Kea leboha ka ho etela Nature.com.Mofuta oa sebatli oo u o sebelisang o na le ts'ehetso e fokolang bakeng sa CSS.Bakeng sa boiphihlelo bo botle ka ho fetisisa, re khothaletsa hore u sebelise sebatli se ntlafalitsoeng (kapa tima mokhoa oa ho lumellana ho Internet Explorer).Khabareng, ho netefatsa tšehetso e tsoelang pele, re tla bonts'a sebaka sena ntle le mekhoa le JavaScript.
Tšilafalo ea mobu ke bothata bo boholo bo bakoang ke mesebetsi ea batho.Kabo ea sebaka sa lintho tse ka bang chefo (PTEs) e fapana libakeng tse ngata tsa litoropo le tse haufi le litoropo.Ka hona, ho thata ho bolela esale pele hore na PTE e na le eng mobung o joalo. Kakaretso ea lisampole tse 115 li ile tsa fumanoa ho tsoa ho Frydek Mistek in the Czech Republic.Calcium, potassium nick) li-concentrations li ne li ikemiselitse ho sebelisa inductively coupled plasma emission spectrometry.Phetoho ea karabelo ke Ni le li-predictors ke Ca, Mg, le K.Matrix ea correlation pakeng tsa karabo ea karabo le phetoho ea ho bolela esale pele e bontša kamano e khotsofatsang pakeng tsa likarolo.Liphetho tsa ho bolela esale pele li bontšitse hore Tšehetso ea Vector Machine Regression (SVMR) e entse hantle, le hoja motso oa eona o hakanngoa hore o bolela phoso ea square / RMSE (mg9) le square23 (mg9). phoso e felletseng (MAE) (166.946 mg/kg) e ne e phahame ho feta mekhoa e meng e sebelisitsoeng.Mehlala e tsoakiloeng bakeng sa Empirical Bayesian Kriging-Multiple Linear Regression (EBK-MLR) e sebetsa hampe, joalo ka ha ho pakoa ke li-coefficients of determination tlase ho 0.1.Mohlala oa Empirical Bayesian Kriging-SVRK e ne e le molemo ka ho fetisisa oa Mochine oa Bayesian Kriging-Sup ka boleng bo tlase ba RMSE (95.479 mg/kg) le MAE (77.368 mg/kg) le coefficient e phahameng ea boikemisetso (R2 = 0.637) le mobu o haufi le teropo.Liphetho li bonts'a hore ho kopanya EBK le SVMR ke mokhoa o sebetsang oa ho bolela esale pele hore ho na le li-concentration tsa Ni mobung oa litoropo le o haufi le toropo.
Nickel (Ni) e nkoa e le micronutrient bakeng sa limela hobane e kenya letsoho ho lokisa naetrojene ea sepakapaka (N) le metabolism ea urea, tseo ka bobeli li hlokahalang bakeng sa ho mela ha peo. Ho phaella ho tlatsetso ea eona ea ho mela ha peo, Ni e ka sebetsa e le inhibitor ea fungal le baktheria le ho khothalletsa tsoelo-pele ea limela. hloka tšebeliso ea manyolo a thehiloeng ho nickel ho ntlafatsa nitrogen fixation2.Ho tsoela pele ho sebelisoa ha menontsha e thehiloeng ho nickel ho nontša mobu le ho eketsa bokhoni ba limela tsa linaoa ho lokisa naetrojene mobung ka ho tsoelang pele ho eketsa motsoako oa nickel mobung. ho nkeloa ha tšepe e le limatlafatsi tse hlokahalang bakeng sa kholo ea limela1. Ho latela Liu3, Ni e fumanoe e le karolo ea 17 ea bohlokoa e hlokahalang bakeng sa nts'etsopele le kholo ea limela. Ho phaella mosebetsing oa nickel ho nts'etsopele le kholo ea limela, batho ba e hloka bakeng sa mefuta e fapaneng ea likopo. Ho feta moo, li-alloys tse thehiloeng ho nickel le lisebelisoa tsa electroplated li sebelisitsoe haholo ho kitchenware, lisebelisoa tsa ballroom, thepa ea indasteri ea lijo, motlakase, terata le cable, li-jet turbines, li-implants tsa ho buoa, masela le kaho ea likepe5. ho feta anthropogenic4,6.Mehloli ea tlhaho ea nickel e kenyelletsa ho foqoha ha seretse se chesang, limela, mello ea meru, le mekhoa ea jeoloji; leha ho le joalo, mehloli ea anthropogenic e kenyelletsa libeteri tsa nickel/cadmium indastering ea tšepe, electroplating, welding ea arc, diesel le mafura a mafura, le mesi e tsoang moeeng e tsoang ho cheso ea mashala le litšila le ho chesoa ha litšila Ho bokellana ha Nickel7,8.Ho latela Freedman le Hutchinson9 le Manyiwa et. 10, mehloli e ka sehloohong ea tšilafalo ea mobu sebakeng se haufi le se haufi haholo ke li-smelters le merafo e thehiloeng ka koporo ea nickel. Mobu o ka holimo o pota-potileng sebaka sa ho hloekisa nickel-copper sa Sudbury Canada se ne se e-na le tšilafalo e phahameng ka ho fetisisa ea tšilafalo ea nickel ho 26,000 mg / kg11. Mobu oa Norway11.Ho latela Alms et al. 12, palo ea HNO3-extractable nickel sebakeng se ka holimo se lengoang sebakeng seo (tlhahiso ea nickel Russia) e ne e tloha ho 6.25 ho ea ho 136.88 mg / kg, e lumellanang le moelelo oa 30.43 mg / kg le motsoako oa motheo oa 25 mg / kg. Ho ea ka kabata 11, mobu oa litoropong oa phosphorus kapa mobu oa phosphorus nakong ea temo. linako tsa lijalo tse latellanang li ka kenya kapa tsa silafatsa mobu.Liphello tse ka bang teng tsa nickel ho batho li ka baka mofetše ka mutagenesis, tšenyo ea chromosomal, moloko oa Z-DNA, ho thibela ho lokisoa ha DNA, kapa mekhoa ea epigenetic13. Litekong tsa liphoofolo, nickel e fumanoe e na le bokhoni ba ho baka mefuta e sa tšoaneng ea lihlahala, 'me mefuta e joalo ea kankere ea nixacer e ka 'na ea e-ba e-epigenetic.
Litlhahlobo tsa tšilafalo ea mobu li atlehile matsatsing a morao tjena ka lebaka la litaba tse ngata tse amanang le bophelo bo botle tse hlahang likamanong tsa mobu le limela, mobu le likamano tsa likokoana-hloko tsa mobu, ho senyeha ha tikoloho, le tlhahlobo ea phello ea tikoloho. e ntlafalitse haholo 'mapa oa mobu oa ho bolela esale pele (PSM).Ho latela Minasny le McBratney16, predictive soil mapping (DSM) has proven to be a prominent subdiscipline of soil science.Lagacherie and McBratney, 2006 define DSM as “thep and fill of spatial soil information systems through the use of in sipatial-leb-option- litsamaiso tsa maikutlo”.McBratney et al. 17 e hlalosa hore DSM kapa PSM ea mehleng ea kajeno ke mokhoa o atlehang ka ho fetisisa oa ho bolela esale pele kapa ho etsa 'mapa oa ho ajoa ha sebaka sa PTEs, mefuta ea mobu le thepa ea mobu.Geostatistics le Machine Learning Algorithms (MLA) ke mekhoa ea DSM ea ho etsa limmapa tsa digitized ka thuso ea lik'homphieutha tse sebelisang lintlha tse bohlokoa le tse fokolang.
Deutsch18 le Olea19 li hlalosa geostatistics e le "pokello ea mekhoa ea lipalo e sebetsanang le kemelo ea litšobotsi tsa sebaka, haholo-holo ho sebelisa mehlala ea stochastic, joalo ka hore na tlhahlobo ea letoto la nako e khetholla data ea nakoana joang." Haholo-holo, geostatistics e kenyelletsa tlhahlobo ea li-variograms, tse lumellang Quantify le ho hlalosa ho its'epaha ha boleng ba sebaka ho tsoa ho dataset ka 'ngoe20.Gumiaux et al. 20 e boetse e bontša hore tlhahlobo ea li-variograms ho geostatistics e itšetlehile ka melao-motheo e meraro, ho kenyelletsa le (a) computing tekanyo ea correlation ea data, (b) ho khetholla le ho etsa computing anisotropy ka dataset disparity le (c) ho phaella ho Ho phaella tabeng ea ho nahanela phoso ea tlhaho ea data ea tekanyo e arohaneng le liphello tsa libaka tse hakantsoeng, mekhoa e mengata e hakantsoeng. li sebelisoa ho geostatistics, ho kenyeletsoa kriging e akaretsang, co-kriging, kriging e tloaelehileng, empirical Bayesian kriging, mokhoa o bonolo oa kriging le mekhoa e meng e tsebahalang ea ho fetolela ho etsa 'mapa kapa ho bolela esale pele PTE, litšobotsi tsa mobu, le mefuta ea mobu.
Machine Learning Algorithms (MLA) ke mokhoa o batlang o le mocha o sebelisang litlelase tse kholoanyane tsa data tseo e seng tsa linear, tse susumetsoang ke dikgato-tharabololo tse sebediswang haholo bakeng sa meepo ya data, ho tsebahatsa dipaterone ho data, le ho sebediswa kgafetsa ho arola mafapheng a mahlale a kang mahlale a mobu le mesebetsi ya ho kgutlela.Pampiri tse ngata tsa dipatlisiso di itshetlehile ka mefuta ya MLA ho bolela esale pele PTE mobung, jwalo ka Tan et al. 22 (meru e sa tloaelehang bakeng sa tekanyo ea tšepe e boima mobung oa temo), Sakizadeh et al. 23 (mohlala o sebelisa mechine ea li-vector ea tšehetso le marang-rang a maiketsetso a maiketsetso) tšilafalo ea mobu ) .Ho phaella moo, Vega et al. 24 (CART bakeng sa ho etsa mohlala oa ho boloka tšepe e boima le adsorption mobung) Sun et al. 25 (tshebediso ya cubist ke kabo ea Cd mobung) le dikgatotharabololo tse ding tse kang k-haufi moahelani, generalized boosted regression, le boosted regression Lifate e boetse e sebelisoa MLA ho bolela esale pele PTE mobung.
Tšebeliso ea li-algorithms tsa DSM ka ho bolela esale pele kapa 'mapa e tobana le mathata a mangata.Bangoli ba bangata ba lumela hore MLA e phahametse geostatistics le ka tsela e fapaneng.Le hoja e le' ngoe e molemo ho feta e 'ngoe, motsoako oa bobeli o ntlafatsa boemo ba ho nepahala ha 'mapa kapa ho bolela esale pele ho DSM15.Woodcock le Gopal26 Finke27; Pontius le Cheuk28 le Grunwald29 ba bua ka mefokolo le liphoso tse ling tse boletsoeng esale pele 'mapa oa mobu.Bo-rasaense ba mobu ba lekile mekhoa e mengata ea ho ntlafatsa katleho, ho nepahala, le ho tseba esale pele ka ho etsa 'mapa oa DSM le ho bolela esale pele 15 e hlakisa hore boitšoaro ba netefatso le ho hloka bonnete tse hlahisoang ke popo ea 'mapa le ho bolela esale pele li lokela ho netefatsoa ka boikemelo ho ntlafatsa boleng ba' mapa.Mefokolo ea DSM e bakoa ke boleng ba mobu o hasantsoeng, o kenyelletsang karolo ea ho hloka botsitso; leha ho le joalo, ho hloka bonnete ba DSM ho ka 'na ha hlaha mehloling e mengata ea phoso, e leng phoso ea covariate, phoso ea mohlala, phoso ea sebaka, le Phoso ea tlhahlobo ea 31. Ho se nepahale ha mohlala ho bakoang ke MLA le mekhoa ea geostatistical ho amahanngoa le ho hloka kutloisiso, qetellong ho lebisang ho fetelletseng ha mokhoa oa sebele32. lipalo tsa mohlala oa lipalo, kapa interpolation33.Haufinyane tjena, ho hlahile mokhoa o mocha oa DSM o khothalletsang ho kopanngoa ha geostatistics le MLA ho etsa limmapa le ho bolela esale pele.Bo-rasaense le bangoli ba 'maloa ba mobu, ba kang Sergeev et al. 34; Subbotina et al. 35; Tarasov et al. 36 le Tarasov et al. 37 e sebelisitse boleng bo nepahetseng ba lipalo-palo le thuto ea mochini ho hlahisa mefuta e nyalisitsoeng e ntlafatsang ts'ebetso ea ponelopele le 'mapa. boleng.E meng ea mefuta ena e nyalisitsoeng kapa e kopantsoeng ea algorithm ke Artificial Neural Network Kriging (ANN-RK), Multilayer Perceptron Residual Kriging (MLP-RK), Generalized Regression Neural Network Residual Kriging (GR- NNRK)36, Artificial Neural Network Kriging-Multilayer Perceptron-MLP-ANNR-3K Processor-MLP-ANNriging le ANNriging Poelo morago38.
Ho ea ka Sergeev et al., ho kopanya mekhoa e fapaneng ea ho etsa mohlala ho na le monyetla oa ho felisa liphoso le ho eketsa katleho ea mohlala oa lebasetere o hlahisoang ho e-na le ho ntshetsa pele ea eona ea mohlala e le 'ngoe.Moelelong ona, pampiri ena e ncha e pheha khang ea hore hoa hlokahala ho sebelisa algorithm e kopantsoeng ea geostatistics le MLA ho theha mefuta e metle ea li-hybrid ho bolela esale pele hore Ni ruime libakeng tsa litoropo le periurban Bay. (EBK) e le mohlala oa motheo le ho o kopanya le Support Vector Machine (SVM) le Multiple Linear Regression (MLR) mehlala.Hybridization of EBK with any MLA is not known.Mefuta e mengata e tsoakiloeng e bonoang ke metsoako ea kriging e tloaelehileng, e setseng, ea ho fokotseha, le MLA.EBK is a geostatistical interpolation method that utilizes a local process as spare tšimo e sa tsitsang/e emeng e nang le liparamente tse hlalositsoeng tsa sebaka ka har'a tšimo, e lumellang phapang ea sebaka39.EBK e sebelisitsoe liphuputsong tse fapaneng, ho kenyeletsoa ho hlahloba kabo ea carbon carbon mobung oa polasi40, ho hlahloba tšilafalo ea mobu41 le ho etsa 'mapa oa thepa ea mobu42.
Ka lehlakoreng le leng, Self-Organising Graph (SeOM) ke algorithm ea ho ithuta e sebelisitsoeng lihloohong tse fapaneng tse kang Li et al. 43, Wang et al. 44, Hossain Bhuiyan et al. 45 le Kebonye et al.46 Fumana litšobotsi tsa sebaka le lihlopha tsa likarolo.Wang et al. 44 hlalosa hore SeOM ke mokhoa o matla oa ho ithuta o tsejoang ka bokhoni ba eona ba ho kopanya le ho nahana ka mathata a sa tsitsang.Ho fapana le mekhoa e meng ea ho lemoha mohlala e kang principal component analysis, fuzzy clustering, hierarchical clustering, and multi-criteria liqeto liqeto, SeOM e molemo ho hlophisa le ho khetholla mekhoa ea PTE.According to Waccording. 44, SeOM e ka arola sebaka sa kabo ea li-neurons tse amanang le ho fana ka pono e phahameng ea data.
Pampiri ena e ikemiselitse ho hlahisa mofuta o matla oa 'mapa o nepahetseng ka ho fetesisa oa ho bolela esale pele litaba tsa nickel mobung oa litoropo le o haufi le litoropo. ka hona, re tla leka ho araba lipotso tsa lipatlisiso tse ka 'nang tsa fana ka mehlala e tsoakiloeng.Leha ho le joalo, mohlala o nepahetse hakae ho bolela esale pele ntho e lebisitsoeng?Hape, ke boemo bofe ba tlhahlobo ea katleho e thehiloeng holim'a netefatso le tlhahlobo ea ho nepahala?Ka hona, lipakane tse khethehileng tsa thuto ena e ne e le (a) ho etsa mohlala o kopantsoeng oa motsoako oa SVMR kapa MLR o sebelisa EBK (mohlala oa ho etsa mohlala o motle ka ho fetisisa) Ke tsepamisitsoeng mobung oa litoropo kapa o haufi le toropo , le (d) ts'ebeliso ea SeOM ho theha 'mapa oa boemo bo holimo oa phetoho ea sebaka sa nickel.
Thuto e ntse e etsoa Czech Republic, haholo-holo seterekeng sa Frydek Mistek sebakeng sa Moravia-Silesian (sheba setšoantšo sa 1). Geography ea sebaka sa boithuto se matsutla haholo 'me boholo ba sona ke karolo ea sebaka sa Moravia-Silesian Beskidy, e leng karolo ea moeli o ka ntle oa Lithaba tsa Carpathian.Sebaka sa boithuto se pakeng tsa 4'10 ° le N1'19 ° 4 le 2'19 ° 0′ E, 'me bophahamo bo pakeng tsa 225 le 327 m; leha ho le joalo, mokhoa oa ho arola Koppen bakeng sa boemo ba leholimo ba sebaka seo o nkoa e le Cfb = boemo ba leholimo bo futhumetseng ba leoatle, Ho na le pula e ngata esita le likhoeling tse omileng.Mocheso o fapana hanyane ho pholletsa le selemo pakeng tsa -5 °C le 24 °C, ka seoelo o theohelang ka tlase ho −14 °C kapa ka holimo ho 30 °C ea selemo le selemo ke 52 ° C, ha ho ntse ho e-na le pula e lekaneng ea 30 ° C. Ho hakanngoa hore sebaka sa phuputso ea sebaka sohle ke lisekoere-k'hilomithara tse 1,208, 'me 39.38% ea mobu o lenngoeng le 49.36% ea meru e koahetsoeng ke meru. (mohlala, bakeng sa ho hanyetsa kutu ea sepakapaka) le litšepe tsa alloy (nickel e eketsa matla a motsoako ha e ntse e boloka ductility ea eona e ntle le toughness), le temo e matla joalo ka kopo ea manyolo a phosphate le tlhahiso ea mehlape ke mehloli ea lipatlisiso e ka bang teng ea nickel sebakeng seo (mohlala, ho eketsa nickel ho likonyana ho eketsa litekanyetso tsa kholo ea likhomo tsa likhomo le ho sebelisa li-nickel libakeng tse tlase). electroplating, ho kenyeletsa electroplating nickel le electroless nickel plating systems.Mehaho ea mobu e khetholloa habonolo ho tloha ho mebala ea mobu, sebopeho, le lihlahisoa tsa carbonate.Mobu oa mobu o mahareng ho isa ho o motle, o tsoang ho lintho tse bonahalang tsa motsoali.Li na le tlhaho ea colluvial, alluvial kapa aeolian. sebaka sa48.Ka bophahamo ho tloha ho 455.1 ho isa ho 493.5 m, li-cambisol li laola Czech Republic49.
'Mapa oa sebaka sa boithuto ['Mapa oa sebaka sa boithuto o entsoe ho sebelisoa ArcGIS Desktop (ESRI, Inc, mofuta oa 10.7, URL: https://desktop.arcgis.com).]
Kakaretso ea lisampole tsa mobu o ka holimo oa 115 li ile tsa fumanoa mobung oa litoropo le o haufi le metse seterekeng sa Frydek Mistek. Mohlala oa mohlala o sebelisitsoeng e ne e le marang-rang a tloaelehileng a nang le lisampole tsa mobu tse arohaneng ka 2 × 2 km ka thōko, 'me mobu o ka holimo o ne o lekanngoa ka botebo ba 0 ho ea ho 20 cm ho sebelisa mochine o tšoaroang ka letsoho oa GPS (Leica Zeno 5 bags, Ziplobele e phuthetsoe ka GPS ka tsela e nepahetseng). ho ea laboratoring.Lisampole li ne li omisitsoe moeeng ho hlahisa lisampole tse silafalitsoeng, tse silafalitsoeng ka mochine oa mochine (Fritsch disc leloala), 'me tsa sieved (sieve size 2 mm) Beha 1 gram ea mobu o omisitsoeng, o nang le homogenized le o sefiloeng ka libotlolong tsa teflon tse ngotsoeng ka ho hlaka.Sejaneng se seng le se seng sa Teflon, 3% 5% ea HC ea Teflon, fana ka 5% HC ea 5%. HNO3 (ho sebelisa mochine o ikemetseng - e le 'ngoe bakeng sa acid e' ngoe le e 'ngoe), koahela hanyenyane 'me u lumelle lisampole hore li eme bosiu bo le bong bakeng sa karabelo (lenaneo la aqua regia) .Beha supernatant holim'a poleiti ea tšepe e chesang (mocheso: 100 W le 160 ° C) bakeng sa 2 h ho nolofatsa ts'ebetso ea tšilo ea lijo ea lisampole, ebe o pholile ho 50 ml ea ho pholile ho 50 ml ea flatric ho 50 ml ea ho pholile. ml e nang le metsi a hloekisitsoeng. Ka mor'a moo, sefa "supernatant" e hlapolotsoeng ka har'a tube ea PVC ea 50 ml e nang le metsi a hloekisitsoeng. Ho phaella moo, 1 ml ea tharollo ea dilution e ile ea hlapolloa ka 9 ml ea metsi a hloekisitsoeng 'me ea hloekisoa ka tube ea 12 ml e lokiselitsoeng bakeng sa PTE pseudo-concentration.The concentrations of PTEs, Crn, Cd, Cd, Cd, Cd, Cd, PTEs (As, Cb, Cd, Cb, C, Z, C, Z, PTEs, CRN, Z, C, Z, C, PTEs, C, Mg, K) e ne e khethiloe ke ICP-OES (Inductively Coupled Plasma Optical Emission Spectroscopy) (Thermo Fisher Scientific, USA) ho latela mekhoa e tloaelehileng le tumellano.Etsa bonnete ba ho netefatsa boleng le taolo (QA/QC) mekhoa (SRM NIST 2711a Montana II Soil) .PTEs e kenyelelitsoe moeli oa thuto ea halofo ea PTE e neng e sa kenyeletsoe. e ne e le 0.0004.(uena).Ho phaella moo, taolo ea boleng le ts'ebetso ea boleng ba boleng bakeng sa tlhahlobo e' ngoe le e 'ngoe e tiisetsoa ka ho hlahloba litekanyetso tsa litšupiso.Ho netefatsa hore liphoso li fokotsehile, ho ile ha etsoa tlhahlobo e habeli.
Empirical Bayesian Kriging (EBK) is one of many geostatistical interpolation techniques used in modelling in different fields such as soil science.Ho fapana le mekhoa e meng ea kriging interpolation, EBK e fapane le mekhoa ea khale ea kriging ka ho nahana ka phoso e hakantsoeng ke mohlala oa semivariogram. Semivariogram.Mekhoa ea ho fetolela e etsa tsela bakeng sa ho hloka botsitso le mananeo a amanang le moralo ona oa semivariogram o etsang karolo e rarahaneng haholo ea mokhoa o lekaneng oa kriging.Tsebetso ea ho fetolela ea EBK e latela mekhoa e meraro e hlahisitsoeng ke Krivoruchko50, (a) mohlala o hakanya boleng ba semivariogram ho tsoa sebakeng se seng le se seng se boletsoeng esale pele sa dataset. semivariogram le (c) ea ho qetela ea A model e khomphuta ho tswa ho etsiso dataset.Molao oa equation oa Bayesian o fanoa e le posterior.
Moo \(Prob\left(A\right)\) e emelang ea pele, \(Prob\left(B\right)\) monyetla o ka thoko o hlokomolohuoa maemong a mangata, \(Prob (B,A)\ ) .Palo ea semivariogram e ipapisitse le molao oa Bayes, o bonts'ang propensity ea observation datasets that of the semiovarisized datasets that can be made the semivariogram. Molao oa Bayes, o bolelang hore na ho na le monyetla o kae oa ho theha dataset ea litebello ho tsoa ho semivariogram.
Mochini oa vector oa ts'ehetso ke mokhoa oa ho ithuta oa mochine o hlahisang hyperplane ea ho arola hantle ho khetholla lihlopha tse ts'oanang empa li sa ikemela ka mokhoa o ikemetseng. Vector Machine Regression - SVMR) e ile ea sebelisoa tlhahlobisong ena.Cherkassky le Mulier53 ba ile ba bula maliboho SVMR e le kernel-based regression, computation ea eona e ileng ea etsoa ho sebelisoa linear regression model with multi-multiple space activities.John et al54 ba tlaleha hore SVMR modelling e sebelisa hyperplane linear regression e lumellang hyperplane linear regression forA. ho Vohland et al. 55, epsilon (ε) -SVMR e sebelisa dataset e koetlisitsoeng ho fumana mohlala oa boemeli e le ts'ebetso ea epsilon-insensitive e sebelisoang ho etsa 'mapa oa data ka mokhoa o ikemetseng le molemo ka ho fetisisa oa epsilon bias ho tloha koetlisong ea data e amanang.Phoso ea sebaka se seng se setiloe e hlokomolohuoa ho tloha boleng ba sebele,' me haeba phoso e kholoanyane ho feta ε(ε), thepa ea mobu ea koetliso e boetse e fokotsa mokhoa o pharaletseng oa ho koetlisa mobu ho fokotsa boholo ba thepa ea mobu. subset of support vectors.The equation e sisintsweng ke Vapnik51 e bontshwa ka tlase.
moo b e emelang moeli oa scalar, \(K\left({x}_{,}{ x}_{k}\right)\) e emetse kernel function, \(\alpha\) e emetse Lagrange multiplier, N E emela dataset ea linomoro, \({x}_{k}\) e emelang ho kenya data, le \.Ornels e sebelisitsoeng ke data, le \.On R) ke senotlolo sa data. ke Gaussian radial basis function (RBF) .RBF kernel e sebelisoa ho fumana mokhoa o nepahetseng oa SVMR, e leng ntho ea bohlokoa ho fumana kotlo e poteletseng ka ho fetisisa e behiloeng C le kernel parameter gamma (γ) bakeng sa data ea koetliso ea PTE. Ntlha ea pele, re ile ra hlahloba setsi sa koetliso ebe re hlahloba ts'ebetso ea mohlala holim'a setsi sa ho netefatsa se sebelisitsoeng.
A multiple linear regression model (MLR) ke mohlala oa regression o emelang kamano e teng lipakeng tsa karabo le mefuta e mengata ea ho bolela esale pele ka ho sebelisa linear pooled parameters tse baloang ka mokhoa oa bonyane ba squares.Ho MLR, a least squares model ke mosebetsi o lebelloang oa thepa ea mobu ka mor'a khetho ea mefuta-futa e hlalosang.Hoa hlokahala ho sebelisa karabo ho theha kamano e sebelisoang e le mokhoa oa ho hlalosa.P. ka mefuta e hlalosang.The MLR equation is
moo y e leng phapang ea karabelo, \(a\) ke sekhechana, n ke palo ea ba lekanyetsang lintho esale pele, \({b}_{1}\) ke phokotso e sa fellang ea li-coefficients, \({x}_{i}\) e emela sephetho kapa tlhaloso e fapaneng,' me \({\varepsilon }_{i}\) e emela phoso, hape e tsejoang e le mohlala.
Mefuta e tsoakiloeng e fumanoe ka sandwiching EBK le SVMR le MLR. Sena se etsoa ka ho ntša litekanyetso tse boletsoeng esale pele ho tsoa ho EBK interpolation. Litekanyetso tse boletsoeng esale pele tse fumanoeng ho tsoa ho Ca, K, le Mg tse kentsoeng li fumanoa ka mokhoa oa ho kopanya ho fumana mefuta e mecha, joalo ka CaK, CaMg, le KMg. CaKMg.Ka kakaretso, mefuta-futa e fumanoeng ke Ca, K, Mg, CaK, CaMg, KMg le CaKMg. Liphetoho tsena li ile tsa fetoha li-predictors tsa rona, tsa thusa ho bolela esale pele likhahla tsa nickel mobung oa litoropo le oa teropo.The algorithm ea SVMR e ile ea etsoa ho li-predictors ho fumana mohlala o tsoakaneng oa Empirical Bayesian Kriging-Supports, EBmilar Machines-Support. ka algorithm ea MLR ho fumana mofuta o tsoakiloeng oa Empirical Bayesian Kriging-Multiple Linear Regression (EBK_MLR). Ka tloaelo, mefuta ea Ca, K, Mg, CaK, CaMg, KMg, le CaKMg e sebelisoa e le li-covariates e le li-predictors tsa Ni content mobung oa litoropo le teropong e fumanehang. ka ho sebelisa kerafo e itlhophisang.Tlhaloso ea mosebetsi oa thuto ena e bontšitsoe setšoantšong sa 2.
Ho sebelisa SeOM e se e le sesebelisoa se tummeng sa ho hlophisa, ho lekola, le ho bolela esale pele lintlha lefapheng la lichelete, tlhokomelo ea bophelo bo botle, indasteri, lipalo-palo, saense ea mobu, le tse ling.SeOM e bōpiloe ka ho sebelisa marang-rang a maiketsetso a neural le mekhoa ea ho ithuta e sa laoleheng bakeng sa mokhatlo, tlhahlobo, le ho bolela esale pele. tlhahlobong ea SeOM li sebelisoa e le n input-dimensional vector variables43,56.Melssen et al. 57 hlalosa ho hokahanngoa ha vector ea ho kenya marang-rang a marang-rang ka lesela le le leng la ho kenya letsoho ho vector ea tlhahiso e nang le vector e le 'ngoe ea boima.Sehlahisoa se hlahisoang ke SeOM ke 'mapa oa mahlakore a mabeli a nang le li-neurone tse fapaneng kapa li-node tse lohiloeng ho limmapa tsa hexagonal, circular, kapa square topological ho latela proximity ea bona.Ho bapisa boholo ba 'mapa bo thehiloeng ho mepotric, OME ea mohlala oa phoso (QEQ) e nang le 0.086 le 0.904, ka ho latellana, e khethiloe, e leng yuniti ea 'mapa oa 55 (5 × 11) .Sebopeho sa neuron se khethoa ho ea ka palo ea li-node ho equation ea empirical.
Palo ea data e sebelisitsoeng thutong ena ke disampole tse 115. Ho ile ha sebelisoa mokhoa o sa reroang oa ho arola lintlha ka lintlha tsa tlhahlobo (25% bakeng sa netefatso) le lihlopha tsa boitsebiso ba koetliso (75% bakeng sa ho lekanya). Lethathamo la boitsebiso ba koetliso le sebelisetsoa ho hlahisa mohlala oa ho fokotsa maemo (calibration), 'me dataset ea teko e sebelisetsoa ho netefatsa bokhoni ba kakaretso58. ts'ebetso ea ho netefatsa sefapano ka makhetlo a leshome, e pheta-phetoang ka makhetlo a mahlano ("Metrics").
Mekhahlelo e fapaneng ea ho netefatsa e ile ea sebelisoa ho fumana mohlala o motle ka ho fetisisa o loketseng ho bolela esale pele likhahla tsa nickel mobung le ho hlahloba ho nepahala ha mohlala le ho netefatsoa ha oona.Mehlala ea Hybridization e ile ea hlahlojoa ho sebelisoa phoso e feletseng (MAE), phoso ea motso oa square (RMSE), le R-squared kapa coefficient determination (R2) . mohlala.RMSE le boholo ba ho fapana ka mehato e ikemetseng e hlalosa matla a ho bolela esale pele a mohlala, ha MAE e etsa qeto ea boleng ba sebele ba palo. Theko ea R2 e tlameha ho ba e phahameng ho hlahloba mohlala o motle ka ho fetisisa oa motsoako ho sebelisa litekanyetso tsa ho netefatsa, ha boleng bo le haufi le ho 1, ho feta ho nepahala.Ho ea ka Li et al. 59, boleng ba criterion ea R2 ea 0.75 kapa ho feta bo nkoa e le selelekela se setle; ho tloha ho 0.5 ho ea ho 0.75 ke ts'ebetso e amohelehang ea mohlala, 'me ka tlase ho 0.5 ke ts'ebetso ea mohlala e sa amoheleheng.Ha u khetha mohlala o sebelisa mekhoa ea ho hlahloba litekanyetso tsa ho netefatsa RMSE le MAE, litekanyetso tse tlaase tse fumanoeng li ne li lekane 'me li nkoa e le khetho e ntle ka ho fetisisa.The equation e latelang e hlalosa mokhoa oa ho netefatsa.
moo n e emelang boholo ba boleng bo hlokometsoeng\({Y}_{i}\) e emelang karabelo e lekantsoeng, mme \({\widehat{Y}}_{i}\) e emela boleng ba karabo e boletsoeng esale pele, ka hona, bakeng sa lipono tsa pele.
Litlhaloso tsa lipalo-palo tsa li-predicor le mefuta-futa ea likarabo li hlahisoa ho Lethathamo la 1, ho bonts'a moelelo, ho kheloha ho tloaelehileng (SD), coefficient of variation (CV), bonyane, boholo, kurtosis, le skewness.Bonyane le boholo ba boleng ba likarolo li fokotseha ka tatellano ea Mg
Likamano tsa mefuta-futa ea li-predicor le likarolo tsa karabelo li bontšitse ho lumellana ho khotsofatsang pakeng tsa likarolo (bona Setšoantšo sa 3) .Khokahano e bontšitse hore CaK e bontšitse kamano e itekanetseng le boleng ba r = 0.53, joalo ka CaNi. Le hoja Ca le K ba bontša likamano tse itekanetseng, bafuputsi ba kang Kingston et al. 68 le Santo69 li fana ka maikutlo a hore maemo a tsona mobung a fapane ka tsela e fapaneng.Leha ho le joalo, Ca le Mg li hanyetsana le K, empa CaK e lumellana hantle.Sena se ka bakoa ke ho sebelisoa ha menontsha e kang potassium carbonate, e leng 56% e phahameng ho potasiamo.Potassium e ne e amana ka mokhoa o itekanetseng le magnesium (KM63 e haufi-ufi, likarolo tsena tse peli tse amanang le potassium). magnesium sulfate, potassium magnesium nitrate, le potash li sebelisoa mobung ho eketsa maemo a bona a khaello.Nickel e ikamahanya ka mokhoa o itekanetseng le Ca, K le Mg le r values = 0.52, 0.63 le 0.55, ka ho latellana.Likamano tse amang calcium, magnesium, le PTEs tse kang calcium nobnesor, calcium nickel, calcium nickel, nickel le magnesium inhibit e fokotsa litlamorao tsa magnesium e feteletseng, 'me ka bobeli magnesium le calcium li fokotsa litlamorao tse chefo tsa nickel mobung.
Matrix ea khokahano bakeng sa likarolo tse bonts'ang kamano pakeng tsa li-predictors le likarabo (Hlokomela: setšoantšo sena se kenyelletsa sekhahla pakeng tsa likarolo, maemo a bohlokoa a thehiloe ho p <0,001).
Setšoantšo sa 4 se bontša kabo ea sebaka sa likarolo.Ho ea ka Burgos et al70, ts'ebeliso ea kabo ea sebaka ke mokhoa o sebelisoang ho lekanya le ho totobatsa libaka tse chesang libakeng tse silafetseng.Maemo a ruisang a Ca setšoantšong sa 4 a ka bonoa karolong e ka leboea-bophirimela ea 'mapa oa kabo ea sebaka. ho ka etsahala hore ebe ka lebaka la tšebeliso ea quicklime (calcium oxide) ho fokotsa acidity ea mobu le tšebeliso ea eona meleng ea tšepe e le oksijene ea alkaline ts'ebetsong ea tšepe. e kanna ea ba ka lebaka la lits'ebetso tsa NPK le potash.Sena se lumellana le lithuto tse ling, tse kang Madaras le Lipavský72, Madaras et al.73, Pulkrabová et al.74, Asare et al.75, ea ileng a hlokomela hore ho tsitsisa ha mobu le kalafo ka KCl le NPK ho ile ha fella ka litaba tse phahameng tsa K mobung. Matlafatso ea Potassium ka leboea-bophirima ho 'mapa oa kabo e ka ba ka lebaka la ts'ebeliso ea menontsha e thehiloeng ho potasiamo joalo ka potassium chloride, potassium sulfate, potassium nitrate, potash le potash ho eketsa litaba tsa potasiamo mobung o futsanehileng.Zádorová et al. 76 le Tlustoš et al. 77 e hlalositse hore ts'ebeliso ea manyolo a thehiloeng ho K e ekelitse litaba tsa K mobung mme e tla eketsa haholo mobu oa limatlafatsi ka nako e telele, haholo-holo K le Mg e bonts'ang sebaka se chesang mobung. Li-hotspots tse leka-lekaneng ka leboea-bophirima ho 'mapa le ka boroa-bochabela ho' mapa. chlorosis. Manyolo a magnesium, joalo ka potassium magnesium sulfate, magnesium sulfate, le Kieserite, a phekola mefokolo (limela li bonahala li pherese, li khubelu kapa li sootho, ho bonts'a khaello ea magnesium) mobung o nang le pH e tloaelehileng6. tlhahiso78.
Kabo ea libaka ['mapa oa kabo ea sebaka o entsoe ho sebelisoa ArcGIS Desktop (ESRI, Inc, Version 10.7, URL: https://desktop.arcgis.com).]
Liphetho tsa index ea ts'ebetso ea mohlala bakeng sa lintlha tse sebelisitsoeng thutong ena li bonts'itsoe ho Lethathamo la 2. Ka lehlakoreng le leng, RMSE le MAE ea Ni ka bobeli li haufi le zero (0.86 RMSE, -0.08 MAE) .Ka lehlakoreng le leng, litekanyetso tsa RMSE le MAE tsa K li amoheleha. Liphetho tsa RMSE le MAE li ne li le kholoanyane bakeng sa calcium le magnesium le liphetho tse fapaneng ka lebaka la data e fapaneng ea MAE le K. RMSE le MAE tsa thuto ena li sebelisa EBK ho bolela esale pele hore Ni li fumanoe li le molemo ho feta liphello tsa John et al. 54 sebelisa synergistic kriging ho noha S concentrations mobung sebelisa data e bokeletsoeng e tšoanang.The EBK outputs re ithutile correlate le tsa Fabijaczyk et al. 41, Yan et al. 79, Beguin et al. 80, Adhikary et al. 81 le Johanne le ba bang. 82, haholo-holo K le Ni.
Ts'ebetso ea mekhoa ea motho ka mong bakeng sa ho lepa li-nickel mobung oa litoropo le o haufi le litoropo e ile ea hlahlojoa ho sebelisoa ts'ebetso ea mehlala (Letlapa la 3) .Boinetefatso ba mohlala le tlhahlobo e nepahetseng e netefalitse hore Ca_Mg_K predictor e kopantsoeng le EBK SVMR model e fane ka ts'ebetso e ntle ka ho fetisisa.Calibration model of square mean Calibration, SVM-square mean model Ca_Mg_K-EBKR2 ea motso oa calibration Calibration RRM2 phoso e feletseng (MAE) e ne e le 0.637 (R2), 95.479 mg/kg (RMSE) le 77.368 mg/kg (MAE) Ca_Mg_K-SVMR e ne e le 0.663 (R2), 235.974 mg/kg (RMSE) le 16kg (lessthewereE) boleng ba R2/2. Ca_Mg_K-SVMR (0.663 mg/kg R2) le Ca_Mg-EBK_SVMR (0.643 = R2); liphetho tsa bona tsa RMSE le MAE li ne li le holimo ho feta tsa Ca_Mg_K-EBK_SVMR (R2 0.637) (sheba Letlapa la 3) . Ho phaella moo, RMSE le MAE ea Ca_Mg-EBK_SVMR (RMSE = 1664.64 le MAE = 1031.49) ea mohlala e kholo ho feta 17. Ca_Mg_K-EBK_SVMR. Ka mokhoa o ts'oanang, RMSE le MAE tsa Ca_Mg-K SVMR (RMSE = 235.974 le MAE = 166.946) li kholoanyane ka 2.5 le 2.2 ho feta tsa Ca_Mg_K-EBK_SVMR li bonts'a hore na data e hlophisoa joang, ho baloa RMSE le MAE ka tatellano. e nang le mohala oa ho lekana hantle.Ho ile ha hlokomeloa RSME e phahameng le MAE.Ho latela Kebonye et al. 46 le john le ba bang. 54, ha RMSE le MAE li le haufi le lefela, liphetho li tla ba betere.SVMR le EBK_SVMR li na le litekanyetso tse phahameng tsa RSME le MAE. Ho ile ha hlokomeloa hore likhakanyo tsa RSME li ne li lula li le holimo ho feta boleng ba MAE, tse bontšang boteng ba ba tsoang kantle ho naha.Ho ea ka Legates, the McCaberate3 (MAE) e khothaletsoa e le sesupo sa boteng ba batho ba tsoang kantle ho naha.Sena se bolela hore ha dataset e fapane haholo, e phahamisa boleng ba MAE le RMSE. Ho nepahala ha tlhahlobo ea netefatso e fapaneng ea Ca_Mg_K-EBK_SVMR e tsoakiloeng ea mohlala bakeng sa ho bolela esale pele Ni content mobung oa litoropo le oa litoropo e ne e le 63.70% ea Litcco. 59, boemo bona ba ho nepahala ke tekanyo e amohelehang ea ts'ebetso ea mohlala.Liphetho tsa hona joale li bapisoa le thuto e fetileng ea Tarasov et al. 36 eo mohlala oa eona oa lebasetere o thehileng MLPRK (Multilayer Perceptron Residual Kriging), e amanang le EBK_SVMR e nepahetseng ea tlhahlobo ea tlhahlobo e tlalehiloeng thutong ea morao-rao, RMSE (210) le The MAE (167.5) e ne e le holimo ho feta liphetho tsa rona thutong ea hona joale (RMSE 95.479, MAE6 ea hona joale ea R2,3 ea hona joale ea R2). (0.637) le ea Tarasov et al. 36 (0.544), ho hlakile hore coefficient of determination (R2) e phahame ka mokhoa ona o tsoakiloeng. Moeli oa phoso (RMSE le MAE) (EBK SVMR) bakeng sa mohlala o tsoakiloeng o ka tlaase ka makhetlo a mabeli. Ka mokhoa o ts'oanang, Sergeev et al.34 o ngotse 0.28 (R2) bakeng sa mofuta o tsoetseng pele oa hybrid (Multila) oa ho ithuta oa morao-rao oa hybrid (Multila) 0.637 (R2).Boemo ba ho bolela esale pele ho nepahala ha mohlala ona (EBK SVMR) ke 63.7%, ha ho nepahala ha ho bolela esale pele ho fumanoe ke Sergeev et al. 34 ke 28%.'Mapa oa ho qetela (setšoantšo sa 5) o entsoeng ka mokhoa oa EBK_SVMR le Ca_Mg_K e le selelekela se bonts'a likhakanyo tsa libaka tse chesang le nickel e itekanetseng ho ea sebakeng sohle sa thuto.Sena se bolela hore bongata ba nickel sebakeng sa thuto ke haholo-holo bo itekanetseng, bo nang le lintlha tse phahameng libakeng tse itseng tse itseng.
'Mapa oa ho bolela esale pele o emeloa ho sebelisoa mofuta o nyalisitsoeng oa EBK_SVMR le ho sebelisoa Ca_Mg_K joalo ka ponelopele.['Mapa oa kabo ea sebaka o entsoe ho sebelisoa RStudio (version 1.4.1717: https://www.rstudio.com/).]
E hlahisitsoe ho Setšoantšo sa 6 ke li-concentrations tsa PTE e le sefofane sa sebopeho se nang le li-neurone tsa motho ka mong.Ha ho le e 'ngoe ea lifofane tsa likarolo tse bontšang mokhoa o tšoanang oa mebala joalokaha ho bontšitsoe.Leha ho le joalo, palo e loketseng ea li-neurons ka 'mapa o mong o entsoeng ke 55.SeOM e hlahisoa ho sebelisoa mebala e sa tšoaneng,' me ha e ntse e tšoana haholoanyane le mebala ea mebala, ho bapisoa le thepa ea lisampole ka bomong, ho latela MCag le likarolo tsa 'mala oa tsona. Ka hona, CaK le CaMg li arolelana lintho tse tšoanang tse nang le li-neurone tse phahameng haholo le mebala e tlaase ho ea ho e leka-lekaneng ea mebala. Ka bobeli li bolela esale pele hore ho na le Niron mobung ka ho bontša mebala e bohareng ho ea holimo ea mebala e kang e khubelu, ea lamunu le e mosehla. mohlala oa likarolo tsa mohlala o bontšitse mokhoa o phahameng oa 'mala o bontšang hore na nickel e ka' na ea e-ba teng mobung (sheba Setšoantšo sa 4) . ea likhakanyo tsa nickel mobung oa litoropo le o haufi le litoropo. Setšoantšo sa 7 se bonts'a mokhoa oa contour ka lihlopha tsa k-mekhoa 'mapeng, li arotsoe ka lihlopha tse tharo ho latela boleng bo boletsoeng esale pele moetsong o mong le o mong. 3 e amohetse mehlala ea 8. Motsoako oa likarolo tse supileng oa planar predictor o ne o nolofalitsoe ho lumella tlhaloso e nepahetseng ea lihlopha.Ka lebaka la mekhoa e mengata ea anthropogenic le ea tlhaho e amang ho thehoa ha mobu, ho thata ho ba le mekhoa e fapaneng e fapaneng ea lihlopha ho 'mapa oa SeOM o abuoang78.
Sephetho sa sefofane se hlahisoang ke mochini o mong le o mong oa Empirical Bayesian Kriging Support Vector Machine (EBK_SVM_SeOM).['Mapa oa SeOM o entsoe ho sebelisoa RStudio (version 1.4.1717: https://www.rstudio.com/).]
Likarolo tse fapaneng tsa sehlopha sa lihlopha ['mapa oa SeOM o entsoe ho sebelisoa RStudio (mofuta oa 1.4.1717: https://www.rstudio.com/).]
Boithuto ba hajoale bo bontša ka ho hlaka mekhoa ea ho etsa mohlala bakeng sa lihloliloeng tsa nickel mobung oa litoropo le metseng e haufi e tiisa kabo ea sebaka sa planar ea likarolo tse bontšitsoeng ke EBK_SVMR (bona Setšoantšo sa 5) . Liphetho li bontša hore mohlala oa ts'ehetso ea mochine oa vector (Ca Mg K-SVMR) o bolela esale pele mahloriso a Ni mobung e le mohlala o le mong, empa litekanyetso tsa ho netefatsa le ho nepahala li bonts'a liphoso tse phahameng haholo ho latela RMSE le mokhoa o mong oa ho sebetsa ka letsoho le MAE. Mohlala oa EBK_MLR o boetse o na le phoso ka lebaka la boleng bo tlase ba coefficient of determination (R2) .Liphetho tse ntle li ile tsa fumanoa ho sebelisoa EBK SVMR le likarolo tse kopantsoeng (CaKMg) tse nang le liphoso tse fokolang tsa RMSE le MAE ka ho nepahala ha 63.7%.Hoa etsahala hore ho kopanya EBK algorithm le mochine oa hybrid concentration algorithm e ka hlahisa algorithm ea ho ithuta ka mochine.TE ho sebelisa Ca Mg K e le li-predictors ho bolela esale pele Nickel sebakeng sa boithuto ho ka ntlafatsa ponelopele ea Ni ka mobung.Sena se bolela hore ts'ebeliso e tsoelang pele ea manyolo a thehiloeng ho nickel le tšilafalo ea mobu ea indasteri ke indasteri ea tšepe e na le tšekamelo ea ho eketsa bongata ba nickel mobung.Phuputso ena e senotse hore mohlala oa EBK o ka fokotsa boemo ba phoso le ho ntlafatsa ho nepahala ha mobu oa litoropong kapa ho ntlafatsa boleng ba mobu. ka kakaretso, re etsa tlhahiso ea ho sebelisa mohlala oa EBK-SVMR ho hlahloba le ho bolela esale pele PTE mobung; ho feta moo, re etsa tlhahiso ea ho sebelisa EBK ho kopanya ka mekhoa e sa tšoaneng ea ho ithuta mochine. leha ho le joalo, ho sebelisa li-covariate tse ngata ho ne ho tla ntlafatsa haholo ts'ebetso ea mohlala, e ka nkoang e le moeli oa mosebetsi oa hona joale.Moeli o mong oa thuto ena ke hore palo ea li-datasets ke 115. Ka hona, haeba ho fanoe ka lintlha tse ngata, ts'ebetso ea mokhoa o hlophisitsoeng o ntlafalitsoeng oa hybridization o ka ntlafatsoa.
PlantProbs.net.Nickel ho Limela le Mobu https://plantprobs.net/plant/nutrientImbalances/sodium.html (E fihletsoe ka la 28 Mmesa 2021).
Kasprzak, KS Nickel e hatela pele ho toxicology ea tikoloho ea sejoale-joale.libaka.toxicology.11, 145-183 (1987).
Cempel, M. & Nikel, G. Nickel: Tlhahlobo ea mehloli ea eona le toxicology ea tikoloho.Polish J. Environment.Stud.15, 375-382 (2006).
Freedman, B. & Hutchinson, TC Kenyeletso e silafatsang e tsoang sepakapakeng le ho bokellana mobung le limela haufi le mochini o qhibilihisang nickel-copper Sudbury, Ontario, Canada.can.J. Bot.58(1), 108-132.https://doi.org/10.1139/b80-014 (1980).
Manyiwa, T. le tse ling
Cabata-Pendias.Kabata-Pendias A. 2011. Latela likarolo mobung le… – Google Scholar https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=Kabata-Pendias+A+2011.+Trace+ Lielemente+mobu+le+lijalo+4th+ed.+New+York+%28NY%29%3A+CRC+Press&btnG= (E fihletsoe ka la 24 Nov 2020).
Almås, A., Singh, B., Agriculture, TS-NJ of & 1995, undefined.Liphello tsa indasteri ea nickel ea Russia ho tsitsipano ea tšepe e boima mobung oa temo le joang bo Soer-Varanger, Norway.agris.fao.org.
Nielsen, GD et al.Nickel absorption le ho boloka metsi a nooang a amana le ho kenngoa ha lijo le nickel sensitivity.toxicology.application.Pharmacodynamics.154, 67-75 (1999).
Costa, M. & Klein, CB Nickel carcinogenesis, phetoho, epigenetics kapa khetho.lipotoloho.Pono ea Bophelo.107, 2 (1999).
Ajman, PC; Ajado, SK; Borůvka, L.; Bini, JKM; Sarkody, VYO; Cobonye, NM; Tlhahlobo ea mekhoa ea lintho tse ka bang chefo: tlhahlobo ea bibliometric.Environmental Geochemistry and Health.Springer Science & Business Media BV 2020.https://doi.org/10.1007/s10653-020-00742-9.
Minasny, B. & McBratney, AB Digital Soil Mapping: Histori e Khutšoanyane le Lithuto tse ling.Geoderma 264, 301–311.https://doi.org/10.1016/j.geoderma.2015.07.017 (2016).
McBratney, AB, Mendonça Santos, ML & Minasny, B. Holima 'mapa oa mobu oa digital.Geoderma 117(1-2), 3-52.https://doi.org/10.1016/S0016-7061(03)00223-4 (2003).
Deutsch.CV Geostatistical Reservoir Modeling,… – Google Scholar https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=CV+Deutsch%2C+2002%2C+Geostatistical+Reservoir+Modeling%2C +Oxford+University+Press%tbcG3C+ 28 Mmesa 2021).
Nako ea poso: Jul-22-2022
