KMS XINJIANG INSTITUTE OF ECOLOGY AND GEOGRAPHY，CAS
| 预报式灌溉决策系统对绿洲棉田土壤与植物的氮磷养分的影响 | |
| Alternative Title | Effects of the predictive irrigation decision-making system on nitrogen and phosphorus nutrients in soil and plant of an oasis cotton field |
| 靳思佳 | |
| Subtype | 硕士 |
| Thesis Advisor | 齐志明 ; 李向义 |
| 2020-08-30 | |
| Degree Grantor | 中国科学院大学 |
| Place of Conferral | 北京 |
| Degree Discipline | 理学硕士 |
| Keyword | 预报式灌溉 RZQWM2 硝态氮 速效磷 情景模拟 predictive irrigation RZQWM2 nitrate nitrogen available phosphorus scenario simulation |
| Abstract | 我国肥料使用量大而效率低，不合理的水肥管理是农田养分损失的重要原因。在养分损失的几个途径中，淋溶损失的贡献率最大， 在干旱地区灌溉水是关键作用因子。 有研究学者新开发了一套基于根区水质模型（ Root Zone Water QualityModel 2， 简称 RZWQM2） 的新型预报式灌溉决策支持系统（ Decision SupportSystem for Irrigation Scheduling，简称 DSSIS），于 2016 年在新疆和田市策勒县的典型绿洲棉田建成测试。 初步测试和评估结果表明， DSSIS 能够在节约灌溉用水的基础上维持作物产量，提高水分利用效率。为了探究 DSSIS 对棉田土壤、植物氮磷养分的影响，在策勒国家试验站开展田间试验， 选取 3 种灌溉决策方式： (1)基于 RZWQM2 模型的灌溉决策支持系统（DSSIS），（2） 基于土壤水分传感器的墒情灌溉决策方式 (Soil moisture sensor， 简称 SMS)，和（3） 基于常规农民经验的经验灌溉决策方式(Experience， 简称 E)；设置 2 个灌溉量水平， 即 100%充分灌溉 ( 100% Full Irrigation， 简称 FI)与 75%亏缺灌溉(75% Deficit Irrigation， 简称DI)。实验研究了不同灌溉决策和灌水量对土壤氮磷分布、棉花生物量、氮磷吸收量和产量的影响。利用田间实测的土壤水分、土壤硝态氮、棉花生物量、产量等数据对 RZWQM2 模型的参数进行了优化，重点评估了 RZWQM2 在策勒绿洲对土壤氮素模拟的适应性。 最后， 进行不同施氮水平的情景模拟， 提出在 DSSIS灌溉决策方式下的最佳施氮水平。主要得出以下结论：（1） 棉花生长期 0～100 cm 土层的硝态氮分布量， DSSIS 和 SMS 显著高于E（P<0.05），棉花收获后 0～205 cm 土层中硝态氮残留量， DSSIS 和 SMS 显著高于 E（P<0.05）。 不同灌溉水平并没有导致硝态氮含量的显著差异，但 DI 水平可提高 0～100 cm 土层的硝态氮分布量， 降低 100～205 cm 土层的硝态氮残留量。 说明砂性土壤棉田中 DSSIS 和 SMS 有利于 205cm 以上土层硝态氮的保持，但仍存在淋失风险。 E 已经对土壤硝态氮进行了严重的淋洗，造成 205cm 以上土层硝态氮含量较低的现象。 速效磷主要分布在 0～30 cm 土层（ 19.43～50.68mg/kg），速效磷在土壤中的分布不受灌溉决策和灌水量的影响。（2）DSSIS 和 E 有助于棉花生物量的增长，但 E 会让棉花的营养生长过盛。DSSIS 和 E 都显著促进了棉花对氮磷养分的吸收（P<0.05）。 DSSIS 比 E 和 SMS的籽棉产量分别提高 33.7%和 12.2%，水分生产力分别提高 80.7%和 9.7%，氮肥料的偏生产力分别提高 12.2%和 25.2%，磷肥料的偏生产力分别提高 12.3%和25.6%。（3） RZWQM2 经过参数优化后，可以较好地模拟策勒绿洲棉田土壤水分、土壤硝态氮、棉花生物量、产量等指标。 RZWQM2 对 DSSIS 条件下不同施氮水平的模拟研究表明，当总氮投入量为 180 kg N/hm2，其中基施有机肥（羊粪） 9230kg/hm2，蕾期追施尿素 27 kg/hm2，初花期追施尿素 27 kg/hm2，盛花期追施尿素36 kg/hm2，花铃期追施尿素 72 kg/hm2，盛铃期追施尿素 72 kg/hm2 时棉花籽棉产量和肥料产量贡献率最高，同时土壤硝态氮的渗漏量相对较少。综上，与 SMS 和 E 相比，基于 RZWQM2 的 DSSIS 不仅具有节水增产的优势，还具有保持土壤养分，减少硝态氮淋失风险的优越性。在 DSSIS 条件下，棉田只需投入 180 kg N/hm2 的肥料，即可获得较高的产量，实现肥料的最大利用，实现氮的最小渗漏损失。因此， DSSIS 可被推荐作为策勒绿洲棉田水分管理的实施办法， RZWQM2 也可以作为一种可靠的研究工具对该区域的农田水肥管理、作物生产进行模拟研究。 |
| Other Abstract | China's fertilizer use is over-dosed and its efficiency is relatively low.Unreasonable water and fertilizer management is a major contributor to the loss offarmland nutrients. Among pathways of nutrient loss, leaching loss accounts for thelargest, and irrigation water is the key factor. Researchers have newly developed anew predictive irrigation decision support system (Decision Support System forIrrigation Scheduling, DSSIS) based on the RZWQM2 model. It was built in a typicaloasis cotton field in Cele County, Hetian City, Xinjiang. Preliminary test and evaluationresults show that DSSIS can maintain crop yield and improve water use efficiency onthe basis of saving irrigation water. In order to explore the effect of DSSIS on soil andplant nitrogen and phosphorus nutrients in cotton fields, a field experiment wasconducted at the Cele National Experimental Station. Three irrigation decision-makingmethods were set up: (1) DSSIS; (2) irrigation decision-making method based on soilmoisture sensor (SMS); and (3) irrigation decision-making method based on theexperience of conventional farmers (E). And 2 irrigation level were set up: 100% fullirrigation (FI) and 75% deficit irrigation (DI). The effects of different irrigation controlmethods and irrigation amount on soil nitrogen and phosphorus distribution, cottonbiomass, nitrogen and phosphorus absorption and yield were studied. The parametersof the RZWQM2 model were calibrated using the data of soil moisture, soil nitratenitrogen, cotton biomass, yield and other data measured in the field, focusing onevaluating the capability of RZWQM2 in simulating soil nitrogen cycle in the CeleOasis. Finally, a scenario simulation of different nitrogen application levels was carriedout, and the optimal nitrogen application level under the DSSIS irrigation decisionmaking method was proposed. The main conclusions are as follows:(1) The distribution of nitrate nitrogen in the 0～100 cm soil layer during thecotton growth period, DSSIS and SMS were significantly higher than E (P<0.05), andthe residual nitrate nitrogen in the 0～205 cm soil layer after cotton harvest, DSSIS and SMS were significantly higher than E (P<0.05). Different irrigation levels did not resultin significant differences in nitrate nitrogen content, while DI level could increase thenitrate nitrogen distribution in the 0～100 cm soil layer and reduce the nitrate nitrogenresidue in the 100～205 cm soil layer. This shows that DSSIS and SMS in sandy soilcotton fields are beneficial to the maintenance of nitrate nitrogen in the soil layer above205 cm, but there is still a risk of leaching. E has carried out a serious leaching of soilnitrate nitrogen, resulting in lower nitrate nitrogen content in the soil layer above 205cm.Available phosphorus is mainly distributed in the 0～30 cm soil layer (19.43～50.68mg/kg), and the distribution of available phosphorus in the soil is not affected byirrigation decision method or irrigation depth.(2) DSSIS and E contribute to the growth of cotton biomass, but E can makevegetative overgrowth. Both DSSIS and E significantly enhanced the absorption ofnitrogen by cotton (P<0.05). The yield of seed cotton under DSSIS increased by 33.7%and 12.2%, respectively, compared with E and SMS, the water productivity increasedby 80.7% and 9.7%, and the partial productivity of nitrogen fertilizer increased by 12.2%and 25.2%, respectively. The partial productivity of fertilizers increased by 12.3% and25.6%, respectively.(3) After parameter calibration, RZWQM2 simulated the soil moisture, soil nitratenitrogen, cotton biomass, yield and other variables of the cotton field in Cele Oasis inan acceptable manner. The RZWQM2 simulation with different nitrogen applicationlevels under DSSIS conditions showed that when the total nitrogen input was 180 kgN/ha, including the organic fertilizer (sheep dung) 9230 kg/ha at pre-planting,topdressing urea 27 kg/ha, urea topping 27 kg/ha at first bloom, topping urea 36 kg/haat full bloom, topping urea 72 kg/ha at full bloom, topping urea 72 kg/ha at full boll.The seed cotton yield and the contribution rate of fertilizer production was the highest,and the leakage of soil nitrate nitrogen was relatively low.To sum up, compared with SMS and E, the DSSIS based on RZWQM2 not onlyhas the advantages of saving water and increasing production, but also can retain soilnutrients and reduce the risk of nitrate nitrogen leaching. Under the DSSIS, the cottonfield only needs 180kg N/ha fertilizer to achieve a higher yield, maximum utilization of fertilizer, and lower nitrogen leaching loss. Therefore, DSSIS can be recommendedas an irrigation control method for water management in the cotton field in the CeleOasis, and RZWQM2 can also be used as a reliable tool for water and fertilizermanagement and crop production in this area. |
| Subject Area | 生态学 |
| Language | 中文 |
| Document Type | 学位论文 |
| Identifier | http://ir.xjlas.org/handle/365004/15427 |
| Collection | 中国科学院新疆生态与地理研究所 研究系统 |
| Affiliation | 中国科学院新疆生态与地理研究所 |
| First Author Affilication | 中国科学院新疆生态与地理研究所 |
| Recommended Citation GB/T 7714 | 靳思佳. 预报式灌溉决策系统对绿洲棉田土壤与植物的氮磷养分的影响[D]. 北京. 中国科学院大学,2020. |
| Files in This Item: | There are no files associated with this item. | |||||
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Edit Comment