Mahalo no ke kipa ʻana iā Nature.com. Loaʻa i ka mana polokalamu kele pūnaewele āu e hoʻohana nei ke kākoʻo palena no CSS. No ka ʻike maikaʻi loa, paipai mākou iā ʻoe e hoʻohana i kahi polokalamu kele pūnaewele i hōʻano hou ʻia (a i ʻole e hoʻopau i ke ʻano hoʻohālikelike ma Internet Explorer). I kēia manawa, no ka hōʻoia ʻana i ke kākoʻo mau ʻana, e hōʻike mākou i ka pūnaewele me ka ʻole o nā kaila a me JavaScript.
He pilikia nui ka haumia o ka lepo i hoʻokumu ʻia e nā hana a ke kanaka. ʻOkoʻa ka hoʻolaha ʻana o nā mea ʻawahia (PTE) ma ka hapa nui o nā wahi kūlanakauhale a me nā wahi peri-urban. No laila, he paʻakikī ke wānana i ka ʻike o nā PTE i loko o ia mau lepo. Ua loaʻa he 115 mau laʻana mai Frydek Mistek ma ka Czech Republic. Ua hoʻoholo ʻia nā ʻano Calcium (Ca), magnesium (Mg), potassium (K) a me nickel (Ni) me ka hoʻohana ʻana i ka inductively coupled plasma emission spectrometry. ʻO ka loli pane ʻo Ni a ʻo nā mea wānana ʻo Ca, Mg, a me K. Hōʻike ka matrix correlation ma waena o ka loli pane a me ka loli wānana i kahi pilina maikaʻi ma waena o nā mea. Ua hōʻike nā hopena wānana ua hana maikaʻi ʻo Support Vector Machine Regression (SVMR), ʻoiai ʻoi aku ka kiʻekiʻe o kāna hewa kumu awelika square (RMSE) (235.974 mg/kg) a me ka hewa kumu piha (MAE) (166.946 mg/kg) ma mua o nā ʻano hana ʻē aʻe i hoʻohana ʻia. Hana maikaʻi ʻole nā kumu hoʻohālike i hui ʻia no Empirical Bayesian Kriging-Multiple Linear Regression (EBK-MLR), e like me ka hōʻike ʻana ma o nā coefficients o ka hoʻoholo ʻana ma lalo o 0.1. ʻO ke kumu hoʻohālike Empirical Bayesian Kriging-Support Vector Machine Regression (EBK-SVMR) ke kumu hoʻohālike maikaʻi loa, me nā waiwai RMSE haʻahaʻa (95.479 mg/kg) a me MAE (77.368 mg/kg) a me ke coefficient hoʻoholo kiʻekiʻe (R2 = 0.637). Hōʻike ʻia ka hopena o ke ʻano hana hoʻohālike EBK-SVMR me ka hoʻohana ʻana i kahi palapala ʻāina hoʻonohonoho ponoʻī. Hōʻike nā neurons clustered ma ka mokulele o ke kumu hoʻohālike hybrid CakMg-EBK-SVMR i nā ʻano kala he nui e wānana i nā ʻano Ni ma nā lepo kūlanakauhale a me nā peri-urban. Hōʻike nā hopena ʻo ka hoʻohui ʻana iā EBK a me SVMR he ʻenehana kūpono no ka wānana ʻana i nā ʻano Ni ma nā lepo kūlanakauhale a me nā peri-urban.
Ua manaʻo ʻia ʻo Nickel (Ni) he micronutrient no nā mea kanu no ka mea he kōkua ia i ka hoʻopaʻa ʻana o ka nitrogen i ka lewa (N) a me ka metabolism urea, ʻo ia mau mea ʻelua e pono ai no ka ulu ʻana o nā hua. Ma waho aʻe o kona hāʻawi ʻana i ka ulu ʻana o nā hua, hiki iā Ni ke hana ma ke ʻano he mea hoʻopaʻa fungal a me bacteria a hoʻolaha i ka ulu ʻana o nā mea kanu. ʻO ka nele o ka nickel i ka lepo e hiki ai i ka mea kanu ke omo iā ia, e hopena ana i ka chlorosis o nā lau. No ka laʻana, pono nā pipi kau a me nā pīni ʻōmaʻomaʻo i ka hoʻopili ʻana i nā mea hoʻouluulu nickel e hoʻomaikaʻi i ka hoʻopaʻa ʻana o ka nitrogen2. ʻO ka hoʻomau ʻana o ka hoʻopili ʻana i nā mea hoʻouluulu nickel e hoʻonui i ka lepo a hoʻonui i ka hiki o nā legumes ke hoʻopaʻa i ka nitrogen i ka lepo e hoʻonui mau i ka nui o ka nickel i ka lepo. ʻOiai he micronutrient ka nickel no nā mea kanu, ʻo kona lawe nui ʻana i ka lepo hiki ke hana i ka ʻino ma mua o ka maikaʻi. ʻO ka ʻawaʻawa o ka nickel i ka lepo e hoʻemi i ka pH lepo a keakea i ka lawe ʻana o ka hao ma ke ʻano he meaʻai koʻikoʻi no ka ulu ʻana o nā mea kanu1. Wahi a Liu3, ua ʻike ʻia ʻo Ni ka mea nui 17 e pono ai no ka ulu ʻana a me ka ulu ʻana o nā mea kanu. Ma waho aʻe o ke kuleana o ka nickel i ka ulu ʻana a me ka ulu ʻana o nā mea kanu, pono nā kānaka iā ia no nā ʻano noi like ʻole. ʻO ka electroplating, ka hana ʻana o nā mea hoʻouluulu nickel ʻO nā mea hoʻohuihui, a me ka hana ʻana i nā mea hoʻā a me nā plugs spark i ka ʻoihana kaʻa e pono ai ka hoʻohana ʻana i ka nickel ma nā ʻoihana ʻoihana like ʻole. Eia kekahi, ua hoʻohana nui ʻia nā mea hoʻohuihui nickel a me nā mea electroplated i nā mea kuke, nā mea pono ballroom, nā lako ʻoihana meaʻai, nā mea uila, nā uea a me nā kaula, nā turbine jet, nā implants ʻokiʻoki, nā textiles, a me ke kūkulu moku5. Ua hoʻopili ʻia nā pae waiwai Ni i loko o ka lepo (ʻo ia hoʻi, nā lepo ʻili) i nā kumu anthropogenic a me nā kumu kūlohelohe, akā ʻo ka mea nui, he kumu kūlohelohe ʻo Ni ma mua o ka anthropogenic4,6. ʻO nā kumu kūlohelohe o ka nickel e komo pū me nā pele pele, nā mea kanu, nā ahi ululāʻau, a me nā kaʻina hana geological; eia naʻe, ʻo nā kumu anthropogenic e komo pū me nā pila nickel/cadmium i loko o ka ʻoihana kila, electroplating, arc welding, diesel a me nā aila wahie, a me nā hoʻokuʻu lewa mai ka puhi ʻana o ka lanahu a me ka puhi ʻana i ka ʻōpala a me ka sludge. Hōʻiliʻili Nickel7,8. Wahi a Freedman lāua ʻo Hutchinson9 a me Manyiwa et al. 10, ʻo nā kumu nui o ka haumia o ka topsoil ma ke kaiapuni koke a me ka ʻaoʻao e pili ana, ʻo ia nā mea hoʻoheheʻe a me nā lua i hoʻokumu ʻia i ka nickel-copper. ʻO ka lepo luna a puni ka hale hana nickel-copper Sudbury ma Kanada ka mea i loaʻa nā pae kiʻekiʻe loa o ka haumia nickel ma 26,000 mg/kg11. I ka hoʻohālikelike ʻana, ua hopena ka haumia mai ka hana nickel ma Rusia i nā ʻano nickel kiʻekiʻe ma ka lepo Norewai11. Wahi a Alms et al. 12, ʻo ka nui o ka nickel hiki ke unuhi ʻia o HNO3 ma ka ʻāina mahiʻai kiʻekiʻe loa o ka ʻāina (ka hana nickel ma Rusia) mai 6.25 a 136.88 mg/kg, e like me ka awelika o 30.43 mg/kg a me ka nui o ka baseline o 25 mg/kg. Wahi a kabata 11, ʻo ka hoʻopili ʻana o nā mea hoʻouluulu phosphorus i nā lepo mahiʻai ma nā ʻāina kūlanakauhale a i ʻole peri-urban i nā kau huaʻai e hiki ke hoʻokomo a hoʻohaumia paha i ka lepo. ʻO nā hopena hiki o ka nickel i loko o ke kanaka e alakaʻi i ka maʻi kanesa ma o ka mutagenesis, ka hōʻino chromosomal, ka hanauna Z-DNA, ka hoʻoponopono ʻana i ka DNA excision i ālai ʻia, a i ʻole nā kaʻina hana epigenetic13. I nā hoʻokolohua holoholona, ua ʻike ʻia he hiki i ka nickel ke hoʻoulu i nā ʻano tumors like ʻole, a hiki i nā hui nickel carcinogenic ke hoʻonui i kēlā mau tumors.
Ua ulu nui nā loiloi haumia lepo i nā manawa i hala iho nei ma muli o nā pilikia olakino like ʻole e kū mai ana mai nā pilina lepo-mea kanu, nā pilina olaola lepo a me ka lepo, ka hoʻohaʻahaʻa ecological, a me ka loiloi hopena kaiapuni. A hiki i kēia lā, ua hana nui a hoʻopau manawa ka wānana spatial o nā mea ʻawahia (PTE) e like me Ni i ka lepo me ka hoʻohana ʻana i nā ʻano kuʻuna. ʻO ka hiki ʻana mai o ka palapala ʻāina lepo kikohoʻe (DSM) a me kona holomua i kēia manawa15 ua hoʻomaikaʻi nui i ka palapala ʻāina lepo wānana (PSM). Wahi a Minasny lāua ʻo McBratney16, ua hōʻoia ʻia ka palapala ʻāina lepo wānana (DSM) he ʻano koʻikoʻi o ka ʻepekema lepo. Ua wehewehe ʻo Lagacherie lāua ʻo McBratney, 2006 iā DSM ʻo "ka hoʻokumu ʻana a me ka hoʻopiha ʻana i nā ʻōnaehana ʻike lepo spatial ma o ka hoʻohana ʻana i nā ʻano nānā in situ a me ka laboratory a me nā ʻōnaehana inference lepo spatial a me non-spatial".McBratney et al. 17 e hōʻike ana ʻo ka DSM a i ʻole PSM o kēia wā ke ʻano hana kūpono loa no ka wānana ʻana a i ʻole ke kaha kiʻi ʻana i ka hoʻolaha spatial o nā PTE, nā ʻano lepo a me nā waiwai lepo. ʻO Geostatistics a me Machine Learning Algorithms (MLA) nā ʻano hana hoʻohālike DSM e hana ana i nā palapala ʻāina kikohoʻe me ke kōkua o nā kamepiula e hoʻohana ana i ka ʻikepili koʻikoʻi a liʻiliʻi.
Ua wehewehe ʻo Deutsch18 lāua ʻo Olea19 i ka geostatistics ma ke ʻano he "hōʻiliʻili o nā ʻano hana helu e pili ana i ka hōʻike ʻana o nā ʻano spatial, me ka hoʻohana nui ʻana i nā hiʻohiʻona stochastic, e like me ke ʻano o ka loiloi moʻo manawa e hōʻike ai i ka ʻikepili manawa." ʻO ke kumu, pili ka geostatistics i ka loiloi ʻana o nā variograms, e ʻae ana i ka Quantify a wehewehe i nā hilinaʻi o nā waiwai spatial mai kēlā me kēia dataset20. Hōʻike hou ʻo Gumiaux et al. 20 ua hoʻokumu ʻia ka loiloi o nā variograms i ka geostatistics ma luna o ʻekolu mau kumumanaʻo, me (a) ka helu ʻana i ka unahi o ka pilina ʻikepili, (b) ka ʻike ʻana a me ka helu ʻana i ka anisotropy i ka ʻokoʻa dataset a me (c) me ka hoʻohui ʻana i ka hewa kūlohelohe o ka ʻikepili ana i hoʻokaʻawale ʻia mai nā hopena kūloko, ua manaʻo ʻia nā hopena o ka wahi. Ke kūkulu nei ma luna o kēia mau manaʻo, hoʻohana ʻia nā ʻano hana interpolation he nui i ka geostatistics, me ka kriging laulā, co-kriging, kriging maʻamau, empirical Bayesian kriging, ke ʻano kriging maʻalahi a me nā ʻano hana interpolation kaulana ʻē aʻe e palapala a wānana paha i ka PTE, nā ʻano lepo, a me nā ʻano lepo.
ʻO nā Algorithms Aʻo Mīkini (MLA) kahi ʻenehana hou e hoʻohana ana i nā papa ʻikepili non-linear nui aʻe, i hoʻoulu ʻia e nā algorithms i hoʻohana nui ʻia no ka ʻeli ʻikepili, ke ʻike ʻana i nā ʻano ma ka ʻikepili, a hoʻopili pinepine ʻia i ka hoʻokaʻawale ʻana i nā kahua ʻepekema e like me ka ʻepekema lepo a me nā hana hoʻihoʻi. Nui nā pepa noiʻi e hilinaʻi nei i nā hiʻohiʻona MLA e wānana i ka PTE i loko o nā lepo, e like me Tan et al. 22 (nā ululāʻau random no ka helu ʻana o ka metala kaumaha i loko o nā lepo mahiʻai), Sakizadeh et al. 23 (ke hoʻohālike ʻana me ka hoʻohana ʻana i nā mīkini vector kākoʻo a me nā pūnaewele neural artificial) ka haumia lepo). Eia kekahi, ʻo Vega et al. 24 (CART no ke hoʻohālike ʻana i ka paʻa ʻana o ka metala kaumaha a me ka adsorption i loko o ka lepo) Sun et al. 25 (ka noi ʻana o cubist ka hoʻolaha ʻana o Cd i loko o ka lepo) a me nā algorithms ʻē aʻe e like me k-nearest neighbor, generalized boosted regression, a me boosted regression Trees i hoʻopili pū iā MLA e wānana i ka PTE i loko o ka lepo.
ʻO ka hoʻopili ʻana o nā algorithms DSM i ka wānana a i ʻole ka palapala ʻāina e kū nei i kekahi mau pilikia. Manaʻo ka nui o nā mea kākau he ʻoi aku ka maikaʻi o MLA ma mua o ka geostatistics a me ka hope. ʻOiai ʻoi aku ka maikaʻi o kekahi ma mua o kekahi, ʻo ka hui pū ʻana o nā mea ʻelua e hoʻomaikaʻi i ka pae o ka pololei o ka palapala ʻāina a i ʻole ka wānana ma DSM15. ʻO Woodcock lāua ʻo Gopal26 Finke27; ʻO Pontius lāua ʻo Cheuk28 lāua ʻo Grunwald29 e ʻōlelo e pili ana i nā hemahema a me kekahi mau hewa i ka palapala ʻāina lepo i wānana ʻia. Ua hoʻāʻo nā ʻepekema lepo i nā ʻano hana like ʻole e hoʻomaikaʻi i ka pono, ka pololei, a me ka wānana o ka palapala ʻāina DSM a me ka wānana. ʻO ka hui pū ʻana o ka maopopo ʻole a me ka hōʻoia ʻana kekahi o nā ʻano like ʻole i hoʻohui ʻia i loko o DSM e hoʻomaikaʻi i ka pono a hōʻemi i nā hemahema. Eia nō naʻe, ua wehewehe ʻo Agyeman et al. 15 e hōʻoia kūʻokoʻa ʻia ke ʻano hōʻoia a me ka maopopo ʻole i hoʻolauna ʻia e ka hana palapala ʻāina a me ka wānana e hoʻomaikaʻi i ka maikaʻi o ka palapala ʻāina. ʻO nā palena o ka DSM ma muli o ka maikaʻi o ka lepo i hoʻopuehu ʻia ma ka ʻāina, kahi e pili ana i kahi ʻāpana o ka maopopo ʻole; eia naʻe, ʻo ka nele o ka maopopo i ka DSM e kū mai paha mai nā kumu hewa he nui, ʻo ia hoʻi ka hewa covariate, ka hewa hoʻohālike, ka hewa wahi, a me ka hewa analytical 31. ʻO nā hewa hoʻohālike i hoʻokomo ʻia i loko o MLA a me nā kaʻina hana geostatistical e pili ana me ka nele o ka hoʻomaopopo ʻana, e alakaʻi ana i ka hoʻomaʻalahi ʻana o ke kaʻina hana maoli32. Me ka nānā ʻole i ke ʻano o ke hoʻohālike ʻana, hiki ke hoʻopili ʻia nā hewa i nā palena hoʻohālike, nā wānana hoʻohālike makemakika, a i ʻole ka interpolation33. I kēia manawa, ua puka mai kahi ʻano DSM hou e paipai ana i ka hoʻohui ʻana o geostatistics a me MLA i ka palapala ʻāina a me ka wānana. Ua hoʻohana kekahi mau ʻepekema lepo a me nā mea kākau, e like me Sergeev et al. 34; Subbotina et al. 35; Tarasov et al. 36 a me Tarasov et al. 37 i ka maikaʻi pololei o ka geostatistics a me ke aʻo mīkini e hana i nā hiʻohiʻona hybrid e hoʻomaikaʻi i ka pono o ka wānana a me ka palapala ʻāina. ʻano maikaʻi. ʻO kekahi o kēia mau hiʻohiʻona algorithm hybrid a i ʻole i hui pū ʻia ʻo 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-Mulilayer Perceptron (ANN-K-MLP)37 a me Co-Kriging a me Gaussian Process Regression38.
Wahi a Sergeev et al., ʻo ka hoʻohui ʻana i nā ʻano hana hoʻohālike like ʻole he hiki ke hoʻopau i nā hemahema a hoʻonui i ka pono o ke kumu hoʻohālike hybrid hopena ma mua o ka hoʻomohala ʻana i kāna kumu hoʻohālike hoʻokahi. Ma kēia ʻano, ke hoʻopaʻapaʻa nei kēia pepa hou he mea pono e hoʻopili i kahi algorithm hui pū ʻia o geostatistics a me MLA e hana i nā kumu hoʻohālike hybrid kūpono e wānana i ka hoʻonui ʻana o Ni ma nā wahi kūlanakauhale a me nā wahi peri-urban. E hilinaʻi kēia haʻawina iā Empirical Bayesian Kriging (EBK) ma ke ʻano he kumu hoʻohālike kumu a hui pū me Support Vector Machine (SVM) a me nā hiʻohiʻona Multiple Linear Regression (MLR). ʻAʻole ʻike ʻia ka Hybridization o EBK me kekahi MLA. ʻO nā hiʻohiʻona hui like ʻole i ʻike ʻia he mau hui pū ʻana o ka maʻamau, koena, regression kriging, a me MLA. ʻO EBK kahi ʻano interpolation geostatistical e hoʻohana ana i kahi kaʻina hana stochastic spatially i kūloko ʻia ma ke ʻano he kahua random non-stationary/stationary me nā palena localization i wehewehe ʻia ma luna o ke kahua, e ʻae ana i ka loli spatial39. Ua hoʻohana ʻia ʻo EBK i nā ʻano haʻawina like ʻole, me ka nānā ʻana i ka hoʻolaha ʻana o ke kalapona organik i nā lepo mahiʻai40, ka loiloi ʻana i ka haumia lepo41 a me ke kaha palapala ʻana i ka lepo. nā waiwai42.
Ma ka ʻaoʻao ʻē aʻe, ʻo Self-Organizing Graph (SeOM) kahi algorithm aʻo i hoʻopili ʻia ma nā ʻatikala like ʻole e like me Li et al. 43, Wang et al. 44, Hossain Bhuiyan et al. 45 a me Kebonye et al. 46 E hoʻoholo i nā ʻano spatial a me ka hui ʻana o nā mea. Ua wehewehe ʻo Wang et al. 44 he ʻenehana aʻo ikaika ʻo SeOM i ʻike ʻia no kona hiki ke hui a noʻonoʻo i nā pilikia non-linear. ʻAʻole e like me nā ʻano hana ʻike ʻano ʻē aʻe e like me ka nānā ʻana i nā ʻāpana nui, ka hui pū ʻana o ka fuzzy, ka hui pū ʻana o ka hierarchical, a me ka hoʻoholo ʻana i nā ʻano multi-criteria, ʻoi aku ka maikaʻi o SeOM i ka hoʻonohonoho ʻana a me ka ʻike ʻana i nā ʻano PTE. Wahi a Wang et al. 44, hiki iā SeOM ke hui pū i ka hoʻokaʻawale ʻana o nā neurons pili a hāʻawi i ka ʻike ʻikepili kiʻekiʻe. E nānā ʻo SeOM i ka ʻikepili wānana Ni e loaʻa ai ke kumu hoʻohālike maikaʻi loa e wehewehe i nā hopena no ka wehewehe pololei.
ʻO ka pahuhopu o kēia pepa ke hana i kahi kumu hoʻohālike palapala ʻāina paʻa me ka pololei kūpono no ka wānana ʻana i ka ʻike nickel i nā lepo kūlanakauhale a me nā peri-urban. Ke kuhi nei mākou ʻo ka hilinaʻi o ke kumu hoʻohālike i hui pū ʻia e hilinaʻi nui ʻia ma ka mana o nā kumu hoʻohālike ʻē aʻe i hoʻopili ʻia i ke kumu hoʻohālike kumu. Ke ʻae nei mākou i nā pilikia e kū nei i ka DSM, a ʻoiai ke kamaʻilio ʻia nei kēia mau pilikia ma nā ʻaoʻao he nui, ʻo ka hui pū ʻana o nā holomua i nā geostatistics a me nā hiʻohiʻona MLA e like me ka hoʻonui ʻia; no laila, e hoʻāʻo mākou e pane i nā nīnau noiʻi e hiki ke loaʻa nā hiʻohiʻona i hui pū ʻia. Eia naʻe, pehea ka pololei o ke kumu hoʻohālike i ka wānana ʻana i ka mea i manaʻo ʻia? Eia kekahi, he aha ka pae o ka loiloi pono e pili ana i ka hōʻoia a me ka loiloi pololei? No laila, ʻo nā pahuhopu kikoʻī o kēia haʻawina ʻo ia ka (a) hana i kahi kumu hoʻohālike hui pū ʻia no SVMR a i ʻole MLR me ka hoʻohana ʻana iā EBK ma ke ʻano he kumu hoʻohālike kumu, (b) hoʻohālikelike i nā kumu hoʻohālike hopena (c) hāpai i ke kumu hoʻohālike hui maikaʻi loa no ka wānana ʻana i nā ʻano Ni ma nā lepo kūlanakauhale a i ʻole peri-urban, a (d) ka hoʻohana ʻana o SeOM e hana i kahi palapala ʻāina kiʻekiʻe o ka loli spatial nickel.
Ke hana ʻia nei ke aʻo ʻana ma ka Repubalika Czech, ʻo ia hoʻi ma ka ʻāpana ʻo Frydek Mistek ma ka ʻāina ʻo Moravia-Silesian (e nānā i ke Kiʻi 1). He ʻano paʻakikī loa ke ʻano o ka ʻāina o ka ʻāina a ʻo ka hapa nui he ʻāpana ia o ka ʻāina ʻo Moravia-Silesian Beskidy, ʻo ia kekahi ʻāpana o ka lihi o waho o nā mauna ʻo Carpathian. Aia ka ʻāina aʻo ma waena o 49° 41′ 0′ N a me 18° 20′ 0′ E, a ʻo ke kiʻekiʻe ma waena o 225 a me 327 m; Eia naʻe, ua helu ʻia ka ʻōnaehana hoʻokaʻawale ʻana o Koppen no ke kūlana aniau o ka ʻāina ʻo Cfb = ke aniau moana mahana. Nui ka ua i nā mahina maloʻo. ʻOkoʻa iki nā mahana i loko o ka makahiki ma waena o −5 °C a me 24 °C, ʻaʻole pinepine e hāʻule ma lalo o −14 °C a i ʻole ma luna o 30 °C, ʻoiai ʻo ka awelika o ka ua makahiki ma waena o 685 a me 752 mm47. ʻO ka wahi i manaʻo ʻia o ka ʻāina holoʻokoʻa he 1,208 kilomita kuea, me 39.38% o ka ʻāina i mahi ʻia a me 49.36% o ka uhi ʻana o ka ululāʻau. Ma ka ʻaoʻao ʻē aʻe, ʻo ka wahi i hoʻohana ʻia ma kēia noiʻi he 889.8 kilomita kuea. Ma loko a puni ʻo Ostrava, hana nui ka ʻoihana kila a me nā hana metala. ʻO nā wili hao, ka ʻoihana kila kahi e hoʻohana ʻia ai ka nickel i nā kila kila (e laʻa no ke kūʻē ʻana i ka palaho lewa) a me nā kila alloy (hoʻonui ka nickel i ka ikaika o ka alloy me ka mālama ʻana i kona ductility maikaʻi a me ka paʻakikī), a me ka mahiʻai ikaika e like me ka hoʻopili ʻana i ka phosphate fertilizer a me ka hana holoholona he mau kumu noiʻi hiki ke loaʻa o ka nickel ma ka ʻāina (e laʻa, me ka hoʻohui ʻana i ka nickel i nā keiki hipa e hoʻonui i ka ulu ʻana o nā keiki hipa a me nā bipi hānai haʻahaʻa). ʻO nā hoʻohana ʻoihana ʻē aʻe o ka nickel ma nā wahi noiʻi e pili ana i kona hoʻohana ʻana i ka electroplating, me ka electroplating nickel a me nā kaʻina hana electroless nickel plating. Hiki ke hoʻokaʻawale maʻalahi i nā waiwai lepo mai ke kala lepo, ka hoʻonohonoho ʻana, a me ka ʻike carbonate. He waena a maikaʻi ke ʻano o ka lepo, i loaʻa mai ka mea makua. He colluvial, alluvial a aeolian paha lākou i ke ʻano. ʻIke ʻia kekahi mau wahi lepo he mottled ma ka ʻili a me ka subsoil, pinepine me ka sima a me ka bleaching. Eia nō naʻe, ʻo nā cambisols a me nā stagnosols nā ʻano lepo maʻamau ma ka ʻāina48. Me nā kiʻekiʻe mai 455.1 a 493.5 m, noho aliʻi nā cambisols ma Czech Republic49.
Palapala ʻāina o ka wahi haʻawina [Ua hana ʻia ka palapala ʻāina o ka wahi haʻawina me ka hoʻohana ʻana iā ArcGIS Desktop (ESRI, Inc, mana 10.7, URL: https://desktop.arcgis.com).]
Ua loaʻa he 115 mau laʻana lepo luna mai nā lepo kūlanakauhale a me nā lepo peri-urban ma ka ʻāpana ʻo Frydek Mistek. ʻO ke ʻano laʻana i hoʻohana ʻia he grid maʻamau me nā laʻana lepo i hoʻokaʻawale ʻia he 2 × 2 km ke kaʻawale, a ua ana ʻia ka lepo luna ma ka hohonu o 0 a 20 cm me ka hoʻohana ʻana i kahi mea GPS lima (Leica Zeno 5 GPS). Hoʻopili ʻia nā laʻana i loko o nā ʻeke Ziploc, i lepili pono ʻia, a hoʻouna ʻia i ka hale hana. Ua hoʻomaloʻo ʻia nā laʻana i ka ea e hana i nā laʻana i pulverized ʻia, pulverized e kahi ʻōnaehana mechanical (Fritsch disc mill), a kānana ʻia (ka nui o ka sieve 2 mm). E kau i 1 gram o nā laʻana lepo maloʻo, homogenized a sieve ʻia i loko o nā ʻōmole teflon i lepili maopopo ʻia. I loko o kēlā me kēia ipu Teflon, e hāʻawi i 7 ml o 35% HCl a me 3 ml o 65% HNO3 (me ka hoʻohana ʻana i kahi dispenser aunoa - hoʻokahi no kēlā me kēia waikawa), e uhi māmā a e ʻae i nā laʻana e kū i ka pō no ka hopena (polokalamu aqua regia). E kau i ka supernatant ma luna o kahi pā metala wela (mahana: 100 W a me 160 °C) no 2 h e hoʻomaʻalahi i ke kaʻina hana ʻeli ʻana o nā laʻana, a laila e hoʻomaʻalili. E hoʻololi i ka supernatant i kahi ʻōmole volumetric 50 ml a hoʻokahe i 50 ml me ka wai deionized. Ma hope o kēlā, kānana i ka supernatant diluted i loko o kahi paipu PVC 50 ml me ka wai deionized. Eia kekahi, ua hoʻokahe ʻia ka 1 ml o ka hopena dilution me 9 ml o ka wai deionized a kānana ʻia i loko o kahi paipu 12 ml i hoʻomākaukau ʻia no ka PTE pseudo-concentration. Ua hoʻoholo ʻia nā ʻano o nā PTE (As, Cd, Cr, Cu, Mn, Ni, Pb, Zn, Ca, Mg, K) e ICP-OES (Inductively Coupled Plasma Optical Emission Spectroscopy) (Thermo Fisher Scientific, USA) e like me nā ʻano maʻamau a me ka ʻaelike. E hōʻoia i nā kaʻina hana Quality Assurance and Control (QA/QC) (SRM NIST 2711a Montana II Soil). Ua kāpae ʻia nā PTE me nā palena ʻike ma lalo o ka hapalua mai kēia haʻawina. ʻO ka palena ʻike o ka PTE i hoʻohana ʻia ma kēia haʻawina ʻo ia 0.0004.(ʻoe). Eia kekahi, ua hōʻoia ʻia ke kaʻina hana kaohi maikaʻi a me ka hōʻoia maikaʻi no kēlā me kēia loiloi ma o ka nānā ʻana i nā kūlana kuhikuhi. No ka hōʻoia ʻana ua hoʻemi ʻia nā hewa, ua hana ʻia kahi loiloi pālua.
ʻO Empirical Bayesian Kriging (EBK) kekahi o nā ʻano hana interpolation geostatistical he nui i hoʻohana ʻia i ke kumu hoʻohālike ma nā ʻano like ʻole e like me ka ʻepekema lepo. ʻAʻole e like me nā ʻano hana interpolation kriging ʻē aʻe, ʻokoʻa ʻo EBK mai nā ʻano hana kriging kuʻuna ma ka noʻonoʻo ʻana i ka hewa i manaʻo ʻia e ke kumu hoʻohālike semivariogram. Ma ka interpolation EBK, ua helu ʻia kekahi mau hiʻohiʻona semivariogram i ka wā o ka interpolation, ma mua o hoʻokahi semivariogram. Hana nā ʻenehana interpolation i ke ala no ka maopopo ʻole a me ka papahana e pili ana me kēia hoʻolālā ʻana o ka semivariogram e hana ana i kahi ʻāpana paʻakikī loa o kahi ʻano kriging lawa. Ke hahai nei ke kaʻina hana interpolation o EBK i nā pae ʻekolu i hāpai ʻia e Krivoruchko50, (a) ke kuhi nei ke kumu hoʻohālike i ka semivariogram mai ka ʻikepili hoʻokomo (b) ka waiwai wānana hou no kēlā me kēia wahi ʻikepili hoʻokomo e pili ana i ka semivariogram i hana ʻia a (c) ua helu ʻia ke kumu hoʻohālike A hope loa mai kahi ʻikepili simulated. Hāʻawi ʻia ke kānāwai hoʻohālikelike Bayesian ma ke ʻano he posterior
Ma kahi o \(Prob\left(A\right)\) e hōʻike ana i ka mua, \(Prob\left(B\right)\) marginal probability i hoʻowahāwahā ʻia i ka hapa nui o nā hihia, \(Prob (B,A)\). Hoʻokumu ʻia ka helu semivariogram ma ke kānāwai o Bayes, kahi e hōʻike ana i ke ʻano o nā ʻikepili nānā e hiki ke hana ʻia mai nā semivariograms. A laila hoʻoholo ʻia ka waiwai o ka semivariogram me ka hoʻohana ʻana i ke kānāwai o Bayes, kahi e hōʻike ana i ka hiki ke hana i kahi ʻikepili o nā nānā ʻana mai ka semivariogram.
ʻO ka mīkini vector kākoʻo he algorithm aʻo mīkini e hana ana i kahi hyperplane hoʻokaʻawale kūpono e hoʻokaʻawale i nā papa like akā ʻaʻole linearly kūʻokoʻa. Ua hana ʻo Vapnik51 i ka algorithm hoʻokaʻawale manaʻo, akā ua hoʻohana ʻia nei e hoʻoponopono i nā pilikia regression-oriented. Wahi a Li et al.52, ʻo SVM kekahi o nā ʻano hana classifier maikaʻi loa a ua hoʻohana ʻia ma nā ʻano like ʻole. Ua hoʻohana ʻia ka ʻāpana regression o SVM (Support Vector Machine Regression - SVMR) i kēia loiloi. Ua hoʻomaka ʻo Cherkassky lāua ʻo Mulier53 iā SVMR ma ke ʻano he regression kernel-based, ka helu ʻana o ia mea i hana ʻia me ka hoʻohana ʻana i kahi kumu hoʻohālike regression linear me nā hana spatial multi-country. Hōʻike ʻo John et al54 e hoʻohana ana ka hoʻohālike SVMR i ka regression linear hyperplane, kahi e hana ai i nā pilina nonlinear a ʻae i nā hana spatial. Wahi a Vohland et al. 55, hoʻohana ka epsilon (ε)-SVMR i ka ʻikepili i aʻo ʻia e loaʻa ai kahi kumu hoʻohālike e like me kahi hana epsilon-insensitive i hoʻopili ʻia e palapala ʻāina i ka ʻikepili kūʻokoʻa me ka bias epsilon maikaʻi loa mai ke aʻo ʻana ma ka ʻikepili pili. Hoʻowahāwahā ʻia ka hewa mamao i hoʻonohonoho mua ʻia mai ka waiwai maoli, a inā ʻoi aku ka nui o ka hewa ma mua o ε(ε), hoʻoponopono nā waiwai lepo iā ia. Hoʻemi pū ke kumu hoʻohālike i ka paʻakikī o ka ʻikepili aʻo i kahi ʻāpana ākea o nā vectors kākoʻo. Hōʻike ʻia ka hoohalike i hāpai ʻia e Vapnik51 ma lalo nei.
kahi e hōʻike ai ʻo b i ka paepae scalar, ʻo \(K\left({x}_{,}{ x}_{k}\right)\) e hōʻike ana i ka hana kernel, ʻo \(\alpha\) e hōʻike ana i ka mea hoʻonui Lagrange, ʻo N e hōʻike ana i kahi ʻikepili helu, ʻo \({x}_{k}\) e hōʻike ana i ka hoʻokomo ʻikepili, a ʻo \(y\) ka hoʻopuka ʻikepili. ʻO kekahi o nā kernels koʻikoʻi i hoʻohana ʻia ʻo ia ka hana SVMR, ʻo ia kahi hana kumu radial Gaussian (RBF). Hoʻopili ʻia ka kernel RBF e hoʻoholo i ke kumu hoʻohālike SVMR kūpono, he mea koʻikoʻi ia e loaʻa ai ka mea hoʻopaʻi hoʻonohonoho maalea loa C a me ka gamma parameter kernel (γ) no ka ʻikepili hoʻomaʻamaʻa PTE. ʻO ka mea mua, ua loiloi mākou i ka set hoʻomaʻamaʻa a laila hoʻāʻo i ka hana hoʻohālike ma ka set hōʻoia. ʻO ka parameter hoʻokele i hoʻohana ʻia ʻo sigma a ʻo ka waiwai hana ʻo svmRadial.
ʻO ke kumu hoʻohālike regression linear maha (MLR) he kumu hoʻohālike regression e hōʻike ana i ka pilina ma waena o ka loli pane a me kahi helu o nā loli wānana ma ka hoʻohana ʻana i nā palena linear pooled i helu ʻia me ka hoʻohana ʻana i ke ʻano liʻiliʻi loa. Ma MLR, ʻo ke kumu hoʻohālike squares liʻiliʻi he hana wānana o nā waiwai lepo ma hope o ke koho ʻana i nā loli wehewehe. Pono e hoʻohana i ka pane e hoʻokumu i kahi pilina linear me ka hoʻohana ʻana i nā loli wehewehe. Ua hoʻohana ʻia ʻo PTE ma ke ʻano he loli pane e hoʻokumu i kahi pilina linear me nā loli wehewehe. ʻO ka hoʻohālikelike MLR
kahi ʻo y ke ʻano pane, ʻo \(a\) ka intercept, ʻo n ka helu o nā mea wānana, ʻo \({b}_{1}\) ka regression hapa o nā coefficients, ʻo \({x}_{ i}\) ke hōʻike nei i kahi mea wānana a i ʻole ka mea wehewehe, a ʻo \({\varepsilon }_{i}\) ke hōʻike nei i ka hewa i loko o ke kumu hoʻohālike, i ʻike ʻia hoʻi ʻo ke koena.
Ua loaʻa nā hiʻohiʻona hui ʻia ma ke kāwili ʻana iā EBK me SVMR a me MLR. Hana ʻia kēia ma ka unuhi ʻana i nā waiwai i wānana ʻia mai ka interpolation EBK. Loaʻa nā waiwai i wānana ʻia mai ka Ca, K, a me Mg i hoʻopili ʻia ma o kahi kaʻina hana hui e loaʻa ai nā loli hou, e like me CaK, CaMg, a me KMg. A laila hui pū ʻia nā mea Ca, K a me Mg e loaʻa ai kahi loli ʻehā, ʻo CaKMg. Ma keʻano holoʻokoʻa, ʻo nā loli i loaʻa ʻo Ca, K, Mg, CaK, CaMg, KMg a me CaKMg. Ua lilo kēia mau loli i kā mākou mau wānana, e kōkua ana i ka wānana ʻana i nā ʻano nickel ma nā lepo kūlanakauhale a me nā peri-urban. Ua hana ʻia ka algorithm SVMR ma nā wānana e loaʻa ai kahi kumu hoʻohālike hui ʻia ʻo Empirical Bayesian Kriging-Support Vector Machine (EBK_SVM). Pēlā nō, ua hoʻopili ʻia nā loli ma o ka algorithm MLR e loaʻa ai kahi kumu hoʻohālike hui ʻia ʻo Empirical Bayesian Kriging-Multiple Linear Regression (EBK_MLR). ʻO ka maʻamau, ʻo nā loli Ca, K, Mg, Hoʻohana ʻia ʻo CaK, CaMg, KMg, a me CaKMg ma ke ʻano he mau covariates ma ke ʻano he mau wānana o ka nui o Ni ma nā lepo kūlanakauhale a me nā lepo peri-urban. A laila e ʻike ʻia ke kumu hoʻohālike i ʻae ʻia (EBK_SVM a i ʻole EBK_MLR) me ka hoʻohana ʻana i kahi pakuhi hoʻonohonoho ponoʻī. Hōʻike ʻia ke kaʻina hana o kēia haʻawina ma ke Kiʻi 2.
Ua lilo ka hoʻohana ʻana iā SeOM i mea hana kaulana no ka hoʻonohonoho ʻana, loiloi ʻana, a me ka wānana ʻana i ka ʻikepili ma ka ʻāpana kālā, mālama ola kino, ʻoihana, helu helu, ʻepekema lepo, a me nā mea hou aku. Hana ʻia ʻo SeOM me ka hoʻohana ʻana i nā pūnaewele neural artificial a me nā ʻano aʻo i mālama ʻole ʻia no ka hoʻonohonoho ʻana, loiloi, a me ka wānana. Ma kēia haʻawina, ua hoʻohana ʻia ʻo SeOM e ʻike i nā ʻano Ni e pili ana i ke kumu hoʻohālike maikaʻi loa no ka wānana ʻana iā Ni ma nā lepo kūlanakauhale a me nā peri-urban. Hoʻohana ʻia ka ʻikepili i hana ʻia ma ka loiloi SeOM ma ke ʻano he n input-dimensional vector variables43,56.Melssen et al. 57 e wehewehe i ka pilina o kahi vector hoʻokomo i loko o kahi pūnaewele neural ma o kahi papa hoʻokomo hoʻokahi i kahi vector hoʻopuka me kahi vector kaumaha hoʻokahi. ʻO ka hopena i hana ʻia e SeOM he palapala ʻāina ʻelua-dimensional i haku ʻia me nā neurons a i ʻole nā nodes like ʻole i ulana ʻia i loko o nā palapala ʻāina topological hexagonal, circular, a square paha e like me ko lākou kokoke. Ke hoʻohālikelike nei i nā nui palapala ʻāina e pili ana i ka metric, quantization error (QE) a me ka topographic error (TE), ua koho ʻia ke kumu hoʻohālike SeOM me 0.086 a me 0.904, kēlā me kēia, ʻo ia hoʻi kahi ʻāpana 55-map (5 × 11). Hoʻoholo ʻia ke ʻano neuron e like me ka helu o nā nodes i ka empirical equation.
ʻO ka helu o ka ʻikepili i hoʻohana ʻia ma kēia haʻawina he 115 mau laʻana. Ua hoʻohana ʻia kahi ala maʻamau e hoʻokaʻawale i ka ʻikepili i loko o ka ʻikepili hoʻāʻo (25% no ka hōʻoia) a me nā ʻikepili hoʻomaʻamaʻa (75% no ka calibration). Hoʻohana ʻia ka ʻikepili hoʻomaʻamaʻa e hana i ke kumu hoʻohālike regression (calibration), a hoʻohana ʻia ka ʻikepili hoʻāʻo e hōʻoia i ka hiki ke hoʻonui ʻia58. Ua hana ʻia kēia e loiloi i ke kūpono o nā ʻano hoʻohālike like ʻole no ka wānana ʻana i ka ʻike nickel i loko o ka lepo. Ua hele nā hoʻohālike āpau i hoʻohana ʻia ma o kahi kaʻina hana cross-validation he ʻumi manawa, i hana hou ʻia i ʻelima mau manawa. Hoʻohana ʻia nā loli i hana ʻia e ka interpolation EBK ma ke ʻano he mau wānana a i ʻole nā loli wehewehe e wānana i ka loli pahuhopu (PTE). Hoʻokele ʻia ke kumu hoʻohālike ma RStudio me ka hoʻohana ʻana i ka waihona puke (Kohonen), waihona puke (caret), waihona puke (modelr), waihona puke ("e1071"), waihona puke ("plyr"), waihona puke ("caTools"), waihona puke ("prospectr") a me nā waihona puke ("Metrics").
Ua hoʻohana ʻia nā ʻano hōʻoia like ʻole e hoʻoholo i ke kumu hoʻohālike maikaʻi loa no ka wānana ʻana i nā ʻano nickel i ka lepo a e loiloi i ka pololei o ke kumu hoʻohālike a me kona hōʻoia. Ua loiloi ʻia nā hiʻohiʻona Hybridization me ka hoʻohana ʻana i ka mean absolute error (MAE), root mean square error (RMSE), a me R-squared a i ʻole coefficient determination (R2). Ho'ākāka ʻo R2 i ka ʻokoʻa o nā proportion i ka pane, i hōʻike ʻia e ke kumu hoʻohālike regression. Hōʻike ka RMSE a me ka nui o ka variance i nā ana kūʻokoʻa i ka mana wānana o ke kumu hoʻohālike, ʻoiai ʻo MAE e hoʻoholo i ka waiwai quantitative maoli. Pono ke kiʻekiʻe o ka waiwai R2 e loiloi i ke kumu hoʻohālike hui maikaʻi loa me ka hoʻohana ʻana i nā palena hōʻoia, ʻo ka kokoke o ka waiwai i 1, ʻoi aku ka kiʻekiʻe o ka pololei. Wahi a Li et al. 59, ʻo ka waiwai R2 criterion o 0.75 a ʻoi aku paha ua manaʻo ʻia he wānana maikaʻi; mai 0.5 a 0.75 ka hana hoʻohālike i ʻae ʻia, a ma lalo o 0.5 ka hana hoʻohālike i ʻae ʻole ʻia. Ke koho ʻana i kahi kumu hoʻohālike me ka hoʻohana ʻana i nā ʻano loiloi pae hōʻoia RMSE a me MAE, ua lawa nā waiwai haʻahaʻa i loaʻa a ua manaʻo ʻia ʻo ia ke koho maikaʻi loa. Hōʻike ka hoohalike ma lalo nei i ke ʻano hōʻoia.
kahi e hōʻike ai ʻo n i ka nui o ka waiwai i ʻike ʻia\({Y}_{i}\) e hōʻike ana i ka pane i ana ʻia, a ʻo \({\widehat{Y}}_{i}\) hoʻi e hōʻike ana i ka waiwai pane i wānana ʻia, no laila, no nā nānā mua ʻana o i.
Hōʻike ʻia nā wehewehe helu o nā mea wānana a me nā loli pane ma ka Papa 1, e hōʻike ana i ka awelika, ka ʻokoʻa maʻamau (SD), ke koina o ka loli (CV), ka palena iki, ka nui loa, kurtosis, a me ke ʻano ʻē. Aia nā waiwai liʻiliʻi a me ka nui loa o nā mea ma ke ʻano emi o Mg
Ua hōʻike ka pilina o nā loli wānana me nā mea pane i kahi pilina kūpono ma waena o nā mea (e nānā i ke Kiʻi 3). Ua hōʻike ka pilina ua hōʻike ʻo CaK i ka pilina kūpono me ka waiwai r = 0.53, e like me CaNi. ʻOiai ʻo Ca lāua ʻo K e hōʻike ana i nā pilina liʻiliʻi me kekahi, ʻo nā mea noiʻi e like me Kingston et al. Hōʻike ʻo 68 a me Santo69 he kūlike ʻole ko lākou pae ma ka lepo. Eia nō naʻe, kūʻē ʻo Ca a me Mg iā K, akā pili maikaʻi ʻo CaK. Hiki paha kēia ma muli o ka hoʻopili ʻana o nā mea hoʻouluulu e like me ka potassium carbonate, ʻo ia hoʻi he 56% kiʻekiʻe i ka potassium. Ua pili pono ʻo Potassium me ka magnesium (KM r = 0.63). I loko o ka ʻoihana mea hoʻouluulu, pili loa kēia mau mea ʻelua no ka mea ua hoʻopili ʻia ka potassium magnesium sulfate, potassium magnesium nitrate, a me ka potash i ka lepo e hoʻonui i ko lākou mau pae hemahema. Pili pono ʻo Nickel me Ca, K a me Mg me nā waiwai r = 0.52, 0.63 a me 0.55, kēlā me kēia. He paʻakikī nā pilina e pili ana i ka calcium, magnesium, a me nā PTE e like me ka nickel, akā naʻe, pale ka magnesium i ka omo ʻana o ka calcium, hoʻemi ka calcium i nā hopena o ka magnesium keu, a hoʻemi ka magnesium a me ka calcium i nā hopena ʻawahia o ka nickel i ka lepo.
ʻO ka matrix correlation no nā mea e hōʻike ana i ka pilina ma waena o nā mea wānana a me nā pane (Hoʻomaopopo: aia kēia kiʻi i kahi scatterplot ma waena o nā mea, ua hoʻokumu ʻia nā pae koʻikoʻi ma p < 0,001).
Hōʻike ka Kiʻi 4 i ka hoʻolaha ʻana o nā mea i ka spatial. Wahi a Burgos et al70, ʻo ka hoʻohana ʻana o ka hoʻolaha spatial kahi ʻenehana i hoʻohana ʻia e helu a hōʻike i nā wahi wela i nā wahi haumia. Hiki ke ʻike ʻia nā pae hoʻonui o Ca ma ke Kiʻi 4 ma ka ʻaoʻao komohana ʻākau o ka palapala ʻāina hoʻolaha spatial. Hōʻike ke kiʻi i nā wahi wela hoʻonui Ca waena a kiʻekiʻe. ʻO ka hoʻonui ʻana o ka calcium ma ke komohana ʻākau o ka palapala ʻāina ma muli paha o ka hoʻohana ʻana i ka quicklime (calcium oxide) e hōʻemi i ka acidity lepo a me kona hoʻohana ʻana i nā wili kila ma ke ʻano he oxygen alkaline i ke kaʻina hana kila. Ma ka ʻaoʻao ʻē aʻe, makemake nā mahiʻai ʻē aʻe e hoʻohana i ka calcium hydroxide i nā lepo acidic e hoʻopau i ka pH, kahi e hoʻonui ai i ka nui o ka calcium o ka lepo71. Hōʻike pū ka Potassium i nā wahi wela ma ke komohana ʻākau a me ka hikina o ka palapala ʻāina. He kaiāulu mahiʻai nui ka Komohana ʻĀkau, a ʻo ke ʻano waena a kiʻekiʻe o ka potassium ma muli paha o nā noi NPK a me potash. Kūlike kēia me nā haʻawina ʻē aʻe, e like me Madaras lāua ʻo Lipavský72, Madaras et al.73, Pulkrabová et al.74, Asare et al.75, nāna i ʻike i ka hoʻopaʻa ʻana o ka lepo a me ʻO ka mālama ʻana me KCl a me NPK i hopena i ka nui o ka K i loko o ka lepo. ʻO ka hoʻonui ʻana o ka Potassium spatial ma ke komohana ʻākau o ka palapala hoʻolaha ma muli paha o ka hoʻohana ʻana i nā mea hoʻouluulu potassium e like me ka potassium chloride, potassium sulfate, potassium nitrate, potash, a me ka potash e hoʻonui i ka nui o ka potassium o nā lepo ʻilihune.Zádorová et al. 76 a me Tlustoš et al. Ua hōʻike ʻia ma ka 77 ua hoʻonui ka hoʻopili ʻana o nā mea hoʻouluulu K i ka nui o ka K i loko o ka lepo a e hoʻonui nui i ka nui o ka meaʻai o ka lepo i ka wā lōʻihi, ʻoiai ʻo K a me Mg e hōʻike ana i kahi wela i loko o ka lepo. ʻO nā wahi wela ma ke komohana ʻākau o ka palapala ʻāina a me ka hikina hema o ka palapala ʻāina. ʻO ka hoʻopaʻa ʻana o Colloidal i loko o ka lepo e hoʻopau i ka nui o ka magnesium i loko o ka lepo. ʻO kona nele i ka lepo e hōʻike ai i nā mea kanu i ka chlorosis intervein melemele. ʻO nā mea hoʻouluulu magnesium, e like me ka potassium magnesium sulfate, magnesium sulfate, a me Kieserite, e mālama i nā hemahema (ʻike ʻia nā mea kanu he poni, ʻulaʻula, a ʻeleʻele paha, e hōʻike ana i ka nele o ka magnesium) i loko o nā lepo me kahi pae pH maʻamau6. ʻO ka hōʻiliʻili ʻana o ka nickel ma nā ʻili lepo kūlanakauhale a me peri-urban paha ma muli o nā hana anthropogenic e like me ka mahiʻai a me ke koʻikoʻi o ka nickel i ka hana kila kila78.
Ka hoʻolaha ʻāpana o nā mea [ua hana ʻia ka palapala hoʻolaha ʻāpana me ka hoʻohana ʻana iā ArcGIS Desktop (ESRI, Inc, Mana 10.7, URL: https://desktop.arcgis.com).]
Ua hōʻike ʻia nā hopena helu hoʻohālike no nā mea i hoʻohana ʻia ma kēia haʻawina ma ka Papa 2. Ma ka ʻaoʻao ʻē aʻe, kokoke ka RMSE a me ka MAE o Ni i ka ʻole (0.86 RMSE, -0.08 MAE). Ma ka ʻaoʻao ʻē aʻe, ua ʻae ʻia nā waiwai RMSE a me MAE o K. ʻOi aku ka nui o nā hopena RMSE a me MAE no ka calcium a me ka magnesium. ʻOi aku ka nui o nā hopena MAE a me RMSE no Ca a me K ma muli o nā ʻikepili like ʻole. ʻOi aku ka maikaʻi o ka RMSE a me MAE o kēia haʻawina e hoʻohana ana iā EBK e wānana iā Ni ma mua o nā hopena o John et al. 54 e hoʻohana ana i ka synergistic kriging e wānana i nā ʻano S i ka lepo me ka hoʻohana ʻana i ka ʻikepili like i hōʻiliʻili ʻia. Pili nā hopena EBK a mākou i aʻo ai me nā mea o Fabijaczyk et al. 41, Yan et al. 79, Beguin et al. 80, Adhikary et al. 81 a me John et al. 82, ʻoiai ʻo K a me Ni.
Ua loiloi ʻia ka hana o nā ʻano hana pākahi no ka wānana ʻana i ka nui o ka nickel ma nā lepo kūlanakauhale a me nā peri-urban me ka hoʻohana ʻana i ka hana o nā hiʻohiʻona (Papa 3). Ua hōʻoia ka hōʻoia hoʻohālike a me ka loiloi pololei ua loaʻa i ka mea wānana Ca_Mg_K i hui pū ʻia me ke kumu hoʻohālike EBK SVMR ka hana maikaʻi loa. ʻO ke kumu hoʻohālikelike Ca_Mg_K-EBK_SVMR kumu hoʻohālike R2, ka hewa kumu awelika huinahā (RMSE) a me ka hewa awelika piha (MAE) he 0.637 (R2), 95.479 mg/kg (RMSE) a me 77.368 mg/kg (MAE) ʻO Ca_Mg_K-SVMR he 0.663 (R2), 235.974 mg/kg (RMSE) a me 166.946 mg/kg (MAE). Eia naʻe, ua loaʻa nā waiwai R2 maikaʻi no Ca_Mg_K-SVMR (0.663 mg/kg R2) a me Ca_Mg-EBK_SVMR (0.643 = R2); ʻOi aku ke kiʻekiʻe o kā lākou mau hopena RMSE a me MAE ma mua o nā mea no Ca_Mg_K-EBK_SVMR (R2 0.637) (e nānā i ka Papa 3). Eia kekahi, ʻo ka RMSE a me ka MAE o ke kumu hoʻohālike Ca_Mg-EBK_SVMR (RMSE = 1664.64 a me MAE = 1031.49) he 17.5 a me 13.4, kēlā me kēia, ʻoi aku ka nui ma mua o nā mea o Ca_Mg_K-EBK_SVMR. Pēlā nō, ʻo ka RMSE a me ka MAE o ke kumu hoʻohālike Ca_Mg-K SVMR (RMSE = 235.974 a me MAE = 166.946) he 2.5 a me 2.2 ka nui ma mua o nā mea o Ca_Mg_K-EBK_SVMR RMSE a me MAE, kēlā me kēia. Hōʻike nā hopena RMSE i helu ʻia i ke ʻano o ka hoʻopaʻa ʻana o ka ʻikepili me ka laina kūpono loa. Ua ʻike ʻia nā RSME kiʻekiʻe a me MAE. Wahi a Kebonye et al. 46 a me john et al. 54, ʻo ka kokoke loa o ka RMSE a me MAE i ka ʻole, ʻoi aku ka maikaʻi o nā hopena. Loaʻa iā SVMR a me EBK_SVMR nā waiwai RSME a me MAE kiʻekiʻe. Ua ʻike ʻia ua ʻoi aku ka kiʻekiʻe o nā kuhi RSME ma mua o nā waiwai MAE, e hōʻike ana i ke alo o nā outliers. Wahi a Legates lāua ʻo McCabe83, ʻo ka nui o ka RMSE e ʻoi aku ai ma mua o ka mean absolute error (MAE) i ʻōlelo ʻia ma ke ʻano he hōʻailona o ke alo o nā outliers. ʻO ke ʻano kēia, ʻoi aku ka heterogeneous o ka dataset, ʻoi aku ka kiʻekiʻe o nā waiwai MAE a me RMSE. ʻO ka pololei o ka loiloi cross-validation o ke kumu hoʻohālike hui ʻia ʻo Ca_Mg_K-EBK_SVMR no ka wānana ʻana i ka ʻike Ni ma nā lepo kūlanakauhale a me suburban he 63.70%. Wahi a Li et al. 59, ʻo kēia pae o ka pololei he helu hana hoʻohālike i ʻae ʻia. Hoʻohālikelike ʻia nā hopena o kēia manawa me kahi noiʻi mua e Tarasov et al. 36 nona ke kumu hoʻohālike hybrid i hana iā MLPRK (Multilayer Perceptron Residual Kriging), e pili ana i ka helu loiloi pololei EBK_SVMR i hōʻike ʻia ma ke aʻo ʻana o kēia manawa, ʻoi aku ka kiʻekiʻe o RMSE (210) a me ka MAE (167.5) ma mua o kā mākou mau hopena ma ke aʻo ʻana o kēia manawa (RMSE 95.479, MAE 77.368). Eia naʻe, i ka hoʻohālikelike ʻana i ka R2 o ke aʻo ʻana o kēia manawa (0.637) me ko Tarasov et al. 36 (0.544), ua maopopo he ʻoi aku ke kiʻekiʻe o ke koina hoʻoholo (R2) ma kēia kumu hoʻohālike i hui ʻia. ʻO ka palena o ka hewa (RMSE a me MAE) (EBK SVMR) no ke kumu hoʻohālike i hui ʻia he ʻelua manawa haʻahaʻa. Pēlā nō, ua hoʻopaʻa ʻo Sergeev et al.34 i 0.28 (R2) no ke kumu hoʻohālike hybrid i hoʻomohala ʻia (Multilayer Perceptron Residual Kriging), ʻoiai ʻo Ni ma ke aʻo ʻana o kēia manawa i hoʻopaʻa ʻia he 0.637 (R2). ʻO ka pae pololei o ka wānana o kēia kumu hoʻohālike (EBK SVMR) he 63.7%, ʻoiai ʻo ka pololei o ka wānana i loaʻa e Sergeev et al. 34 he 28%. ʻO ka palapala ʻāina hope loa (Kiʻi 5) i hana ʻia me ka hoʻohana ʻana i ke kumu hoʻohālike EBK_SVMR a me Ca_Mg_K ma ke ʻano he wānana e hōʻike ana i nā wānana o nā wahi wela a me ke kaulike a hiki i ka nickel ma luna o ka wahi aʻo holoʻokoʻa. ʻO ke ʻano kēia, ʻo ka nui o ka nickel ma ka wahi aʻo he awelika nui ia, me nā ʻoi aku ka kiʻekiʻe ma kekahi mau wahi kikoʻī.
Hōʻike ʻia ka palapala wānana hope loa me ka hoʻohana ʻana i ke kumu hoʻohālike hybrid EBK_SVMR a me ka hoʻohana ʻana iā Ca_Mg_K ma ke ʻano he wānana. [Ua hana ʻia ka palapala hoʻolaha spatial me ka hoʻohana ʻana iā RStudio (mana 1.4.1717: https://www.rstudio.com/).]
Ua hōʻike ʻia ma ke Kiʻi 6 nā ʻano PTE ma ke ʻano he mokulele haku mele i haku ʻia me nā neurons pākahi. ʻAʻohe o nā mokulele ʻāpana i hōʻike i ke ʻano kala like e like me ka mea i hōʻike ʻia. Eia nō naʻe, ʻo ka helu kūpono o nā neurons no kēlā me kēia palapala i kahakiʻi ʻia he 55. Hana ʻia ʻo SeOM me ka hoʻohana ʻana i nā ʻano kala like ʻole, a ʻo ka like o nā ʻano kala, ʻoi aku ka like o nā waiwai o nā laʻana. Wahi a kā lākou unahi kala pololei, ua hōʻike nā mea pākahi (Ca, K, a me Mg) i nā ʻano kala like me nā neurons kiʻekiʻe hoʻokahi a me ka hapa nui o nā neurons haʻahaʻa. No laila, ua kaʻana like ʻo CaK lāua ʻo CaMg i kekahi mau ʻano like me nā neurons kiʻekiʻe loa a me nā ʻano kala haʻahaʻa a waena. Ua wānana nā hiʻohiʻona ʻelua i ka nui o Ni i ka lepo ma ka hōʻike ʻana i nā hues waena a kiʻekiʻe o nā kala e like me ka ʻulaʻula, ʻalani a me ka melemele. Hōʻike ke kumu hoʻohālike KMg i nā ʻano kala kiʻekiʻe he nui e pili ana i nā ʻāpana kikoʻī a me nā ʻāpana kala haʻahaʻa a waena. Ma kahi unahi kala kikoʻī mai ka haʻahaʻa a kiʻekiʻe, ua hōʻike ke ʻano hoʻolaha planar o nā ʻāpana o ke kumu hoʻohālike i kahi ʻano kala kiʻekiʻe e hōʻike ana i ka hiki ke hoʻohui ʻia o ka nickel i ka lepo (e nānā i ke Kiʻi 4). Hōʻike ka mokulele ʻāpana kumu hoʻohālike CakMg i kahi ʻano kala like ʻole mai ka haʻahaʻa a kiʻekiʻe e like me ke kala pololei. unahi. Eia kekahi, ua like ka wānana o ke kumu hoʻohālike no ka ʻike nickel (CakMg) me ka hoʻolaha spatial o ka nickel i hōʻike ʻia ma ke Kiʻi 5. Hōʻike nā kiʻi ʻelua i nā ʻāpana kiʻekiʻe, waena a me ka haʻahaʻa o nā ʻano nickel ma nā lepo kūlanakauhale a me nā peri-urban. Hōʻike ke Kiʻi 7 i ke ʻano contour ma ka hui k-means ma ka palapala ʻāina, i māhele ʻia i ʻekolu mau hui e pili ana i ka waiwai i wānana ʻia i kēlā me kēia kumu hoʻohālike. Hōʻike ke ʻano contour i ka helu kūpono o nā hui. No nā laʻana lepo 115 i hōʻiliʻili ʻia, ua loaʻa i ka māhele 1 ka hapa nui o nā laʻana lepo, 74. Ua loaʻa i ka Cluster 2 he 33 mau laʻana, ʻoiai ua loaʻa i ka cluster 3 he 8 mau laʻana. Ua hoʻomaʻalahi ʻia ka hui wānana planar ʻehiku-ʻāpana e ʻae i ka wehewehe pololei ʻana o ka cluster. Ma muli o nā kaʻina hana anthropogenic a me nā mea kūlohelohe e pili ana i ka hoʻokumu ʻana o ka lepo, he paʻakikī ke loaʻa nā ʻano cluster i hoʻokaʻawale pono ʻia ma kahi palapala ʻāina SeOM i hoʻolaha ʻia78.
Ka hoʻopuka ʻana o ka mokulele ʻāpana e kēlā me kēia loli Empirical Bayesian Kriging Support Vector Machine (EBK_SVM_SeOM). [Ua hana ʻia nā palapala ʻāina SeOM me ka hoʻohana ʻana iā RStudio (mana 1.4.1717: https://www.rstudio.com/).]
Nā ʻāpana hoʻokaʻawale hui like ʻole [Ua hana ʻia nā palapala ʻāina SeOM me ka hoʻohana ʻana iā RStudio (mana 1.4.1717: https://www.rstudio.com/).]
Hōʻike maopopo ke aʻo ʻana o kēia manawa i nā ʻano hana hoʻohālike no nā ʻano nickel i nā lepo kūlanakauhale a me nā peri-urban. Ua hoʻāʻo ke aʻo ʻana i nā ʻano hana hoʻohālike like ʻole, e hoʻohui ana i nā mea me nā ʻano hana hoʻohālike, e loaʻa ai ke ala maikaʻi loa e wānana ai i nā ʻano nickel i ka lepo. Ua hōʻike nā hiʻohiʻona spatial planar compositional SeOM o ke ʻano hana hoʻohālike i kahi ʻano kala kiʻekiʻe mai ka haʻahaʻa a i ke kiʻekiʻe ma kahi pālākiō kala pololei, e hōʻike ana i nā ʻano Ni i ka lepo. Eia nō naʻe, ua hōʻoia ka palapala hoʻolaha spatial i ka hoʻolaha spatial planar o nā ʻāpana i hōʻike ʻia e EBK_SVMR (e nānā i ke Kiʻi 5). Hōʻike nā hopena i ka wānana o ke kumu hoʻohālike regression mīkini vector kākoʻo (Ca Mg K-SVMR) i ka ʻano o Ni i ka lepo ma ke ʻano he kumu hoʻohālike hoʻokahi, akā hōʻike nā ʻōkuhi hōʻoia a me ka loiloi pololei i nā hewa kiʻekiʻe loa ma ke ʻano o RMSE a me MAE. Ma ka ʻaoʻao ʻē aʻe, ua hewa pū ke ʻano hana hoʻohālike i hoʻohana ʻia me ke kumu hoʻohālike EBK_MLR ma muli o ka waiwai haʻahaʻa o ka coefficient of determination (R2). Ua loaʻa nā hopena maikaʻi me ka hoʻohana ʻana iā EBK SVMR a me nā mea i hui pū ʻia (CaKMg) me nā hewa RMSE a me MAE haʻahaʻa me ka pololei o 63.7%. Ua ʻike ʻia ʻo ka hoʻohui ʻana i ka algorithm EBK me kahi algorithm aʻo mīkini hiki ke hana i kahi algorithm hybrid e hiki ke wānana i ka nui o nā PTE i ka lepo. Hōʻike nā hopena i ka hoʻohana ʻana iā Ca Mg K ma ke ʻano he mau wānana e wānana i nā nui Ni ma kahi o ke aʻo ʻana hiki ke hoʻomaikaʻi i ka wānana o Ni i nā lepo. ʻO ke ʻano kēia, ʻo ka hoʻopili mau ʻana o nā mea hoʻouluulu nickel a me ka haumia ʻoihana o ka lepo e ka ʻoihana kila he ʻano ia e hoʻonui ai i ka nui o ka nickel i ka lepo. Ua hōʻike kēia haʻawina e hiki i ke kumu hoʻohālike EBK ke hōʻemi i ka pae o ka hewa a hoʻomaikaʻi i ka pololei o ke kumu hoʻohālike o ka hoʻolaha spatial lepo ma nā lepo kūlanakauhale a peri-urban paha. Ma keʻano laulā, ke manaʻo nei mākou e hoʻopili i ke kumu hoʻohālike EBK-SVMR e loiloi a wānana i ka PTE i ka lepo; eia kekahi, ke manaʻo nei mākou e hoʻohana iā EBK e hybridize me nā algorithms aʻo mīkini like ʻole. Ua wānana ʻia nā nui Ni me ka hoʻohana ʻana i nā mea e like me nā covariates; akā naʻe, ʻo ka hoʻohana ʻana i nā covariates hou aʻe e hoʻomaikaʻi nui i ka hana o ke kumu hoʻohālike, hiki ke manaʻo ʻia he palena o ka hana o kēia manawa. ʻO kekahi palena o kēia haʻawina, ʻo ia ka helu o nā ʻikepili he 115. No laila, inā hāʻawi ʻia nā ʻikepili hou aku, hiki ke hoʻomaikaʻi ʻia ka hana o ke ʻano hybridization i manaʻo ʻia.
PlantProbs.net.Nickel i nā mea kanu a me ka lepo https://plantprobs.net/plant/nutrientImbalances/sodium.html (Loaʻa ʻia ma 28 ʻApelila 2021).
Kasprzak, KS Nickel holomua i ka toxicology kaiapuni hou.surroundings.toxicology.11, 145–183 (1987).
Cempel, M. & Nikel, G. Nickel: He loiloi o kona mau kumu a me ka toxicology kaiapuni. Polish J. Environment.Stud.15, 375–382 (2006).
ʻO Freedman, B. & Hutchinson, TC Ka hoʻokomo ʻana o ka mea haumia mai ka lewa a me ka hōʻiliʻili ʻana i ka lepo a me nā mea kanu kokoke i kahi mea hoʻoheheʻe nickel-keleawe ma Sudbury, Ontario, Kanada.can.J. Bot.58(1), 108-132.https://doi.org/10.1139/b80-014 (1980).
Manyiwa, T. et al. Nā metala kaumaha i ka lepo, nā mea kanu a me nā pilikia e pili ana me nā ruminants e ʻai ana ma kahi kokoke i ka lua keleawe-nickel Selebi-Phikwe ma Botswana. surroundings.Geochemistry.Health https://doi.org/10.1007/s10653-021-00918-x (2021).
ʻO Cabata-Pendias.Kabata-Pendias A. 2011. Nā mea trace i loko o ka lepo a me… – Google Scholar https://scholar.google.com/scholar?hl=en&as_sdt=0%2C5&q=Kabata-Pendias+A.+2011.+Trace+ Elements+in+soils+and+plants.+4th+ed.+New+York+%28NY%29%3A+CRC+Press&btnG= (Loaʻa ʻia ma 24 Nov 2020).
ʻO Almås, A., Singh, B., Agriculture, TS-NJ o & 1995, undefined. Nā hopena o ka ʻoihana nickel Lūkini ma nā ʻano metala kaumaha ma nā lepo mahiʻai a me nā mauʻu ma Soer-Varanger, Norway.agris.fao.org.
ʻO Nielsen, GD et al. Pili ka omo ʻana a me ka paʻa ʻana o ka nickel i ka wai inu i ka ʻai ʻana i ka meaʻai a me ka ʻike nickel.toxicology.application.Pharmacodynamics.154, 67–75 (1999).
Costa, M. & Klein, CB Nickel carcinogenesis, mutation, epigenetics a i ʻole ke koho ʻana. puni. Health Perspective.107, 2 (1999).
ʻAjman, PC; Ajado, SK; Borůvka, L.; Bini, JKM; Sarkody, VYO; Cobonye, NM; Ka nānā ʻana i nā ʻano o nā mea ʻawahia: he loiloi bibliometric. Environmental Geochemistry and Health. Springer Science & Business Media BV 2020. https://doi.org/10.1007/s10653-020-00742-9.
Minasny, B. & McBratney, AB Palapala ʻĀina Lepo Kikohoʻe: He Moʻolelo Pōkole a me kekahi mau Haʻawina. Geoderma 264, 301–311. https://doi.org/10.1016/j.geoderma.2015.07.017 (2016).
McBratney, AB, Mendonça Santos, ML & Minasny, B. Ma ka palapala ʻāina kikohoʻe.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%2C+376+pages.+&btnG= (Loaʻa ʻia ma 28 ʻApelila 2021).
Ka manawa hoʻouna: Iulai-22-2022
