Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Doctorat → Chine → 2016 → The Effect of Long-term Fertilization on Farmland Soil in Loess Plateau

Northwest A&F University (2016)

The Effect of Long-term Fertilization on Farmland Soil in Loess Plateau

丁少男

Titre : The Effect of Long-term Fertilization on Farmland Soil in Loess Plateau

Auteur : 丁少男

Grade : Doctoral Dissertation 2016

Université : Northwest A&F University

Résumé
The farmland in the loess hilly region is an important part of the northern farmland in China. In order to increase crop yield, the fertilizer input in many places is very large, and the effects of different fertilization treatments on soil are different. Determining the comprehensive effect of long-term fertilization on soil, which could provide suggestion for the farmland management and agricultural activities, soil nutrient index,labile organic carbon and microbial properties at 0-20 cm and 20-40 cm layers under different fertilization treatments(N, P, N+P, M, M+N, M+P, M+N+P and BL, CK) were determined at an alluvial terrace farmland in the hilly region of Loess Plateau. Through the characteristics of soil nutrient, soil organic carbon pool and effects of nutrient transformation and energy flow response to different fertilization treatment, and then using the correlation analysis, cluster analysis, principal component analysis method to compare different fertilization treatments, mainly get the following conclusions:1. After 16 years tillage, compare to CK, long-term fertilization with N, P and NP fertilizer had no significant effect on organic matter(OM), total nitrogen(TN) and alkaline hydrolysis nitrogen(AN), and the application of P fertilizer could significantly increase the total phosphorus(TP) and available phosphorus(AP) content in the soil. The treatment of organic fertilizer could significantly increase the content of OM, TN, AN and AK in soil, and the average increase was 30%, 31%, 27% and 85%, respectively. The nutrient distribution trend at the 20-40 cm layer is not as obvious as the 0-20 cm layer. Fertilization mainly affected soil surface layer.2. Of different fertilization treatments, MNP was the highest crop yield and the yield difference between organic fertilizer and chemical fertilizer increased with the fertilization age. Single application of N or NP fertilizer could significantly increase crop yield, but the effect of P fertilizer was not significant. The contents of N, P and K in different parts of soybean and maize were significantly different under different fertilization treatments. The N and P content in soybean seed is about 10 times of its pod. The contents of N and P in maize seed is far greater than the content in the stem, and K content is far less than the stem content and organic fertilizer will increase a lot K content in maize stem.3. The differences in soil particle composition is not significant under different fertilization treatment, the average percent content of surface and subsurface clay, silt, sand was 17.92%, 60.72%, 21.37%. The dominant size of soil particles and micro aggregates were 0.02 0.05 mm. The fractal dimension of soil particle size in different fertilization treatments had no significant difference, and the value of MNP treatment was the least. Soil particle size fractal dimension is the extremely significant positive correlate with clay(< 0.002mm) and fine silt(0.002 0.02mm), and is significantly negatively correlate with coarse silt(0.02 0.05mm) and fine sand(0.05 0.2 mm) ; soil degree of aggregation is significantly negative correlate with soil micro- aggregates of <0.05 mm size, and significant positive correlate with soil micro aggregates of >0.05 mm size.4. Compare to CK, manure fertilizer can improve the soil WSOC, WSTN content much more than chemical fertilizer and treatment with P fertilizer will significantly increase WSTP content at surface layer. The surface soil C : N value is about 10, and the lower layer is a 9 under long-term fertilization, there is no significant difference among different treatments with small variation ; While WSOC/WSTN is significant different with the big variation of 15.7-28.6, the coefficient of variation is more than 10%. The response of WSOC to fertilization was more sensitive than that of organic carbon. The soil available Cu, Zn, Mn, Fe content in CK were 0.90, 1.03, 8.59, 6.31 mg kg-1. The soil microelement contents with long term application of organic manure is significantly higher than pure chemical fertilizer, average increase in Zn, Mn and Fe was 34.3%, 31.5% and 40.4%. Long term N fertilizer could result in a slight loss of available Zn and Mn, but no significant difference was achieved. The soil in the loess hilly region is not short of available Cu, and the available Zn, Mn and Fe are all at the edge of deficit, and the application of organic manure can improve the poor condition of microelements in soil. WSOM is significantly correlate with available Mn, Zn and Fe, while available Cu is just significantly correlate with WSOC.5. F1, F2, F4 and F3 accounted for 47%, 27%, 18%, and 8% of total organic carbon, respectively. The labile fractions(F1 and F2) showed advantages in quantity over passive fractions(F3+F4) in this region. F1 was a more sensitive index than the other C fractions through SI analysis, and could reflect the labile C fractions changes. Long-term fertilization with inorganic fertilizer could increase F4 fraction and F4 tend to be stabilized in deeper soil layers. The manure fertilizer not only increase labile fractions(F1) in a short time, but also could increase passive fraction(F4) in a long term.6. Urease, invertase and alkaline phosphatase enzyme activity at surface layer is higher than the subsurface layer except catalase ; Inorganic fertilizer did not significantly affect soil enzyme activity while organic fertilizer significantly do, fertilization differences affects soil enzyme activity. The response of enzyme activity per unit carbon to different fertilization was not consistent with the pattern of the traditional enzyme activity. The higher the OM content was, the lower the enzyme activity per unit carbon was. There was a high significant correlation between urease, invertase, alkaline phosphatase activity and labile fractions of OC, and the labile component of OC significantly affected the enzyme activity. Single application of N fertilizer could increase the content of G- and total PLFA, and single application of P fertilizer could increase the content of G-, actinomycetes and total PLFA, the effect of NP treatment on microbial respiration and community structure was not obvious. The application with organic fertilizer had significant effects on the contents of G+, G- bacteria, actinomycetes and total PLFA. M treatment affected the total PLFA content and the induced respiration, while organic and inorganic fertilizer could significantly change the microbial community structure and soil respiration intensity.7. Through principal component analysis with a total of 24 soil indicators, 4 principal components can be used to characterize, the cumulative variance is 90%. The main factors of the 4 kinds of main components are carbon-nitrogen factor, nutrient factor, phosphorus factor and soil structure factor, and the variance contribution rate is 60.12%, 17.15%, 6.48%, 6.29%, respectively. The results of principal component analysis showed that phosphorus is a limiting factor to the quality of farmland in loess area. MP treatment in the field of farmland, which is the highest quality of soil comprehensively. Application of chemical fertilizer alone will cause loss of soil fertility, is not conducive to the sustainable development of agriculture, the potential of organic fertilizer for the improvement of farmland is very great.

Mots clés : loess hilly region; long term fertilization; microelements; soil enzyme activity; oxidizable organic carbon fraction;

Présentation (CNKI)

Page publiée le 29 janvier 2017, mise à jour le 11 septembre 2017