途。44Conclusions

總結(jié)Takeasampleofforageusinggoodsamplingprotocol–thesampleneedstoberepresentativeofthematerialfedtothecows.嚴格遵守標準操作程序科學采樣，因為所采集的樣品必須充分代表奶牛所攝入的飼料Forageandfeedsource