Zikomo chifukwa chochezera Nature.com.Msakatuli wa msakatuli omwe mukugwiritsa ntchito ali ndi chithandizo chochepa cha CSS.Kuti mukhale ndi chidziwitso chabwino kwambiri, tikukulimbikitsani kuti mugwiritse ntchito msakatuli wosinthidwa (kapena muzimitsa mawonekedwe ogwirizana mu Internet Explorer).Pakalipano, kuti titsimikizire kuti kupitirizabe kuthandizidwa, tidzawonetsa malo opanda masitayelo ndi JavaScript.
Kuwonongeka kwa nthaka ndi vuto lalikulu lomwe limadza chifukwa cha ntchito za anthu.Kugawa kwa malo kwa zinthu zomwe zingayambitse poizoni (PTEs) zimasiyana m'madera ambiri a m'matauni ndi ozungulira midzi.Choncho, n'zovuta kulosera zamtundu wa PTEs m'nthaka yoteroyo.Chiwerengero cha zitsanzo 115 chinapezedwa kuchokera ku Frydek Mistek ku Czech Republic.), magnesium nickium) Kukhazikika kunatsimikiziridwa pogwiritsa ntchito inductively cod plasma emission spectrometry.Mayankho osiyanasiyana ndi Ni ndi zoloserazo ndi Ca, Mg, ndi K.Matrix ogwirizana pakati pa kuyankha kosiyana ndi kusinthika kwamtsogolo kumasonyeza mgwirizano wokhutiritsa pakati pa zinthu.Zotsatira zoloserazo zinasonyeza kuti Support Vector Machine Regression (SVMR) inachita bwino, ngakhale kuti muzu wake wa 9 mg ndi square23 (mg45) amatanthauza zolakwa za square/23 (mg45). zolakwika mtheradi (MAE) (166.946 mg/kg) zinali zapamwamba kuposa njira zina zomwe zimagwiritsidwa ntchito.Zitsanzo zosakanikirana za Empirical Bayesian Kriging-Multiple Linear Regression (EBK-MLR) zimagwira ntchito bwino, monga momwe zikuwonetsedwera ndi ma coefficients of determination osachepera 0.1.The Empirical Bayesian Kriging-Sups model (EBK-Sup) chitsanzo chabwino kwambiri cha Bayesian Kriging-Sup yokhala ndi RMSE (95.479 mg/kg) ndi MAE (77.368 mg/kg) ndi kutsimikiza kwapamwamba (R2 = 0.637). Njira yowonetsera ya EBK-SVMR ikuwonetsedwa pogwiritsa ntchito mapu odzipangira okha. Ma neuron ophatikizika mu ndege ya hybrid modelKM-SvR amawonetsa mitundu ingapo ya NiEB-concentration CakMg ndi dothi lozungulira mizinda.Zotsatirazi zikuwonetsa kuti kuphatikiza EBK ndi SVMR ndi njira yothandiza yolosera za kuchuluka kwa Ni mu dothi lakumatauni ndi lakutali.
Nickel (Ni) amaonedwa kuti ndi micronutrient kwa zomera chifukwa zimathandiza kuti mumlengalenga nayitrogeni fixation (N) ndi urea kagayidwe, zonse zofunika kuti mbewu kumera. Kuphatikiza pa chothandizira kumera mbewu, Ni akhoza kuchita ngati fungal ndi bakiteriya inhibitor ndi kulimbikitsa kukula kwa zomera. kumafunika kuthira feteleza wa nickel kuti muwonjezere nitrogen fixation2.Kugwiritsa ntchito feteleza wa nickel mopitilira muyeso kuti nthaka itukuke komanso kuonjezera mphamvu ya nyemba zokonza nayitrojeni m'nthaka mosalekeza kumawonjezera kuchuluka kwa nickel m'nthaka. kutengeka kwa chitsulo monga chakudya chofunikira pakukula kwa zomera 1. Malinga ndi Liu3, Ni yapezeka kuti ndi chinthu chofunika kwambiri cha 17 chofunikira pakukula kwa zomera. Kuwonjezera pa ntchito ya nickel pakukula ndi kukula kwa zomera, anthu amafunikira kuti azigwiritsidwa ntchito zosiyanasiyana. Kuonjezera apo, ma aloyi opangidwa ndi faifi ndi zinthu zopangidwa ndi electroplated akhala akugwiritsidwa ntchito kwambiri mu kitchenware, zida za ballroom, zinthu zamakampani azakudya, magetsi, waya ndi chingwe, ma turbine a jet, implants opangira opaleshoni, nsalu, ndi kupanga zombo 5. kuposa anthropogenic4,6.Magwero achilengedwe a nickel amaphatikizapo kuphulika kwa mapiri, zomera, moto wa nkhalango, ndi zochitika za geological; komabe, magwero a anthropogenic akuphatikizapo mabatire a nickel / cadmium mumakampani azitsulo, electroplating, kuwotcherera kwa arc, dizilo ndi mafuta amafuta, komanso mpweya wotuluka mumlengalenga wochokera kumoto wamalasha ndi zinyalala zowotcha Nickel accumulation7,8.Malingana ndi Freedman ndi Hutchinson9 ndi Manyiwa et. 10, magwero akuluakulu a kuipitsidwa kwa nthaka pamtunda wapafupi ndi pafupi ndi malo omwe ali pafupi ndi nickel-copper-based smelters ndi migodi.Nthaka yapamwamba yozungulira Sudbury nickel-copper refinery ku Canada inali ndi chiwerengero chapamwamba kwambiri cha kuipitsidwa kwa nickel pa 26,000 mg / kg11. Dothi la Norway11.Malingana ndi Alms et al. 12, kuchuluka kwa faifi tambala HNO3-extractable m'dera pamwamba zolimidwa nthaka (nickel kupanga mu Russia) ranged kuchokera 6.25 kuti 136.88 mg/kg, lolingana ndi tanthauzo la 30.43 mg/kg ndi m'munsi ndende ya 25 mg/kg. Malinga ndi kabata 11 m'tawuni feteleza mu dothi la phosphorous mu m'tauni kapena m'nthaka yaulimi Nyengo zotsatizana za mbewu zimatha kulowetsa kapena kuipitsa nthaka.Zotsatira za nickel mwa anthu zimatha kuyambitsa khansa kudzera mu mutagenesis, kuwonongeka kwa chromosomal, kubadwa kwa Z-DNA, kutsekedwa kwa DNA excision kukonza, kapena epigenetic process13.Muzoyesa nyama, nickel yapezeka kuti ili ndi kuthekera koyambitsa zotupa zosiyanasiyana, ndipo ma carcinogenic nickel nickel complexes amatha kukhala ndi zotupa.
Kuwunika kwa kuipitsidwa kwa dothi kwakula kwambiri posachedwapa chifukwa cha zovuta zambiri zokhudzana ndi thanzi zomwe zimabwera chifukwa cha maubwenzi a nthaka ndi zomera, nthaka ndi nthaka, kuwonongeka kwa chilengedwe, ndi kuwunika kwa chilengedwe. very improved predictive soil mapping (PSM) .Malinga ndi Minasny ndi McBratney16, predictive soil mapping (DSM) has proven to be a prominent subdiscipline of soil science.Lagacherie and McBratney, 2006 define DSM as “creation and fill of spatial soil information systems through the use of spatiatory soils and labour Machitidwe opangira ".McBratney et al. 17 ikufotokoza kuti DSM kapena PSM yamakono ndiyo njira yothandiza kwambiri yolosera kapena kupanga mapu ogawa malo a PTEs, mitundu ya nthaka ndi katundu wa nthaka.Geostatistics ndi Machine Learning Algorithms (MLA) ndi njira zowonetsera DSM zomwe zimapanga mapu a digito mothandizidwa ndi makompyuta pogwiritsa ntchito deta yofunika komanso yochepa.
Deutsch18 ndi Olea19 amatanthauzira geostatistics ngati "kusonkhanitsa kwa manambala omwe amakhudzana ndi kuyimira kwa malo, makamaka pogwiritsa ntchito zitsanzo zachikale, monga momwe kusanthula kwanthawi kumasonyezera zanthawi." Makamaka, geostatistics imakhudzanso kuwunika kwamitundu yosiyanasiyana, yomwe imalola Kuwerengera ndikutanthauzira kudalira kwamitengo yamalo kuchokera ku dataset20.Gumiaux et al. 20 ikuwonetsanso kuti kuwunika kwa ma variograms mu geostatistics kumatengera mfundo zitatu, kuphatikiza (a) kuwerengera kuchuluka kwa kulumikizana kwa data, (b) kuzindikira ndi kupanga anisotropy pakusiyana kwa dataset ndi (c) kuwonjezera pa Kuphatikiza kutengera cholakwika chobadwa nacho cha data yoyezera yomwe imasiyanitsidwa ndi zotsatirapo zambiri zapagawo, njira zofananirazi. amagwiritsidwa ntchito mu geostatistics, kuphatikiza kriging wamba, co-kriging, kriging wamba, empirical Bayesian kriging, njira yosavuta ya kriging ndi njira zina zodziwika bwino zomasulira mapu kapena kulosera PTE, mawonekedwe a nthaka, ndi mitundu ya nthaka.
Machine Learning Algorithms (MLA) ndi njira yatsopano yomwe imagwiritsa ntchito magulu akuluakulu a data omwe sali ofanana, omwe amalimbikitsidwa ndi ma aligorivimu omwe amagwiritsidwa ntchito makamaka pamigodi ya data, kuzindikira machitidwe mu data, ndikugwiritsidwa ntchito mobwerezabwereza kumagulu a sayansi ya nthaka ndi ntchito zobwerera. 22 (nkhalango zosawerengeka zowerengera zitsulo zolemera mu dothi laulimi), Sakizadeh et al. 23 (kujambula pogwiritsa ntchito makina othandizira vekitala ndi maukonde opangira neural) kuwononga nthaka) .Kuonjezera apo, Vega et al. 24 (NTHAWI yowonetsera kusungirako zitsulo zolemera ndi kutsatsa m'nthaka) Sun et al. 25 (ntchito ya cubist ndi kugawa Cd mu nthaka) ndi ma aligorivimu ena monga k-oyandikana nawo, generalized boosted regression, ndi boosted regression Mitengo inagwiritsanso ntchito MLA kulosera PTE m'nthaka.
Kugwiritsa ntchito ma algorithms a DSM pakulosera kapena kupanga mapu kumakumana ndi zovuta zingapo.Olemba ambiri amakhulupirira kuti MLA ndi yapamwamba kuposa geostatistics ndi mosemphanitsa.Ngakhale kuti imodzi ili yabwino kuposa ina, kuphatikiza kwa ziwirizi kumawongolera kuchuluka kwa kulondola kwa mapu kapena kulosera mu DSM15.Woodcock ndi Gopal26 Finke27; Pontius ndi Cheuk28 ndi Grunwald29 amathirira ndemanga pa zofooka ndi zolakwika zina pakupanga mapu a nthaka.Asayansi a nthaka ayesa njira zosiyanasiyana kuti akwaniritse bwino, kulondola, ndi kuneneratu kwa mapu a DSM ndi kulosera. 15 ikuwonetsa kuti machitidwe otsimikizira ndi kusatsimikizika koyambitsidwa ndi kupanga mapu ndi kulosera kuyenera kutsimikiziridwa payekhapayekha kuti kuwongolera mapu.Zocheperako za DSM ndichifukwa cha dothi lomwazika, lomwe limaphatikizapo chigawo cha kusatsimikizika; komabe, kusowa kwachidziwitso mu DSM kungabwere kuchokera kuzinthu zambiri zolakwika, zomwe ndi zolakwika za covariate, zolakwika zachitsanzo, zolakwika za malo, ndi Zolakwika za analytical 31.Modelling zolakwika zomwe zinapangidwa mu MLA ndi geostatistical process zimagwirizanitsidwa ndi kusowa kwa kumvetsetsa, potsirizira pake kumabweretsa kuwonjezereka kwa ndondomeko yeniyeni32. Mosasamala kanthu za chikhalidwe, kulephera kwachitsanzo kungathe kufotokozedwa ndi chitsanzo cha parameter, chomwe chikhoza kuperekedwa kwa chitsanzo cha chitsanzo. masamu chitsanzo kulosera, kapena interpolation33.Posachedwapa, njira yatsopano ya DSM yatulukira yomwe imalimbikitsa kusakanikirana kwa geostatistics ndi MLA mu mapu ndi kulosera.Asayansi angapo a nthaka ndi olemba, monga Sergeev et al. 34; Subbotina et al. 35; Tarasov et al. 36 ndi Tarasov et al. 37 agwiritsa ntchito luso lolondola la geostatistics ndi kuphunzira pamakina kuti apange mitundu yosakanizidwa yomwe imapititsa patsogolo luso la kulosera komanso kupanga mapu. Zina mwa mitundu iyi yosakanizidwa kapena yophatikizika ya algorithm ndi 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-Multilayer Perceptron (ANN-Multilayer Perceptron-Musri-Gan-3) Kubwerera m'mbuyo38.
Malinga ndi Sergeev neri Al., kuphatikiza njira zosiyanasiyana zowonetsera zimatha kuthetsa zolakwika ndikuwonjezera luso lachitsanzo chosakanizidwa m'malo mopanga mtundu wake umodzi.Pankhani iyi, pepala latsopanoli likunena kuti ndikofunikira kugwiritsa ntchito ma aligorivimu ophatikizana a geostatistics ndi MLA kuti apange zitsanzo zabwino kwambiri zosakanizidwa kuneneratu za kutukuka kwa Ni m'matauni ndi m'mphepete mwa nyanja. (EBK) monga chitsanzo choyambira ndikusakaniza ndi Support Vector Machine (SVM) ndi Multiple Linear Regression (MLR) zitsanzo.Hybridization ya EBK ndi MLA iliyonse sichidziwika.Mitundu yambiri yosakanikirana yomwe imawoneka ndi yophatikizika ya kriging wamba, yotsalira, yobwereranso, ndipo MLA.EBK ndi njira yomasulira ya geostatistical yomwe imagwiritsa ntchito njira yapakati pa malo omwe amagwiritsidwa ntchito ngati malo apakati gawo losasunthika / losasunthika lokhala ndi zizindikiro zodziwika bwino za malo pamunda, kulola kusintha kwa malo39.EBK yakhala ikugwiritsidwa ntchito m'maphunziro osiyanasiyana, kuphatikizapo kusanthula kagayidwe ka carbon organic mu dothi lafamu40, kuyesa kuwononga nthaka41 ndi kupanga mapu a nthaka42.
Kumbali ina, Self-Organising Graph (SeOM) ndi njira yophunzirira yomwe yakhala ikugwiritsidwa ntchito m'nkhani zosiyanasiyana monga Li et al. 43, Wang ndi al. 44, Hossain Bhuiyan et al. 45 ndi Kebonye et al.46 Dziwani za malo ndi magulu a zinthu.Wang et al. 44 ikufotokoza kuti SeOM ndi njira yamphamvu yophunzirira yomwe imadziwika kuti imatha kupanga magulu ndi kulingalira mavuto osagwirizana ndi mzere.Mosiyana ndi njira zina zozindikiritsa chitsanzo monga kusanthula kwachigawo chachikulu, kusanjana kosadziwika bwino, kusakanikirana kwa hierarchical, ndi kupanga zisankho zamitundu yambiri, SeOM ili bwino pakukonzekera ndi kuzindikira PTE mapatani. 44, SeOM ikhoza kugawa m'magulu agawidwe a ma neuroni okhudzana ndikupereka mawonekedwe apamwamba kwambiri.
Pepalali likufuna kupanga chithunzi cholimba cha mapu olondola kwambiri polosera zamtundu wa faifi tambala m'dothi la m'tawuni ndi m'mphepete mwa tawuni. Choncho, tidzayesa kuyankha mafunso ofufuza omwe angapereke zitsanzo zosakanikirana.Komabe, kodi chitsanzocho ndi cholondola bwanji pa kulosera zomwe mukufuna? Kukhazikika kwa Ni mu dothi la m'tauni kapena m'mphepete mwa tawuni , ndi (d) kagwiritsidwe ntchito ka SeOM kuti apange mapu osinthika kwambiri a kusintha kwa malo a faifi tambala.
Kafukufukuyu akuchitika ku Czech Republic, makamaka m’chigawo cha Frydek Mistek m’chigawo cha Moravia-Silesian (onani Chithunzi 1) . Malo ochitira kafukufukuyu ndi ovuta kwambiri ndipo makamaka ali mbali ya chigawo cha Moravia-Silesian Beskidy, chomwe chili mbali ya kunja kwa mapiri a Carpathian.Malo ophunzirira ali pakati pa 4’0 ndi 2′19 ° 4 ndi 2′19° 0'E, ndipo kutalika kuli pakati pa 225 ndi 327 m; komabe, dongosolo la Koppen la mkhalidwe wa nyengo wa derali limatchedwa Cfb = nyengo yotentha ya m’nyanja ya m’nyanja, Kumakhala mvula yambiri ngakhale m’miyezi yowuma.Kutentha kumasiyana pang’ono m’chaka chonse pakati pa −5 °C ndi 24 °C, sikawirikawiri kutsika pansi -14 °C kapena pamwamba pa 30 °C ndi 5 ° C pachaka, pamene pafupifupi 7 ° C ndi 5 ° C; M47. The akuti kafukufuku dera lonse ndi 1,208 makilomita lalikulu, ndi 39.38% ya nthaka yolimidwa ndi 49.36% ya nkhalango coverage. Komano, dera ntchito mu phunziro ili ndi za 889.8 lalikulu kilomita.Mu ndi mozungulira Ostrava, makampani zitsulo ndi zitsulo ntchito mu zitsulo zitsulo ntchito nickless zitsulo ndi nickless kwambiri. (mwachitsanzo chifukwa kukana dzimbiri mumlengalenga) ndi aloyi zitsulo (nickel kumawonjezera mphamvu ya aloyi pamene kukhalabe ductility wabwino ndi toughness), ndi ulimi kwambiri monga phosphate ntchito fetereza ndi kupanga ziweto ndi kafukufuku angathe magwero faifi tambala m'dera (mwachitsanzo, kuwonjezera faifi tambala kwa anaankhosa kuonjezera mitengo kukula mu ng'ombe ntchito m'madera otsika ntchito kafukufuku wa ng'ombe ndi otsika ntchito kafukufuku ng'ombe). Electroplating, kuphatikizapo electroplating nickel ndi electroless nickel plating process.Nthaka imasiyanitsidwa mosavuta ndi mtundu wa nthaka, kapangidwe kake, ndi carbonate content.Nthaka ya nthaka ndi yapakati mpaka yabwino, yochokera ku zinthu za makolo.Iwo ndi colluvial, alluvial kapena aeolian mu chilengedwe.Madera ena a nthaka amawoneka mottled pamwamba ndi subsoil, nthawi zambiri amakhala ndi konkriti ndi casgnobleaching. dera48.Ndi mtunda woyambira 455.1 mpaka 493.5 m, ma cambisol amalamulira Czech Republic49.
Mapu a malo ophunzirira [Mapu a malo ophunzirira adapangidwa pogwiritsa ntchito ArcGIS Desktop (ESRI, Inc, mtundu 10.7, URL: https://desktop.arcgis.com).]
Zitsanzo zonse za 115 za dothi lapamwamba zinapezedwa kuchokera ku dothi la m'tauni ndi m'mphepete mwa tawuni ku Frydek Mistek. Chitsanzo chomwe chinagwiritsidwa ntchito chinali gridi yokhazikika yokhala ndi zitsanzo za nthaka zomwe zimatalikirana ndi 2 × 2 km motalikirana, ndipo dothi lapamwamba linayesedwa pa kuya kwa 0 mpaka 20 masentimita pogwiritsa ntchito chipangizo cha GPS chogwira dzanja (Leica Zeno 5 bags, Ziplobele zojambulidwa bwino mu phukusi la GPS ndi Ziplobele zojambulidwa bwino). ku laboratory.Zitsanzozo zinali zowumitsidwa ndi mpweya kuti zipange zitsanzo zowonongeka, zowonongeka ndi makina opangira makina (Fritsch disc mphero), ndi sieved (sieve size 2 mm) Ikani 1 gramu ya dothi zouma, homogenized ndi sieved mu mabotolo a teflon olembedwa bwino. HNO3 (pogwiritsa ntchito dispenser zodziwikiratu - mmodzi wa asidi aliyense), kuphimba mopepuka ndi kulola zitsanzo kuima usiku mmene (aqua regia pulogalamu) .Ikani supernatant pa otentha zitsulo mbale (kutentha: 100 W ndi 160 °C) kwa 2 h kuti atsogolere chimbudzi ndondomeko ya zitsanzo, ndiye kuziziritsa 50 ml ya voliyumu ndi digestible 50 milliliters flavour. ml ndi madzi deionized. Pambuyo pake, fyuluta kuchepetsedwa supernatant mu 50 ml PVC chubu ndi madzi deionized. Komanso, 1 ml ya dilution njira kuchepetsedwa ndi 9 ml ya madzi deionized ndi amasefedwera mu 12 ml chubu wokonzekera PTE pseudo-concentration.The ndende ya PTEs, Cnr, Cbn, C, PTEs, CNR, CNB, PTEs, CNB, PTEs, CNB Mg, K) adatsimikiziridwa ndi ICP-OES (Inductively Coupled Plasma Optical Emission Spectroscopy) (Thermo Fisher Scientific, USA) malinga ndi njira zovomerezeka ndi mgwirizano.Sungani ndondomeko ya Quality Assurance and Control (QA / QC) (SRM NIST 2711a Montana II Dothi) .PTEs zomwe zinagwiritsidwa ntchito pofufuza malire a PTE omwe anagwiritsidwa ntchito m'munsimu. anali 0.0004.(inu) .Kuonjezera apo, kuwongolera khalidwe ndi kutsimikizira khalidwe la kusanthula kulikonse kumatsimikiziridwa mwa kusanthula ndondomeko zowonetsera.Kuonetsetsa kuti zolakwikazo zinachepetsedwa, kusanthula kawiri kunachitika.
Empirical Bayesian Kriging (EBK) ndi imodzi mwa njira zambiri zofotokozera za geostatistical interpolation zomwe zimagwiritsidwa ntchito pojambula m'madera osiyanasiyana monga sayansi ya nthaka. Mosiyana ndi njira zina zomasulira za kriging, EBK imasiyana ndi njira zachikhalidwe poganizira zolakwika zomwe zimaganiziridwa ndi chitsanzo cha semivariogram. Njira za semivariogram.Kutanthauzira kumapanga njira ya kusatsimikizika ndi mapulogalamu okhudzana ndi kukonzekera uku kwa semivariogram yomwe imapanga gawo lovuta kwambiri la njira yokwanira ya kriging.Njira yomasulira ya EBK imatsatira njira zitatu zomwe Krivoruchko50, (a) chitsanzocho chimayerekezera semivariogram pamtengo wolowetsedwa wa dataset kuchokera ku dataset yatsopano semivariogram ndi (c) chitsanzo chomaliza cha A chimawerengedwa kuchokera ku dataset yofananira. Lamulo la equation la Bayesian limaperekedwa ngati posterior
Kumene \(Prob\left(A\right)\) imayimira zoyambilira, \(Prob\left(B\right)\) mpata wapambali umanyalanyazidwa nthawi zambiri, \(Prob (B,A)\ ) .Kuwerengera kwa semivariogram kumachokera ku lamulo la Bayes, lomwe likuwonetsa kuthekera kwa kuwunika kwa datasets zomwe zimatha kupangidwa kuchokera ku semivariogram. Lamulo la Bayes, lomwe limafotokoza momwe ndizotheka kupanga zowonera kuchokera ku semivariogram.
Makina othandizira vekitala ndi njira yophunzirira makina yomwe imapanga njira yabwino kwambiri yolekanitsira ma hyperplane kuti athe kusiyanitsa magulu ofanana koma osadziyimira okha. Vector Machine Regression - SVMR) idagwiritsidwa ntchito pakuwunikaku.Cherkassky ndi Mulier53 adachita upainiya SVMR ngati kernel-based regression, mawerengedwe ake omwe adagwiritsidwa ntchito pogwiritsa ntchito mzere wowongolera womwe uli ndi ntchito zamayiko ambiri. ku Vohland et al. 55, epsilon (ε) -SVMR imagwiritsa ntchito deta yophunzitsidwa kuti ipeze chitsanzo choyimira ngati ntchito ya epsilon-insensitive yomwe imagwiritsidwa ntchito kuti ipange mapu a deta pawokha ndi epsilon bias yabwino kuchokera ku maphunziro pa data yogwirizana.Kulakwitsa kwa mtunda wokonzedweratu kumanyalanyazidwa kuchokera ku mtengo weniweni, ndipo ngati cholakwikacho chiri chachikulu kuposa ε(ε), katundu wa nthaka amachepetsanso mtengo wamtengo wapatali wochepetsera. kagawo kakang'ono ka ma vectors othandizira. Equation yomwe idaperekedwa ndi Vapnik51 ikuwonetsedwa pansipa.
pomwe b imayimira scalar threshold, \(K\left({x}_{,}{ x}_{k}\right)\) imayimira ntchito ya kernel, \(\alpha\) imayimira Lagrange multiplier, N Imayimira nambala, \({x}_{k}\) imayimira kulowetsa kwa data, ndipo \.yrnels yogwiritsidwa ntchito ndi makiyi a data, ndi \ (yrne \ SV) ndi makiyi omwe amagwiritsidwa ntchito. ndi Gaussian radial basis function (RBF) .RBF kernel imagwiritsidwa ntchito kuti mudziwe njira yabwino kwambiri ya SVMR, yomwe ndi yofunika kwambiri kuti mupeze chilango chodziwika bwino cha C ndi kernel parameter gamma (γ) pa data yophunzitsira ya PTE.Choyamba, tinayesa ndondomeko yophunzitsira ndikuyesa chitsanzo cha chitsanzo pa ndondomeko yovomerezeka yomwe imagwiritsidwa ntchito.
Multilinear regression model (MLR) ndi njira yochepetsera yomwe imayimira mgwirizano pakati pa kuyankha kwa mayankho ndi mitundu ingapo yolosera pogwiritsa ntchito mizere yolumikizana ndi mizere yowerengedwera pogwiritsa ntchito njira yocheperako. ndi zosintha zofotokozera.Chiwerengero cha MLR ndi
pomwe y ndi kusinthika kwa mayankho, \(a\) ndiko kuphatikizika, n ndi chiwerengero cha zolosera, \({b}_{1}\) ndikuchepetsa pang'ono kwa ma coefficients, \({x}_{i}\) amayimira kulosera kapena kusinthika kofotokozera, ndipo \({\varepsilon}_{i}\) imayimiranso cholakwika chomwe chimadziwikanso ngati chitsanzo.
Mitundu yosakanikirana idapezedwa ndi sandwiching EBK yokhala ndi SVMR ndi MLR. Izi zimachitika potengera zomwe zidanenedweratu kuchokera ku EBK kumasulira. Zomwe zidanenedweratu zomwe zimapezedwa kuchokera ku interpolated Ca, K, ndi Mg zimapezedwa kudzera munjira yophatikizira kupeza zosinthika zatsopano, monga CaK, CaMg, ndi KMg, KMg ndi KMg. CaKMg.Ponseponse, zosinthika zomwe zidapezeka ndi Ca, K, Mg, CaK, CaMg, KMg ndi CaKMg. Zosinthazi zidakhala zolosera zathu, zimathandizira kulosera kuchuluka kwa nickel m'madothi akumidzi ndi m'mphepete mwa tawuni. The algorithm ya SVMR idachitidwa pazolosera kuti apeze mtundu wosakanikirana Empirical Bayesian Kriging-Supports Machines-Suplar (EBmilar SV) kudzera mu aligorivimu ya MLR kuti mupeze mtundu wosakanikirana wa Empirical Bayesian Kriging-Multiple Linear Regression (EBK_MLR). Nthawi zambiri, zosintha za Ca, K, Mg, CaK, CaMg, KMg, ndi CaKMg zimagwiritsidwa ntchito ngati zofananira monga zolosera za Ni zomwe zili m'matauni ndi m'mphepete mwatawuni dothi lovomerezeka kapena EBKM lovomerezeka kapena EBKM lovomerezeka. pogwiritsa ntchito graph yodzipangira yokha.Kayendedwe ka kafukufukuyu akuwonetsedwa pa chithunzi 2.
Kugwiritsa ntchito SeOM kwakhala chida chodziwika bwino chokonzekera, kuyesa, ndi kulosera deta muzachuma, zaumoyo, zamakampani, ziwerengero, sayansi ya nthaka, ndi zina zambiri.SeOM imapangidwa pogwiritsa ntchito maukonde a neural ochita kupanga komanso njira zophunzirira zosayang'aniridwa za bungwe, kuyesa, ndi kulosera. pakuwunika kwa SeOM amagwiritsidwa ntchito ngati n zosintha zamitundu yolowera-dimensional43,56.Melssen et al. 57 ikufotokoza kugwirizana kwa vector yolowera mu neural network kudzera mu gawo limodzi lolowera ku vekitala yotulutsa ndi vector imodzi yolemera. Zomwe zimatulutsidwa ndi SeOM ndi mapu awiri-dimensional omwe ali ndi ma neuroni osiyanasiyana kapena mfundo zolukidwa mu hexagonal, circular, kapena square topological map molingana ndi kuyandikira kwawo. ndi 0.086 ndi 0.904, motsatira, amasankhidwa, omwe ndi mapu a 55 (5 × 11) .
Chiwerengero cha deta chomwe chimagwiritsidwa ntchito mu phunziroli ndi zitsanzo za 115. Njira yowonongeka inagwiritsidwa ntchito kugawanitsa deta mu data yoyesera (25% kuti itsimikizidwe) ndi ma seti a deta yophunzitsira (75% kuti ayesedwe).Dataseti yophunzitsira imagwiritsidwa ntchito popanga chitsanzo cha regression (calibration), ndipo dataset yoyesera imagwiritsidwa ntchito kutsimikizira luso la generalization58.Izi zidachitidwa kuti ziyesere zonse zomwe zimagwiritsidwa ntchito m'nthaka zamitundu yosiyanasiyana. njira khumi zovomerezeka zovomerezeka, zobwerezedwa kasanu.Zosintha zomwe zimapangidwa ndi kutanthauzira kwa EBK zimagwiritsidwa ntchito ngati zolosera kapena zofotokozera zofotokozera kuti ziwonetsere zomwe mukufuna (PTE) .Modeling imayendetsedwa mu RStudio pogwiritsa ntchito laibulale ya phukusi (Kohonen), library (caret), library (modelr), library ("e1071"), library ("plyr" library), library ("plyr") library ("Metrics").
Magawo osiyanasiyana ovomerezeka adagwiritsidwa ntchito kuti adziwe chitsanzo chabwino kwambiri cholosera kuchuluka kwa nickel m'nthaka ndikuyesa kulondola kwa chitsanzocho ndi kutsimikizira kwake.Zitsanzo za Hybridization zinayesedwa pogwiritsa ntchito mean absolute error (MAE), root mean square error (RMSE), ndi R-squared kapena coefficient determination (R2) . chitsanzo.RMSE ndi kukula kwa kusiyana m'miyeso yodziyimira payokha imalongosola mphamvu yolosera zachitsanzo, pamene MAE imatsimikizira mtengo weniweni wochuluka.Mtengo wa R2 uyenera kukhala wokwera kwambiri kuti uyese chitsanzo chabwino kwambiri chosakanikirana pogwiritsa ntchito zizindikiro zovomerezeka, kuyandikira kwa mtengowo ndi 1, kukwezeka kwapamwamba.Malingana ndi Li et al. 59, mtengo wa R2 wa 0.75 kapena kupitilira apo amaonedwa kuti ndiwolosera bwino; kuchokera ku 0.5 kufika ku 0.75 ndi ntchito yovomerezeka yachitsanzo, ndipo pansi pa 0.5 ndi ntchito yosavomerezeka yachitsanzo. Posankha chitsanzo pogwiritsa ntchito njira zowunikira zovomerezeka za RMSE ndi MAE, zikhalidwe zotsika zomwe zinapezedwa zinali zokwanira ndipo zinkaonedwa kuti ndizo zabwino kwambiri.
pomwe n imayimira kukula kwa mtengo wowonedwa\({Y}_{i}\) imayimira yankho loyezedwa, ndipo \({\widehat{Y}}_{i}\) imayimiranso mtengo woyankhidwa wonenedweratu, chifukwa chake, pazowonera zoyamba.
Malongosoledwe a zolosera ndi kuyankha kwamitundu yosiyanasiyana akufotokozedwa mu Gulu 1, kusonyeza tanthauzo, kupatuka kokhazikika (SD), coefficient of variation (CV), osachepera, pazipita, kurtosis, ndi skewness.Zochepa komanso zochulukirapo za zinthuzo zili mu dongosolo lochepera la Mg Chifukwa cha miyeso yosiyana yoyezedwa ya zinthu zomwe zatsatiridwa, kugawa kwa data kwa zinthu kumawonetsa kupotoza kosiyana.Kupindika ndi kurtosis kwa zinthuzo kunali pakati pa 1.53 mpaka 7.24 ndi 2.49 mpaka 54.16, motsatira. Peaked.Ma CV owerengeka a zinthu zimasonyezanso kuti K, Mg, ndi Ni amasonyeza kusinthasintha kwapakati, pamene Ca ali ndi kusiyana kwakukulu kwambiri.Ma CV a K, Ni ndi Mg amafotokoza kugawa kwawo yunifolomu.Kuwonjezerapo, kugawa kwa Ca sikufanana ndipo magwero akunja angakhudze msinkhu wake wolemeretsa.
Kugwirizana kwa zowonetseratu zowonetseratu ndi zinthu zomwe zimayankhidwa zimasonyeza mgwirizano wokhutiritsa pakati pa zinthu (onani Chithunzi 3) .Kugwirizanako kunasonyeza kuti CaK inasonyeza kugwirizanitsa kwapakati ndi r value = 0.53, monga CaNi. 68 ndi Santo69 zikusonyeza kuti milingo yawo m'nthaka ndi yosiyana mosiyanasiyana.Komabe, Ca ndi Mg amatsutsana ndi K, koma CaK amalumikizana bwino.Izi zikhoza kukhala chifukwa cha kugwiritsa ntchito feteleza monga potaziyamu carbonate, yomwe ili pamwamba pa 56% mu potaziyamu.Potaziyamu inali yogwirizana kwambiri ndi magnesiamu (KM6 3, potassium pafupi ndi mafakitale okhudzana ndi 0). magnesium sulfate, potaziyamu magnesium nitrate, ndi potashi amagwiritsidwa ntchito ku dothi kuti awonjezere kuchepa kwawo. Nickel imagwirizana kwambiri ndi Ca, K ndi Mg ndi r values ​​= 0.52, 0.63 ndi 0.55, motero. amachepetsa zotsatira za magnesium owonjezera, ndipo onse magnesium ndi calcium amachepetsa poizoni wa faifi tambala m'nthaka.
Matrix olumikizirana pazinthu zomwe zikuwonetsa ubale pakati pa zolosera ndi mayankho (Zindikirani: chithunzichi chimaphatikizapo scatterplot pakati pa zinthu, milingo yofunikira imachokera pa p <0,001).
Chithunzi cha 4 chikuwonetsa kugawidwa kwa zinthu kwa malo. Malingana ndi Burgos et al70, kugwiritsa ntchito kugawa kwa malo ndi njira yomwe imagwiritsidwa ntchito powerengera ndi kuwonetsa malo otentha m'madera oipitsidwa.Magawo olemera a Ca mu chithunzi 4 akhoza kuwonedwa kumpoto chakumadzulo kwa mapu ogawa malo.Mapuwa akuwonetsa kupititsa patsogolo kwa calcium kwapamwamba kwambiri kumapiri olemera kwambiri a calcium. mwina chifukwa cha ntchito quicklime (kashiamu okusayidi) kuchepetsa acidity nthaka ndi ntchito zitsulo mphero monga alkaline mpweya mu steelmaking ndondomeko.Komanso, alimi ena amakonda kugwiritsa ntchito calcium hydroxide mu dothi acidic neutralize pH, amenenso kumawonjezera kashiamu zili mu nthaka71.Potaziyamu amasonyezanso mawanga otentha kumpoto chakumadzulo ndi kumpoto, chakum'maŵa ndi chigawo chachikulu chaulimi ndi mapu okwera kwambiri a potaziyamu. zikhoza kukhala chifukwa cha ntchito za NPK ndi potashi.Izi zimagwirizana ndi maphunziro ena, monga Madaras ndi Lipavský72, Madaras et al.73, Pulkrabová et al.74, Asare et al.75, omwe adawona kuti kukhazikika kwa nthaka ndi mankhwala ndi KCl ndi NPK kunapangitsa kuti nthaka ikhale ndi K wambiri m'nthaka. Kulemera kwa Potaziyamu kumpoto chakumadzulo kwa mapu ogawa kungakhale chifukwa chogwiritsa ntchito feteleza wa potaziyamu monga potassium chloride, potaziyamu sulfate, potaziyamu nitrate, potashi, ndi potashi kuti awonjezere potaziyamu m'dothi losauka.Zádorová et al. 76 ndi Tlustoš et al. 77 inanena kuti kugwiritsa ntchito feteleza wopangidwa ndi K kumawonjezera kuchuluka kwa K m'nthaka ndipo kungapangitse kuti nthaka ikhale ndi michere yambiri m'nthaka, makamaka K ndi Mg kusonyeza malo otentha m'nthaka. Malo otentha kwambiri kumpoto chakumadzulo kwa mapu ndi kum'mwera chakum'mawa kwa mapu. chlorosis. Magnesium opangidwa ndi feteleza, monga potaziyamu magnesium sulfate, magnesium sulfate, ndi Kieserite, amathandizira zofooka (zomera zimawoneka zofiirira, zofiira, kapena zofiirira, zomwe zikuwonetsa kuchepa kwa magnesium) m'nthaka yokhala ndi pH yanthawi zonse6. kupanga78.
Kugawa kwazinthu [mapu ogawa malo adapangidwa pogwiritsa ntchito ArcGIS Desktop (ESRI, Inc, Version 10.7, URL: https://desktop.arcgis.com).]
Zotsatira za ndondomeko yachitsanzo ya zinthu zomwe zimagwiritsidwa ntchito mu phunziroli zikuwonetsedwa mu Table 2. Kumbali inayi, RMSE ndi MAE ya Ni zonse zili pafupi ndi zero (0.86 RMSE, -0.08 MAE) .Komanso, zonse za RMSE ndi MAE za K ndizovomerezeka. Zotsatira za RMSE ndi MAE zinali zazikulu za calcium ndi magnesium. RMSE ndi MAE a phunziroli pogwiritsa ntchito EBK kulosera za Ni adapezeka kuti ndi abwino kuposa zotsatira za John et al. 54 pogwiritsa ntchito synergistic kriging kulosera za S m'nthaka pogwiritsa ntchito deta yosonkhanitsidwa yomweyi.Zotsatira za EBK zomwe tidaphunzira zimagwirizana ndi za Fabijaczyk et al. 41, Yan et al. 79, Beguin et al. 80, Adhikary et al. 81 ndi Yohane et al. 82, makamaka K ndi Ni.
Kuchita kwa njira zapayekha zolosera za faifi tambala m'dothi la m'tauni ndi m'mphepete mwa tawuni kunawunikidwa pogwiritsa ntchito machitidwe a zitsanzo (Table 3).Kutsimikizira kwachitsanzo ndi kulondola kolondola kunatsimikizira kuti Ca_Mg_K predictor pamodzi ndi EBK SVMR model inapereka ntchito yabwino kwambiri.Calibration model Calibration model RMRM_EBKRMRMRMRMRMRMRM2RMRMRMRM2RMRM2 zolakwa mtheradi (MAE) anali 0.637 (R2), 95.479 mg/kg (RMSE) ndi 77.368 mg/kg (MAE) Ca_Mg_K-SVMR anali 0.663 (R2), 235.974 mg/kg (RMSE) ndi 16kg (zosawerengeka MA, zabwino R2/2000 mg). Ca_Mg_K-SVMR (0.663 mg/kg R2) ndi Ca_Mg-EBK_SVMR (0.643 = R2); zotsatira zawo za RMSE ndi MAE zinali zapamwamba kuposa za Ca_Mg_K-EBK_SVMR (R2 0.637) (onani Table 3) .Kuonjezera apo, RMSE ndi MAE ya Ca_Mg-EBK_SVMR (RMSE = 1664.64 ndi MAE = 1031.49) chitsanzo ndi 134, zazikulu ndi 17.5. Ca_Mg_K-EBK_SVMR. Momwemonso, RMSE ndi MAE a Ca_Mg-K SVMR (RMSE = 235.974 ndi MAE = 166.946) ndi 2.5 ndi 2.2 zazikulu kuposa za Ca_Mg_K-EBK_SVMR momwe zotsatira zakhazikitsira RMSE motsatana. ili ndi mzere wokwanira bwino.Zapamwamba RSME ndi MAE zinawonedwa.Malingana ndi Kebonye et al. 46 ndi Yohane et al. 54, kuyandikira kwa RMSE ndi MAE kufika ku ziro, zotsatira zake zimakhala bwino.SVMR ndi EBK_SVMR ali ndi misinkhu yapamwamba ya RSME ndi MAE. Zinaonedwa kuti kuyerekezera kwa RSME kunali kopambana nthawi zonse kusiyana ndi ma MAE, kusonyeza kukhalapo kwa otuluka.Malinga ndi Legates, a McCabso amatanthauza kupitirira8 kulakwitsa kwa McCabe3 (MAE) ikulimbikitsidwa ngati chisonyezero cha kukhalapo kwa operekera kunja.Izi zikutanthauza kuti zambiri zamtundu wa dataset, zimakhala zapamwamba kwambiri za MAE ndi RMSE.Kulondola kwazitsulo zovomerezeka za Ca_Mg_K-EBK_SVMR chitsanzo chosakanikirana cholosera za Ni zomwe zili m'madothi a m'tawuni ndi m'midzi inali 63.70% ya Litcco. 59, mlingo uwu wa kulondola ndi chitsanzo chovomerezeka cha machitidwe.Zotsatira zomwe zilipo zikufaniziridwa ndi kafukufuku wakale wa Tarasov et al. 36 yemwe mtundu wake wosakanizidwa udapanga MLPRK (Multilayer Perceptron Residual Kriging), yokhudzana ndi EBK_SVMR yowunikira kulondola kwazomwe zanenedwa mu kafukufuku wapano, RMSE (210) ndi The MAE (167.5) zinali zapamwamba kuposa zotsatira zathu mu kafukufuku wapano (RMSE 95.479, MAE68 yaposachedwa ya R2). (0.637) ndi Tarasov et al. 36 (0.544), zikuwonekeratu kuti coefficient of determination (R2) ndi yapamwamba mu chitsanzo ichi chosakanikirana.Mphepete mwa zolakwika (RMSE ndi MAE) (EBK SVMR) ya chitsanzo chosakanikirana ndi chocheperapo kawiri. Momwemonso, Sergeev et al.34 analemba 0.28 (R2) kwa hybrid hybrid model (Multila yojambula pakalipano yophunzira) 0.637 (R2) .Mlingo wolosera molondola wa chitsanzo ichi (EBK SVMR) ndi 63.7%, pamene kulondola kwaneneratu komwe kunapezedwa ndi Sergeev et al. 34 ndi 28%.Mapu omaliza (mkuyu 5) adapangidwa pogwiritsa ntchito chitsanzo cha EBK_SVMR ndi Ca_Mg_K monga chowonetseratu chikuwonetseratu za malo otentha ndi nickel yamtengo wapatali pa malo onse ophunzirira.Izi zikutanthauza kuti kuchuluka kwa nickel m'dera la phunziroli makamaka kumakhala kocheperako, komwe kumakhala kwakukulu m'madera ena enieni.
Mapu olosera omaliza akuimiridwa pogwiritsa ntchito mtundu wosakanizidwa wa EBK_SVMR ndikugwiritsa ntchito Ca_Mg_K monga cholosera.[Mapu ogawa malo adapangidwa pogwiritsa ntchito RStudio (mtundu 1.4.1717: https://www.rstudio.com/).]
Zomwe zikuwonetsedwa mu Chithunzi 6 ndizokhazikika za PTE monga ndege yopangidwa ndi ma neuroni.Palibe ndege zamagulu zomwe zimasonyeza mtundu wofanana wa mtundu monga momwe zasonyezedwera. Momwemonso, CaK ndi CaMg amagawana zofananira ndi ma neuroni apamwamba kwambiri komanso mitundu yotsika mpaka yocheperako.Zitsanzo zonsezi zimalosera kuchuluka kwa Ni munthaka powonetsa mitundu yapakatikati mpaka yayitali yamitundu yofiira, lalanje ndi yachikasu.Mtundu wa KMg ukuwonetsa mitundu yambiri yamitundu yayitali kutengera kulondola kwatsatanetsatane, kugawa kwamtundu wapakatikati ndi kutsika kwamtundu wapakatikati. chitsanzo cha zigawo za chitsanzo chinasonyeza mtundu wapamwamba wa mtundu womwe umasonyeza kuchuluka kwa nickel m'nthaka (onani Chithunzi 4) .Ndege ya gawo lachitsanzo la CakMg imasonyeza mitundu yosiyanasiyana yamitundu kuchokera kumunsi kupita kumtunda malinga ndi mtundu wolondola wa mtundu. za kuchuluka kwa faifi tambala m’dothi la m’matauni ndi m’mphepete mwa tawuni. Chithunzi 7 chikusonyeza njira yozungulira m’magulu a k-njira pa mapu, yogawidwa m’magulu atatu potengera mtengo wonenedweratu wa chitsanzo chilichonse. 3 inalandira zitsanzo za 8. Kuphatikizika kwa zigawo zisanu ndi ziwiri za planar predictor kunaphweka kuti zilole kutanthauzira kolondola kwamagulu.Chifukwa cha njira zambiri za anthropogenic ndi zachilengedwe zomwe zimakhudza mapangidwe a nthaka, zimakhala zovuta kuti tisiyanitse bwino masango mu mapu a SeOM ogawidwa mu78.
Magawo apandege amapangidwa ndi makina aliwonse a Empirical Bayesian Kriging Support Vector Machine (EBK_SVM_SeOM).[Mapu a SeOM adapangidwa pogwiritsa ntchito RStudio (mtundu 1.4.1717: https://www.rstudio.com/).]
Zigawo zosiyanasiyana zamagulu [mapu a SeOM adapangidwa pogwiritsa ntchito RStudio (mtundu 1.4.1717: https://www.rstudio.com/).]
Kafukufuku wapano akuwonetsa momveka bwino njira zopangira nickel mu dothi lakumatauni ndi m'mphepete mwa tawuni. Kafukufukuyu adayesa njira zosiyanasiyana zofananira, kuphatikiza zinthu ndi njira zofananira, kuti apeze njira yabwino yodziwira kuchuluka kwa faifi tambala m'nthaka. imatsimikizira kugawa kwapang'onopang'ono kwa magawo omwe amawonetsedwa ndi EBK_SVMR (onani Chithunzi 5) .Zotsatirazi zikuwonetsa kuti makina othandizira makina oyendetsa makina (Ca Mg K-SVMR) amaneneratu kuchuluka kwa Ni mu dothi ngati chitsanzo chimodzi, koma zovomerezeka ndi zolondola zowunikira zikuwonetsa zolakwika zazikulu kwambiri ponena za RMSE yogwiritsira ntchito njira yogwiritsira ntchito, OOn chitsanzo ndi dzanja la employe. Chitsanzo cha EBK_MLR chilinso cholakwika chifukwa cha mtengo wotsika wa coefficient of determination (R2) . Zotsatira zabwino zinapezedwa pogwiritsa ntchito EBK SVMR ndi zinthu zophatikizika (CaKMg) ndi zolakwika zochepa za RMSE ndi MAE zolondola 63.7%.Zikuoneka kuti kuphatikiza EBK algorithm ndi makina osakanizidwa ophunzirira nthaka algorithms omwe amatha kuwonetseratu ma algorithms a hybrid mu nthaka.TE kugwiritsa ntchito Ca Mg K monga zolosera kulosera za kuchuluka kwa nickel m'dera la kafukufuku kungawongolere kulosera kwa Nickel mu dothi. Izi zikutanthauza kuti kugwiritsa ntchito feteleza wopangidwa ndi faifi kosalekeza ndi kuipitsa nthaka m'mafakitale ndi mafakitale azitsulo kumakhala ndi chizolowezi chowonjezera kuchuluka kwa nickel m'nthaka. zambiri, tikupempha kugwiritsa ntchito chitsanzo cha EBK-SVMR kuti tiyese ndikuwonetseratu PTE m'nthaka; kuonjezerapo, tikuganiza kuti tigwiritse ntchito EBK kuti tigwirizane ndi makina osiyanasiyana ophunzirira makina. komabe, kugwiritsa ntchito ma covariates ambiri kungathandize kwambiri ntchito yachitsanzo, yomwe ingaganizidwe kuti ndi yochepetsera ntchito yomwe ilipo.
PlantProbs.net.Nickel mu Zomera ndi Dothi https://plantprobs.net/plant/nutrientImbalances/sodium.html (Idafikira pa 28 Epulo 2021).
Kasprzak, KS Nickel ikupita patsogolo mu toxicology yamakono ya chilengedwe.surroundings.toxicology.11, 145-183 (1987).
Cempel, M. & Nikel, G. Nickel: Ndemanga ya magwero ake ndi chilengedwe toxicology.Polish J. Environment.Stud.15, 375-382 (2006).
Freedman, B. & Hutchinson, TC Zowononga zowononga kuchokera mumlengalenga ndi kudzikundikira munthaka ndi zomera pafupi ndi chitsulo chosungunula nickel-copper ku Sudbury, Ontario, Canada.can.J. Bot.58(1), 108-132.https://doi.org/10.1139/b80-014 (1980).
Manyiwa, T. et al. Zitsulo zolemera mu nthaka, zomera ndi zoopsa zomwe zimagwirizanitsidwa ndi zoweta msipu pafupi ndi mgodi wa nickel wa Selebi-Phikwe ku Botswana.surroundings.Geochemistry.Health https://doi.org/10.1007/s10653-021-00918-x (2021).
Cabata-Pendias.Kabata-Pendias A. 2011. Trace elements in soil and… – Google Scholar https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=Kabata-Pendias+A+2011.+Trace+ Zinthu+mudothi+ndi+zomera.+4th+ed.+New+York+%28NY%29%3A+CRC+Press&btnG= (Idafikira pa 24 Nov 2020).
Almås, A., Singh, B., Agriculture, TS-NJ of & 1995, undefined.Zotsatira zamakampani a nickel aku Russia pazambiri zazitsulo zolemera mu dothi laulimi ndi udzu ku Soer-Varanger, Norway.agris.fao.org.
Nielsen, GD et al.Mayamwidwe a Nickel ndi kusungidwa m'madzi akumwa amagwirizana ndi kudya ndi nickel sensitivity.toxicology.application.Pharmacodynamics.154, 67-75 (1999).
Costa, M. & Klein, CB Nickel carcinogenesis, mutation, epigenetics kapena select.surroundings.Health Perspective.107, 2 (1999).
Ajman, PC; Ajado, SK; Borůvka, L.; Bini, JKM; Sarkody, VYO; Cobonye, ​​NM; Kusanthula kachitidwe ka zinthu zomwe zitha kukhala poizoni: kuwunika kwa 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: Mbiri Yachidule ndi Maphunziro Ena.Geoderma 264, 301–311.https://doi.org/10.1016/j.geoderma.2015.07.017 (2016).
McBratney, AB, Mendonça Santos, ML & Minasny, B. Pa mapu a nthaka ya digito.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%t7nG3C+ 28 Epulo 2021).