中国农业科学 ›› 2017, Vol. 50 ›› Issue (24): 4714-4724.doi: 10.3864/j.issn.0578-1752.2017.24.005

• 耕作栽培·生理生化·农业信息技术 • 上一篇    下一篇

基于SPAD值的水稻施氮叶值模型构建及应用效果

李杰1,2,冯跃华1,牟桂婷1,许桂玲1,罗强鑫1,罗康杰1,黄世凤1,石欣1,管正策1,叶勇1,黄佑岗1

 
  

  1. 1贵州大学农学院,贵阳 550025;2黔东南民族职业技术学院,贵州凯里 556000
  • 收稿日期:2017-04-20 出版日期:2017-12-16 发布日期:2017-12-16
  • 通讯作者: 冯跃华,E-mail:fengyuehua2006@126.com
  • 作者简介:李杰,E-mail:guizhoutianxin@163.com
  • 基金资助:
    国家自然科学基金(31360311)、国家公益性行业(农业)科研专项经费(201503118-03)、贵州省农业科技攻关项目(黔科合NY [2013]3005)、贵州省作物学省级重点学科建设计划(黔学位合字ZDXK[2014]8)、贵州省普通高等学校粮油作物遗传改良与生理生态特色重点实验室项目(黔教合KY字[2015]333)

Construction and Application Effect of the Leaf Value Model Based on SPAD Value in Rice

LI Jie1,2, FENG YueHua1, MOU GuiTing1, XU GuiLing1, LUO QiangXin1, LUO KangJie1, HUANG ShiFeng1, SHI Xin1, GUAN ZhengCe1, YE Yong1, HUANG YouGang1   

  1. 1College of Agronomy, Guizhou University, Guiyang 550025; 2Qiandongnan Vocational and Technical College for Nationalities, Kaili 556000,Guizhou
  • Received:2017-04-20 Online:2017-12-16 Published:2017-12-16

摘要: 【目的】研究分析不同地力条件、施氮量、SPAD值衍生指标、产量之间的关系,实现简便、快速、无损地推荐水稻施氮量,构建基于SPAD值的水稻施氮叶值模型。【方法】2015年和2016年以杂交籼稻Q优6号为试验材料,设4种施氮水平(0、75、150、225 kg·hm-2),探讨产量、SPAD值衍生指标与田块表观供氮量之间的关系,并对初步构建的叶值模型进行变量施氮应用效果研究。【结果】产量与抽穗期田块表观供氮量之间具有极显著的曲线关系,两年拟合度R2分别为0.5523,0.7148。在其拟合关系下2年度最高产量分别为9 264.93 kg·hm-2、11 167.97 kg·hm-2,相差1 903.14 kg·hm-2,2016年产量相比2015年增加20.54%;达到最高产量的表观总吸氮量较为接近,分别为575.27 kg·hm-2,546.71 kg·hm-2,仅相差28.56 kg·hm-2,2016年表观总吸氮量相比2015年减少4.96%。不同年度的拔节期和抽穗期,SPAD值衍生指标中SPADL3(顶3叶SPAD值)、SPADL4(顶4叶SPAD值)、SPADmean(顶部4张叶片的平均SPAD值)、SPADL3×L4/mean(顶3叶SPAD值×顶4叶SPAD值/顶部4张叶片的平均SPAD值)与田块表观供氮量之间具有显著或极显著的线性关系。单张叶片中,SPADL3与拔节期田块表观供氮量,SPADL1与抽穗期田块表观供氮量线性拟合的系数在年份间变化均较小,分别为0.0156,0.0154;0.0172,0.0173。年份间,2016年SPADL3×L4/mean与田块表观供氮量线性拟合的系数和b值比2015年的拔节期依次增加了-28.70%,17.41%;抽穗期依次增加了-15.34%,56.11%。叶值模型施氮总量为表观总吸氮量与土壤表观供氮量之差,通过SPAD值衍生指标可以估测土壤表观供氮量,且抽穗期时SPAD值衍生指标与田块表观供氮量的线性拟合度较拔节期时的高。拔节期时,SPADL4与SPADL3×L4/mean,SPADL3与SPADmean之间估测推荐的施氮总量较为接近,且SPADL4、SPADL3×L4/mean估测的施氮量高出SPADL3、SPADmean50%左右。基于叶值模型的变量施氮效果表明,变量区产量高出对照区产量820.68 kg·hm-2,变量区的氮素偏生产力、农学利用率和贡献率均明显高于对照区,分别高出13.74%,103.45%,104.12%。确定叶值模型的一般表达式为:Nw=Nz-[(Ys-b)/k-Ng],式中Nw表示施氮总量(kg·hm-2),Nz表示水稻品种表观总吸氮量(kg·hm-2),Ys表示叶片SPAD值衍生指标,Ng表示追肥之前已经施入的氮量(kg·hm-2),k、b是田块表观供氮量(Nx)与叶片SPAD值衍生指标线性关系中的斜率和截距,而田块表观供氮量等于土壤表观供氮量(Nt)与人工已施氮量(Ng)之和【结论】应用叶值模型的变量施氮减少了产量差,提高了产量以及氮素农学利用率、偏生产力和贡献率。

关键词: 水稻, 产量, SPAD值衍生指标, 田块表观供氮量, 叶值模型, 变量施氮

Abstract: 【Objective】The relationship of different plots, nitrogen application amount, SPAD value derivative index and yield was studied to construct linear model of nitrogen rate based on SPAD value in rice, so as to easily, quickly, perfectly recommend nitrogen rate.【Method】The experiment was conducted by using Qyou6 as the materials with four nitrogen levels (0, 75, 150, 225 kg?hm-2) in 2015 and 2016. We compared the relationship between yield and apparent nitrogen supply of field, and between SPAD value derivative index and apparent nitrogen supply of field in this experiment. And the effect of variable nitrogen application based on leaf value model was also studied.【Result】The results showed that, curve relationship between yield and apparent nitrogen supply of field at heading stage was extremely significant positive correlation, and R2 of two years was 0.5523, 0.7148, respectively. In this fitting relationship, the higher yield of every years was 9 264.93 kg?hm-2, 11 167.97 kg?hm-2. The difference of maximum yield was 1 903.14 kg?hm-2 in two years, and yield in 2016 was 20.54% higher than that in 2015. But the apparent total nitrogen uptake of the highest yield in different years were close, which was 575.27 kg?hm-2, 546.71 kg?hm-2, respectively. And the difference of apparent total nitrogen uptake was only 28.56 kg?hm-2, and apparent total nitrogen uptake was -4.96% higher than in 2015. The linear relationships between SPADL3 (SPAD value of third leaf), SPADL4 (SPAD value of fourth leaf), SPADmean (average SPAD value of four leaves), SPADL3×L4/mean (product of SPAD value of third leaf and SPAD value of fourth leaf divided by average SPAD value of four leaves)and apparent nitrogen supply of field were significant or extremely significant under different growth period and year. On single leaf, the change of slope of linear relationship between SPADL3 and apparent nitrogen supply was smaller at jointing stage in years, and value of slope was 0.0156 in 2015, 0.0154 in 2016, respectively. At heading stage, the slope of linear relationship between SPADL1 and apparent nitrogen supply in 2015 and 2016 was 0.0172, 0.0173, respectively. At joining and heading stage of different years, the slope and origin of linear relationship between SPADL3×L4/meanand apparent nitrogen supply of 2016 increased by  -28.70%, 17.41%; -15.34%, 56.11% than that of 2015, respectively. Constructing a new rice nitrogen model based on SPAD value, which was called the leaf value model. The total nitrogen rate of leaf value model was difference between apparent total nitrogen uptake and apparent nitrogen supply of soil, and it could estimate apparent nitrogen supply of soil by measured value of SPAD value derivative index. And R2 of linear fitting between SPAD value derivative index and apparent nitrogen supply of field at heading stage was higher than that at jointing stage. At joining stage, the total nitrogen rate which was estimated by SPADL4 or SPADL3×L4/mean was close with that was estimated by SPADL3 or SPADmean, and the former was 50% higher than the later. The effect of variable nitrogen application based on leaf value model showed that yield of variable area was 820.68 kg?hm-2 higher than that of control area, and nitrogen partial productivity and nitrogen agronomic efficiency of variable area was 13.74%, 103.45% higher than that of control area, respectively. A general expression for leaf value model was Nw=Nz-[(Ys-b)/k-Ng], where Nw, Nz, Ys, Ng meant total nitrogen application rate (kg?hm-2), apparent total nitrogen uptake (kg?hm-2), SPAD value derivative index, base-tiller fertilizer amount (kg?hm-2), respectively, and k, b was slope and origin of linear relationship betweent apparent nitrogen supply of field (Nx) and SPAD value derivative index (Ys), and Nx was sum of nitrogen content of soil (kg?hm-2) and already manure amount. 【Conclusion】Variable nitrogen application based on leaf value model could reduce yield difference and increase yield, nitrogen agronomic efficiency, nitrogen partial productivity and nitrogen contribution rate.

Key words: rice, yield, SPAD value derivative index, apparent nitrogen supply of field, leaf value model, variable nitrogen