-
农作物高产稳产和施肥模式的环境友好是农业可持续发展的关键,但其稳定性受到诸多因素的影响。在现实生产中,土壤养分管理存在偏施化肥、有机肥施用不足和养分比例失调等不合理现象,导致作物产量稳定性下降。因此,探讨在田块尺度上长期不同施肥模式对作物产量稳定性影响以及如何维持作物持续高产稳产具有重要意义。长期定位试验同时具备时间上的长期性和气候上的代表性,是评价农田生态系统可持续性的有效技术手段。Calderini等[1]和Verón等[2]通过回归模型分析了阿根廷等21个国家的产量趋势与稳定性;Ladha等[3]对33个长期肥料定位试验的总结表明,有85%试验点施用氮磷钾化肥的水稻和小麦产量保持稳定,6%试验点呈显著下降趋势;Manna等[4]研究表明,有机肥与化肥配施和平衡施用氮磷钾化肥可有效增加大豆–小麦轮作系统作物产量稳定性。我国北方不同区域和主要土壤类型长期施肥对小麦和玉米产量稳定性和可持续性的影响[5-11],及南方黄壤、红壤和水稻土等主要土壤类型长期施肥对水稻、玉米、甘薯产量稳定性和可持续性影响[12-18]均开展了许多深入系统研究,根据产量趋势指数、产量变异系数和产量可持续性指数等指标[19],研究结果大都揭示了长期氮磷钾推荐施肥或氮磷钾化肥配施有机肥有利于维持和提高土壤肥力,作物产量年际间变化幅度较小,有利于作物高产稳产,是维持农田生态系统系统可持续性的最优施肥模式。但在已有研究文献中,针对赤红壤旱地长期不同施肥模式对花生–甘薯轮作制产量效应的研究还鲜见报道,趋势产量指数、产量变异系数和产量可持续性指数对评价花生和甘薯产量稳定性的适宜性尚待深入探讨。
花生和甘薯是福建省除了水稻以外种植面积最大的两种大田作物,常年种植面积分别在10万hm2和20万hm2左右。种植区域主要分布于闽东南沿海和闽西的丘陵耕地,花生–甘薯轮作制是该区域旱耕地的主要耕作制度,在区域农业生产中具有重要地位。为此,本研究根据闽东南赤红壤旱地花生–甘薯轮作制连续15年和13年不同施肥模式的两个定位试验的产量演变特征,探讨不同施肥模式对产量稳定性变化的影响及其定量评价方法,旨在为赤红壤旱地花生-甘薯轮作制实现高产稳产提供最佳养分管理技术依据。
-
定位试验设置在福建省莆田市秀屿区笏石镇东华村,北纬25°19′4.31″,东经119°05′ 9.10″。地处北回归线北侧边缘,东濒海洋,属亚热带海洋性季风气候,日照充足,平均年太阳辐射量110.41千卡/cm2,年日照时数平均1996 h,年均日照率45%。气候温和湿润,年均气温16℃~21℃之间,年降雨量90~1300 mm,农业生产上适宜一年二熟制或三熟制作物栽培。两个定位试验田的供试土壤均为赤红壤发育的旱地土壤,母岩为花岗岩,土壤类型均为灰赤沙土,土层深厚,土壤质地为沙壤土。
-
试验从2005年3月开始至2020年已有15年。采用常规方法[20]测定供试土壤主要理化性状为pH 5.90、有机质17.8 g/kg、全氮1.13 g/kg、全磷0.60 g/kg、全钾13.7 g/kg、缓效钾345 mg/kg、碱解氮128 mg/kg、Olsen-P 46.3 mg/kg、速效钾 86.1 mg/kg;代换性钙和镁含量分别为950 mg/kg和120 mg/kg,有效硫为10.4 mg/kg,有效硼和有效锌含量分别为0.40 mg/kg和3.20 mg/kg。试验设6个处理 (表1),其中,习惯施肥是根据试验农户的常年施肥习惯确定,推荐施肥则是根据花生、甘薯参考文献[21]确定。小区间用水泥筑永久性田埂作分离,其中田埂埋入地下30 cm,露出田面20 cm,各小区均有进出水口连接排灌水沟。试验设4次重复,随机区组排列,小区面积16.2 m2。
表 1 花生–甘薯轮作制长期定位试验处理和施肥量 (kg/hm2)
Table 1. Treatments and fertilizer input (kg/hm2) in the two long-term peanut–sweet potato rotation experiments
长期定位试验
Long-term experiment处理编号
Treatment number处理
Treatment花生 Peanut 甘薯 Sweet potato N P2O5 K2O 有机肥 Manure N P2O5 K2O 有机肥 Manure 化肥
Chemical fertilizer1 CK 0 0 0 0 0 0 2 FP 90 45 75 225 45 150 3 RF 75 60 90 180 45 225 4 RF-N 0 60 90 0 45 225 5 RF-P 75 0 90 180 0 225 6 RF-K 75 60 0 180 45 0 化肥和有机肥配施
Chemical and organic fertilizer1 CK 0 0 0 0 0 0 0 0 2 RF 75 60 90 0 180 60 225 0 3 RF + CM 50 40 60 1995 120 40 150 4140 4 RF + PM 50 50 71 6585 120 35 177 15795 5 RF + S 50 57 38 2745 120 51 101 6600 注(Note):FP—农民习惯施肥 Farmer's practice; RF—推荐施肥 Recommended fertilization; RF-N—推荐施肥减氮 Without N in the recommended fertilization; RF-P—推荐施肥减磷 Without P in the recommended fertilization; RF-K—推荐施肥减钾 Without K in the recommended fertilization; CM—商品有机肥 Commercial organic fertilizer (N + P2O5 + K2O ≥ 5%, N 1.7%); RF + CM、RF + PM 和 RF + S 分别表示推荐施氮量的 1/3 由商品有机肥、猪粪和稻草替代 RF + CM, RF + PM and RF + S represent the replacement of one-third of recommended N input with commercial organic manure, pig manure, and rice straw, respectively. 供试氮肥为尿素 (N 46%),磷肥为过磷酸钙 (P2O5 12%),钾肥为氯化钾 (K2O 60%),试验地不施有机肥。根据基础土壤有效养分测定结果,在适当年份补施必要的中微量元素肥料。花生施肥方法:基施为50%的氮肥和钾肥、100%的磷肥;50%的氮钾肥肥在苗期追施。甘薯的施肥方法为:基施50%的氮肥和钾肥、100%的磷肥;40%的氮肥在第一次追肥 (苗期) 施用;10%的氮和50%钾肥作第二次追肥 (夹边肥) 施用。每季作物收获时各小区单收单称,分别记录农产品鲜重和晒干重。
-
试验从2007年3月开始至2020年共有13年,基础土样主要理化性状:pH 5.73、有机质17.1 g/kg、土壤全氮1.72 g/kg、全磷0.67 g/kg、全钾12.5 g/kg、缓效钾192 mg/kg,碱解氮151 mg/kg、Olsen-P 15.6 mg/kg、速效钾84.5 mg/kg;代换性钙和镁含量分别为1359 mg/kg和127 mg/kg,有效硫为16.3 mg/kg,有效硼和有效锌含量分别为1.00 mg/kg和2.10 mg/kg。试验包括5个处理 (表1),其中,处理2为供试花生和甘薯的化肥推荐施肥量[21];处理3中的供试有机肥由市场购入,其N + P2O5 + K2O ≥ 5%;处理4和处理5的供试有机肥为当地农家腐熟猪粪和稻草,猪粪养分含量为N 0.38%、P 0.16%和K 0.30% (干基),稻草养分为N 0.91%、P 0.13%和K 1.89% (干基)。处理3~5与处理2为等量氮磷钾设计,其中化学氮肥的1/3由有机肥替代。供试化肥同1.2.1。每个处理重复3次,随机区组排列,小区面积19.6 m2。田埂制作、供试花生和甘薯的施肥管理、收获和计产方法与化肥定位试验相同。
-
产量稳定性指标采用植物生态学中常用的变异系数 (coefficient of variation,CV) 表示,衡量年际间产量的变异程度,CV值越大则说明产量稳定性越低。计算公式为[22]:
$CV = \frac{\sigma }{{\overline Y }} \times {\rm{100{\text \%} }}$ 产量可持续性采用可持续产量指数来表征,指数越高则说明该系统的可持续越好。计算公式为[23]:
$SYI = \frac{{\overline Y - \sigma }}{{Y_{\rm{max}}}}$ 式 (1) 和 (2) 中,σ为标准差 (kg/hm2),
$\overline Y $ 为定位试验某一处理的产量均值 (kg/hm2),Ymax为该定位试验处理的最高产量 (kg/hm2)。 -
现有文献资料中,产量趋势指数是根据年际产量与试验年限散点图拟合的趋势线,并依据斜率[年变化值,kg/ (hm2·yr)]大小来评价产量随着时间变化的状况[24]。但考虑到作物产量水平受到诸多不可控因子的影响,导致年际间实际产量水平出现较大的波动,直接用年际产量及其年份进行拟合难以得到具有统计学意义的回归模型。因此,将年际作物产量看成是灰色量,应用灰色系统理论的一次累加生产原理[25],建立一元灰色线性趋势模型。设Y是一次累加生成的作物产量 (kg/hm2),t为对应定位试验时间 (年),建立一元线性回归模型:
$Y = a + bt$ 根据微积分原理,式 (3) 的导数等于b,即b为年际可得趋势产量均值。通过计算式 (3) 回归方程b值的置信区间,即可对不同施肥模式的作物趋势产量做出定量评价。
本文采用MATLAB R2015b软件免费试用版进行数据处理。试验数据汇总、二因素方差分析和一元线性回归分析应用该软件的统计工具箱进行相关计算和统计分析[26];产量变异系数、产量可持续性指数和文中图形则采用MATLAB语言编程计算和绘制。显著性选用P < 0.05 水平。
-
两个定位试验的花生和甘薯历年实收产量结果表明,不施肥处理在试验开始2~3年内产量迅速下降,过后则逐渐稳定在最低水平上,其它所有施肥处理的产量水平均高于该处理。在化肥施肥模式定位试验中,推荐施肥无论是花生还是甘薯的历年产量始终高于其它施肥模式,其次是习惯施肥。在化肥配施有机肥定位试验中,花生历年产量均是化肥配施猪粪处理高于其它处理;在2015年以前,化肥配施稻草的甘薯产量高于其它处理,但之后化肥配施猪粪处理的产量有高于配施稻草处理的趋势。表2结果显示,各处理第1~5年的平均产量明显高于第6~10年的平均产量,但化肥试验最近5年和化肥与有机肥配施试验最近3年的平均产量有上升的趋势。
表 2 定位试验不同年限轮作体系中各施肥处理花生和甘薯的产量
Table 2. Yield of peanut and sweet potato in two different fertilization experiments and different experimental years
长期试验
Experiment处理
Treatment花生 Peanut (kg/hm2, dry base) 甘薯 Sweet potato (kg/hm2, fresh base) 1~5年
1–5 years6~10年
6–10 years11~15年
11–15 years*平均
Mean1~5年
1–5 years6~10年
6–10 years11~15年
11–15 years*平均
Mean化肥**
Chemical fertilizer applicationCK 1503 ± 764 d 776 ± 351 d 984 ± 538 e 1101 ± 603 e 5649 ± 951 e 6858 ± 2570 e 7033 ± 1934 e 6092 ± 2161 e FP 3674 ± 324 ab 2974 ± 659 ab 3544 ± 277 b 3411 ± 536 b 15760 ± 1400 b 17262 ± 5070 b 16063 ± 3319 b 16178 ± 3356 b RF 4229 ± 305 a 3367 ± 674 a 4147 ± 399 a 3936 ± 608 a 18640 ± 1235 a 20073 ± 6198 a 18976 ± 3155 a 19070 ± 3796 a RF-N 2659 ± 623 c 1747 ± 767 c 2080 ± 481 d 2183 ± 695 d 10210 ± 1773 d 10503 ± 2639 d 10085 ± 1518 d 10020 ± 2101 d RF-P 3236 ± 492 bc 2432 ± 649 b 2791 ± 373 cd 2829 ± 580 c 13833 ± 2177 c 14091 ± 4193 c 14529 ± 2198 c 13667 ± 2889 c RF-K 3251 ± 392 b 2336 ± 601 bc 2870 ± 449 c 2844 ± 594 c 12989 ± 3017 cd 13196 ± 3897 c 12569 ± 2218 c 12798 ± 2894 c 化肥和有机肥配施***
Chemical and organic fertilizerCK 2151 ± 580 b 986 ± 375 c 1131 ± 200 c 1449 ± 926 c 5871 ± 1778 d 8214 ± 1826 b 7928 ± 3472 c 6701 ± 2139 c RF 3558 ± 523 a 3000 ± 584 b 3033 ± 1056 b 3223 ± 532 b 11490 ± 2409 c 16905 ± 4107 a 14640 ± 2077 b 14250 ± 3831 b RF + CM 3765 ± 536 a 3160 ± 624 ab 3217 ± 649 ab 3393 ± 567 ab 12457 ± 2921 bc 18299 ± 3804 a 15476 ± 1893 ab 15355 ± 3950 b RF + PM 4027 ± 478 a 3308 ± 536 a 3478 ± 675 a 3613 ± 540 a 13116 ± 2837 b 19337 ± 3722 a 19643 ± 3647 a 16681 ± 4563 ab RF + S 3646 ± 483 a 2927 ± 577 b 3226 ± 638 ab 3269 ± 546 b 14624 ± 2518 a 19464 ± 5762 a 17829 ± 1447 ab 17085 ± 4370 a 注(Note):*—化肥有机肥配合试验为第三阶段时间为 11~13 年; **—化肥长期地位试验花生产量方差分析的处理间 F 值和年际间 F 值分别为 213.9 和 26.8; 甘薯产量的处理间 F 值和年际间 F 值则分别为 161.8 和 21.7; ***—化肥有机肥配合长期定位试验的花生产量方差分析的处理间和年际间的 F 值分别为 132.8 和 23.4; 甘薯产量的处理间和年际间的 F 值则分别为 60.6 和 15.4; 同列数据后不同字母表示同一试验处理间差异达 5% 显著水平。*—The period for the chemical and organic fertilizer combination experiment was 11 to 13 years; **—In the chemical fertilization experiment, the inter-treatment and inter-annual F values for peanut yield were 213.9 and 26.8, and those of sweet potato yield were 161.8 and were 21.7, respectively; ***—In the chemical and organic fertilizer combination experiment, the inter-treatment and inter-annual F values for peanut yield were 132.8 and 23.4, and those of sweet potato were 60.6 and 15.4, respectively. Values followed by different small letters in each column indicate significant differences among treatments in the same experiment at 0.05 level. FP—农民习惯施肥 Farmer's practice; RF—推荐施肥 Recommended fertilization; RF-N—推荐施肥减氮 Without N in the recommended fertilization; RF-P—推荐施肥减磷 Without P in the recommended fertilization; RF-K—推荐施肥减钾 Without K in the recommended fertilization; CM—商品有机肥 Commercial organic fertilizer (N + P2O5 + K2O ≥ 5%, N 1.7%); RF + CM、RF + PM 和 RF + S分别表示推荐施氮量的 1/3 由商品有机肥、猪粪和稻草替代 RF + CM, RF + PM and RF + S represent the replacement of one-third of recommended N input with commercial organic manure, pig manure, and rice straw, respectively. 处理和年际的二因素方差分析和LSD法多重比较结果 (表2) 表明,花生和甘薯化肥推荐施肥模式的产量均值显著高于其它施肥模式,其比习惯施肥平均分别增产13.3%和15.2%,比不施肥平均分别增产72.0%和68.1%。花生施用氮、磷、钾肥平均分别增产44.6%、28.1%和27.7%,甘薯则平均分别增产47.5%、28.3%和32.9%。花生和甘薯施氮肥的增产效果显著高于磷肥和钾肥;花生的磷、钾肥增产效果差异不显著,甘薯的钾肥增产效果则显著高于磷肥。在化肥配施有机肥定位试验中,化肥配施猪粪处理的花生产量显著高于化肥推荐施肥和化肥配施稻草处理,但与化肥配施商品有机肥处理的产量差异不显著;与化肥推荐施肥处理相比,化肥配施猪粪或商品有机肥或稻草处理的花生平均增产率分别为12.1%、5.3%和1.4%。化肥配施稻草处理的甘薯产量均值显著高于化肥推荐施肥和化肥配施商品有机肥处理,但与化肥配施猪粪处理的产量差异不显著;与化肥推荐施肥处理相比,化肥配施稻草或猪粪或商品有机肥处理的花生平均增产率分别为19.9%、17.1%和7.8%。
因此,连续15年和13年的两个定位试验的产量均值表明,化肥推荐施肥处理的花生和甘薯产量较高,花生推荐施肥配施猪粪、甘薯推荐施肥施稻草或猪粪可进一步提高产量水平。
-
根据表1中两个定位试验历年各处理花生和甘薯实收产量计算各重复小区的产量可持续性指数 (SYI) 结果 (表3),处理和年际二因素方差分析表明,在化肥定位试验中,花生和甘薯均以RF的SYI最大,分别为0.729 ± 0.019和0.501 ± 0.028,均显著高于FP处理;而FP处理的SYI显著高于RF-N、RF-P、RF-K处理;RF-N处理的SYI显著低于RF-P和RF-K,后两处理间的SYI差异不显著。在化肥有机肥配施定位试验中,RF + PM处理的花生产量SYI为0.689 ± 0.013,显著高于CK和RF,与RF + CM差异不显著;RF + S处理的甘薯SYI为0.514 ± 0.029,显著高于CK、RF和RF + CM,但与RF + PM差异不显著。
表 3 长期不同施肥处理对轮作体系中花生和甘薯产量可持续性指数的影响
Table 3. Effects of different fertilization patterns on sustainable yield index (SYI) of peanut and sweet potato
化肥长期定位试验 Chemical fertilizer experiment 化肥有机肥配施长期定位试验 Chemical and organic fertilizer experiment 处理 Treatment 花生 Peanut 甘薯 Sweet potato 处理 Treatment 花生 Peanut 甘薯 Sweet potato CK 0.101 ± 0.006 e 0.128 ± 0.020 e CK 0.117 ± 0.004 c 0.182 ± 0.003 d FP 0.631 ± 0.014 b 0.421 ± 0.016 b RF 0.605 ± 0.012 b 0.421 ± 0.011 c RF 0.729 ± 0.019 a 0.501 ± 0.028 a RF+CM 0.630 ± 0.023 ab 0.462 ± 0.011 b RF-N 0.324 ± 0.032 d 0.262 ± 0.008 d RF+PM 0.689 ± 0.013 a 0.492 ± 0.023 ab RF-P 0.491 ± 0.021 c 0.355 ± 0.022 c RF+S 0.609 ± 0.033 b 0.514 ± 0.029 a RF-K 0.492 ± 0.020 c 0.328 ± 0.013 c F值 F value 1380.4** 288.5** F值 F value 359.7** 216.5** 注(Note):FP—农民习惯施肥 Farmer's practice; RF—推荐施肥 Recommended fertilization; RF-N—推荐施肥减氮 Without N in the recommended fertilization; RF-P—推荐施肥减磷 Without P in the recommended fertilization; RF-K—推荐施肥减钾 Without K in the recommended fertilization; CM—商品有机肥 Commercial organic fertilizer (N + P2O5 + K2O ≥ 5%, N 1.7%); RF + CM、RF + PM和RF + S 分别表示推荐施氮量的 1/3 由商品有机肥、猪粪和稻草替代 RF + CM, RF + PM and RF + S represent the replacement of one-third of recommended N input with commercial organic manure, pig manure, and rice straw, respectively. 同列数据后不同字母表示同一试验处理间差异达 5% 显著水平 Values followed by different small letters in each column indicate significant differences among treatments at 0.05 level. 因此,RF + CM和RF + PM处理的花生产量可持续性指数没有显著差异,RF + PM和RF + S处理的甘薯产量可持续性指数之间没有显著差异。
-
实现高产稳产是农业生产的关键目标之一,但由于作物产量水平受到众多不可控因素的影响,使历年实际产量水平出现较大幅度的随机波动,常规建模方法难以给出有统计学意义的定量评价指标。然而,根据灰色系统建模理论[25],若对同一个施肥模式的历年实收产量进行一次累加生成,在二维坐标上以对应试验时间 (年) 绘制散点图,例如以试验1的甘薯产量 (图1) 为例,表明二者间存在典型的线性关系。为此,应用式 (3) 模型进行一元线性回归分析 (表4)。
图 1 不同施肥模式甘薯累加产量的变化
Figure 1. Relationship between cumulative yields of sweet potato and experimental years
表 4 轮作系统中长期不同施肥处理的花生和甘薯产量趋势灰色线性模型
Table 4. Linear grey model of peanut and sweet potato yield trends as affected by fertilization patterns
定位试验
Experiment处理
Treatments花生 Peanut 甘薯 Sweet potato Y = a + bt F值
F valueR2 b的95%置信区间
95% confidence interval of bY = a + bt F值
F valueR2 b的95%置信区间
95% confidence interval of b化肥
Chemical fertilizer experimentCK Y = 2598 + 889t 1055.5** 0.988 829~948 Y = –2844 + 6472t 1263.7** 0.990 6078~6865 FP Y = 1095 + 3276t 6928.3** 0.998 3191~3361 Y = –2322 + 16477t 3258.2** 0.996 15853~17100 RF Y = 1096 + 3780t 5363.0** 0.998 3668~3891 Y = –3980 + 19408t 4234.4** 0.997 18763~20052 RF-N Y = 2715 + 1959t 2578.9** 0.995 1876~2042 Y = 1403 + 10072t 3567.3** 0.996 9708~10436 RF-P Y = 2162 + 2650t 5583.1** 0.998 2574~2727 Y = 1270 + 13678t 3089.7** 0.996 13147~14210 RF-K Y = 1993 + 2652t 3663.2** 0.996 2557~2747 Y = 2989 + 12637t 2934.5** 0.996 12133~13141 化肥有机肥配施
Chemical and organic fertilizer experimentCK Y = 4208 + 1159t 606.2** 0.982 1056~1263 Y = –1209 + 6818t 1153.8** 0.991 6376~7260 RF Y = 1990 + 3093t 8566.9** 0.999 3028~3158 Y = –7392 + 14807t 1728.1** 0.994 14023~15591 RF + CM Y = 1951 + 3276t 9789.1** 0.999 3203~3349 Y = –7286 + 15901t 1683.4** 0.994 15048~16754 RF + PM Y = 1905 + 3492t 9474.6** 0.999 3413~3571 Y = –10408 + 17177t 1252.5** 0.991 16109~18245 RF + S Y = 1581 + 3154t 10332.0** 0.999 3086~3222 Y = –7092 + 17567t 2231.6** 0.995 16749~18386 注(Note):一元回归方程中,Y 表示产量,t表示试验年限,b表示斜率,即为年际趋势产量。In the unitary regression equation, Y is the yield, t is the experimental years, b is the slope, indicating the inter-annual yield trend. FP—农民习惯施肥 Farmer's practice; RF—推荐施肥 Recommended fertilization; RF-N—推荐施肥减氮 Without N in the recommended fertilization; RF-P—推荐施肥减磷 Without P in the recommended fertilization; RF-K—推荐施肥减钾 Without K in the recommended fertilization; CM—商品有机肥 Commercial organic fertilizer (N + P2O5 + K2O ≥ 5%, N 1.7%); RF + CM、RF + PM 和 RF + S 分别表示推荐施氮量的 1/3 由商品有机肥、猪粪和稻草替代 RF + CM, RF + PM and RF + S represent the replacement of one-third of recommended N input with commercial organic manure, pig manure, and rice straw, respectively.
**—P < 0.01.表4的回归分析表明,两个定位试验不同施肥模式的22个花生荚果、甘薯薯块累加产量的回归模型F值均达到极显著水平以上,且描述产量变异方差解释能力的拟合优度R2都高达98.2%以上,显示均具有极佳的拟合效果。供试作物年际趋势产量均值,即一元回归模型b值的参数估计结果表明,在化肥不同施肥模式定位试验中,花生和甘薯趋势产量均值大小顺序均为推荐施肥 > 习惯施肥 > 不施磷肥和不施钾肥 > 不施氮肥 > 不施肥。b值的95%置信区间表明,花生化肥推荐施肥的置信区间为3668~3891 kg/hm2,习惯施肥则为3191~3361 kg/hm2;甘薯推荐施肥的置信区间为18763~20052 kg/hm2,习惯施肥则为15853~17100 kg/hm2,显示推荐施肥和习惯施肥的产量在95%置信区间内不重叠,揭示了化肥推荐施肥的趋势产量显著高于习惯施肥。
在化肥配施有机肥定位试验中,花生趋势产量均值大小顺序是化肥配施猪粪 > 化肥配施商品有机肥 > 化肥配施稻草 > 化肥推荐施肥 > 不施肥;甘薯趋势产量均值大小顺序是化肥配施稻草 > 化肥配施猪粪 > 化肥配施商品有机肥 > 化肥推荐施肥 > 不施肥。化肥配施猪粪处理的花生趋势产量的95%置信区间为3413~3571 kg/hm2,化肥配施商品有机肥处理的花生趋势产量则为3203~3349 kg/hm2,二者在该置信区间内不重叠,显示前者施肥处理的趋势产量显著高于后者。化肥配施猪粪、化肥配施稻草处理的甘薯趋势产量在95%置信区间里出现较大的重叠,而与化肥配施商品有机肥、化肥推荐施肥和不施肥等处理的趋势产量置信区间重叠很少或不重叠,揭示了化肥配施稻草或猪粪的甘薯趋势产量显著高于其它3种施肥处理。
-
国内外的研究表明[27-28],产量可持续性指数 (SYI) 是评价不同养分管理系统可持续性的一个重要指标。应用 (2) 式计算不同施肥处理长期定位试验的产量可持续性,综合比较2个定位试验的结果表明,长期不施肥处理的花生和甘薯产量可持续性指数最小,说明不施肥条件下产量容易大幅波动;化肥推荐施肥处理的花生和甘薯可持续获得较高的产量,花生推荐施肥配施猪粪处理、甘薯推荐施肥配施稻草处理可进一步持续提高产量水平,而且具有最大的SYI值 (表4)。杨生茂等[29]和李秀英等[30]在玉米和小麦上研究表明,氮磷钾化肥推荐施肥和化肥配施有机肥均使产量稳产性显著提高。本研究表明,在推荐施肥基础上,等氮磷钾养分投入下的有机无机肥配施花生和甘薯产量年际间波动明显减少,产量可持续性指数显著高于单施化肥,具有较好的稳产性,与前人的相关研究结果[31-35]一致。
从统计学的角度看,产量可持续性指数计算式的分母,Ymax取值应是计算产量均值和标准差的同一个产量系列的最大值,据此反映该产量数据系列的年际波动情况。按照该方法计算2个定位试验各施肥模式的SYI结果 (表5) 表明,在化肥不同施肥模式定位试验中,花生和甘薯的习惯施肥和推荐施肥的SYI指数均没有显著差异;在化肥配施有机肥定位试验中,不同配施模式间在花生和甘薯产量SYI同样均无显著差异。相关结果与表3不一致,其原因是SYI仅反映同一个施肥处理的历年产量数据波动情况,而与产量高低没有直接关系。
表 5 长期不同施肥处理下轮作体系中花生和甘薯产量的变异系数和可持续性指数 (SYI)
Table 5. The coefficient of variation (CV) and sustainable yield index of peanut and sweet potato under different treatments in two long-term experiments
定位试验
Experiment处理
Treatment花生 Peanut 甘薯 Sweet potato CV (%) SYI CV (%) SYI 化肥
Chemicalfertilizer experimentCK 55.3 ± 2.0 a 0.175 ± 0.014 d 36.9 ± 9.6 a 0.339 ± 0.114 b FP 15.7 ± 0.7 d 0.711 ± 0.027 a 22.1 ± 2.0 b 0.500 ± 0.030 a RF 15.6 ± 1.3 d 0.730 ± 0.026 a 21.4 ± 2.1 b 0.501 ± 0.028 a RF-N 32.5 ± 1.6 b 0.407 ± 0.029 c 21.8 ± 3.2 b 0.579 ± 0.056 a RF-P 21.1 ± 1.3 c 0.579 ± 0.042 b 22.4 ± 2.3 b 0.523 ± 0.052 a RF-K 21.3 ± 2.2 c 0.598 ± 0.050 b 23.4 ± 1.7 b 0.519 ± 0.033 a F值 F value 565.0** 199.4** 9.8** 9.8** 化肥有机肥配施
Chemical and organic fertilizer experimentCK 64.1 ± 1.0 a 0.138 ± 0.004 b 33.2 ± 1.2 a 0.442 ± 0.030 b FP 16.7 ± 0.7 b 0.637 ± 0.018 a 27.3 ± 2.2 b 0.489 ± 0.020 ab RF + CM 17.6 ± 2.1 b 0.631 ± 0.034 a 26.1 ± 0.6 b 0.515 ± 0.005 a RF + PM 15.4 ± 1.2 b 0.670 ± 0.027 a 27.6 ± 2.1 b 0.507 ± 0.031 a RF + S 17.3 ± 1.4 b 0.643 ± 0.037 a 26.0 ± 2.1 b 0.513 ± 0.026 a F值F value 641.6** 10.2** 213.2** 5.6* 注(Note):FP—农民习惯施肥 Farmer's practice; RF—推荐施肥 Recommended fertilization; RF-N—推荐施肥减氮 Without N in the recommended fertilization; RF-P—推荐施肥减磷 Without P in the recommended fertilization; RF-K—推荐施肥减钾 Without K in the recommended fertilization; CM—商品有机肥Commercial organic fertilizer (N + P2O5 + K2O ≥ 5%,N 1.7%); RF + CM、RF + PM和RF + S分别表示推荐施氮量的1/3由商品有机肥、猪粪和稻草替代 RF + CM, RF + PM and RF + S represent the replacement of one-third of recommended N input with commercial organic manure, pig manure, and rice straw, respectively. 同列数据后不同小写字母表示同一定位试验处理间该指标差异显著达 P < 0.05 水平 Values followed by different small letters in each column indicate significant differences among treatments in the same experiment at 0.05 level. 在定位试验产量稳定性和可持续性评价中,产量变异系数也是常用指标。在两个长期定位中化肥定位试验的花生产量外,产量变异系数指标仅能明确区分化肥定位试验中不同施肥处理的花生产量的稳定性,难以区分而该定位试验的甘薯产量和化肥有机肥配合试验中的花生和甘薯产量的稳定性 (表5),显示该指标只能粗略地反映产量系列的波动状况。
目前,许多文献巧妙地将Ymax改为同一个试验点各施肥处理得到的最高产量,较好地改善了SYI的应用价值,满足农田生态系统稳定性评价中的高产目标[36]。但是,式 (2) 模型终究是个经验式,在未来研究中Ymax取值的专业合理性还需要做更深入地阐述。
-
在长期施肥定位试验中,产量变异系数和产量可持续性指数反映了不同施肥模式的历年作物产量波动状态,但在农业生产中,作物产量水平及其长期趋势更受人们关注。人们经常根据同一施肥模式的历年实际产量水平与对应试验年限构建一元线性趋势线,以确定年际产量的递增或递减状况[19]。按照该方法构建不同施肥模式的花生和甘薯年际产量的一元线性趋势线 (表6),回归分析表明,除了化肥配施有机肥定位试验的花生3个处理外,其它19个回归方程均未达到统计显著水平;描述产量变异方差解释能力的拟合优度R2都很小,即使能通过显著性检验的3个回归方程,R2值也小于43%,结果揭示了这种趋势线用于描述不同施肥模式的长期产量趋势没有统计学意义。其原因,在于作物产量是个随机变量,且年际间出现较大波动幅度,常规回归分析方法难以适应。
表 6 一元线性回归对长期不同施肥处理花生和甘薯产量趋势的拟合精度
Table 6. Fitting accuracy of linear regression on the yield trendofpeanut and sweet potato under long-term fertilization
长期定位试验
Experiment处理
Treatment花生 Peanut 甘薯 Sweet potato F值 F value R2 F值 F value R2 化肥
Chemical fertilizer application experimentCK 1.9 0.127 0.0 0.000 FP 0.0 0.000 0.0 0.000 RF 0.1 0.004 0.1 0.007 RF-N 1.5 0.101 0.6 0.043 RF-P 1.3 0.091 0.2 0.013 RF-K 0.5 0.034 0.3 0.020 化肥有机肥配施
Chemical and organic fertilizer experimentCK 8.3* 0.429 0.3 0.003 FP 5.2* 0.320 0.8 0.064 RF + CM 4.9* 0.308 0.7 0.060 RF + PM 4.7 0.299 2.4 0.179 RF + S 2.0 0.156 0.7 0.056 注(Note):FP—农民习惯施肥 Farmer's practice; RF—推荐施肥Recommended fertilization; RF-N—推荐施肥减氮 Without N in the recommended fertilization; RF-P—推荐施肥减磷 Without P in the recommended fertilization; RF-K—推荐施肥减钾 Without K in the recommended fertilization; CM—商品有机肥 Commercial organic fertilizer (N + P2O5 + K2O ≥ 5%, N 1.7%); RF + CM、RF + PM 和 RF + S 分别表示推荐施氮量的 1/3 由商品有机肥、猪粪和稻草替代 RF + CM, RF + PM and RF + S represent the replacement of one-third of recommended N input with commercial organic manure, pig manure, and rice straw, respectively. 针对这种状况,本研究将作物年际产量看成灰色量,根据灰色系统理论的一次累加生成原理[25],将有关施肥模式的年际产量逐年累加,与对应的试验年份进行一元线性回归分析。图1和表4的结果表明,各施肥模式的年际累加产量与试验年限均呈现典型的线性关系,一元线性回归方程均达到极显著水平以上。结果揭示了花生和甘薯化肥推荐施肥的趋势产量显著高于习惯施肥;化肥配施猪粪的花生趋势产量显著高于其它施肥模式;化肥配施稻草或猪粪的甘薯趋势产量显著高于其它3种施肥模式,较好地解决了趋势产量的定量评价问题。
灰色系统理论[25]认为,任何随机变量都是在一定幅值范围和一定时区内变化的灰色量,可通过累加生成等缓冲算子,弱化其不确定性,使离乱的原始数据中蕴涵的积分特性或规律清晰地呈现出来,实现对不确定性系统的运行行为和演化规律的正确描述。本文根据连续15年和13年的两个定位试验,对同一个施肥处理的历年产量数据经过一次累加生成,使各年际的离乱产量数据与试验年限呈现一条典型的线性关系,从而实现了对其精确拟合。这样,就能把趋势产量的确定建立在严格的定量技术基础上,使每个施肥模式的长期趋势产量估算都具有鲜明的统计学意义。
本研究仅采用了最简单的一元灰色线性模型,在实际应用中还可以根据原始产量数据的长期动态特征,诸如具有上升或下降趋势的产量数列,构建GM (1,1) 模型或其改进模型,从而实现精确建模和定量分析。
-
综合比较2个定位试验不同施肥模式的各年度产量均值、产量可持续性指数和趋势产量水平,化肥推荐施肥有利于提高赤红壤旱地花生–甘薯轮作制产量的稳定性,花生化肥推荐施肥配施猪粪、甘薯化肥推荐施肥配施稻草或猪粪可进一步提高长期趋势产量水平及其可持续性。
长期不同施肥模式下赤红壤旱地轮作体系内花生甘薯产量的稳定性研究
Yield stability of peanut and sweet potato under long-term combined application of chemical and organic fertilizers in latosolic red soil
-
摘要:
【目的】 研究不同施肥模式下赤红壤旱地花生–甘薯轮作制中两种作物产量的稳定性,可从不同角度加深理解合理养分管理技术的增产机理。 【方法】 化肥定位试验和化肥配施有机肥定位试验在闽东南典型赤红壤旱地分别连续进行了15年和13年,种植制度均为花生和甘薯轮作。化肥定位试验处理包括农民习惯施肥、推荐施肥以及在推荐施肥基础上去掉N、P、K的处理,以不施肥为对照;化肥有机肥定位试验处理包括农民习惯施肥、推荐施肥以及以商品有机肥、猪粪和稻草替代推荐1/3施氮量处理,以不施肥为对照。根据历年花生和红薯产量,计算了产量的可持续性指数和变异指数,构建一元灰色线性模型,计算了不同施肥模式的长期趋势产量。 【结果】 长期化肥定位试验中,推荐施肥处理的花生和甘薯产量均值显著高于其它5个处理,在长期有机无机配合试验中,三个有机无机配施处理的产量均显著高于推荐施肥处理。花生和甘薯化肥推荐施肥的产量可持续指数 (SYI) 分别为0.729 ± 0.019和0.501 ± 0.028,化肥配施猪粪花生的SYI为0.689 ± 0.013,显著高于推荐施肥和化肥配施稻草处理,与配施商品有机肥差异不显著;化肥配施稻草甘薯的SYI为0.514 ± 0.029,显著高于推荐施肥处理和化肥配施商品有机肥处理,与化肥配施猪粪的差异不显著。一元灰色线性模型显示,化肥推荐施肥的花生和甘薯趋势产量均值分别为3780 kg/hm2和19408 kg/hm2,均显著高于其他化肥处理。化肥配施农家猪粪的花生趋势产量均值为3492 kg/hm2,显著高于推荐施肥和其余两个有机无机配施处理;化肥配施稻草处理甘薯的趋势产量均值为17567 kg/hm2,显著高于推荐施肥处理和化肥配施商品有机肥处理,与化肥配施农家猪粪处理没有显著差异。产量可持续系数较变异系数可更加有效地区分不同处理之间的产量波动差异,与产量的绝对高低没有关系。 【结论】 化肥推荐施肥有利于提高赤红壤旱地轮作体系中花生和甘薯的产量,用有机肥替代推荐施肥中氮肥的1/3可以进一步提高花生和甘薯的产量。有机无机配施不仅提高了作物的产量,还不同程度显著提高了产量的稳定性和长期增产趋势。化肥配施猪粪或商品有机肥均可以提高花生产量的稳定性和增产潜力,长期而化肥配合配施稻草或猪粪对甘薯产量及其稳定性最为有利。 Abstract:【Objectives】 Researching the effects of fertilization patterns on stability of crop yield is important for understanding the optimum nutrient management. 【Methods】 Two different experiments involving application of chemical fertilizer or combined application of chemical and organic fertilizers were conducted in an upland (latosolic red soil) of Southeast Fujian for 15 and 13 years, respectively. The planting system in the two experiments was peanut–sweet potato rotation. The chemical fertilizer experiment is composed of no fertilizer (control, CK), farmer's practice (FP), recommended fertilizer (RF) and removal of either N, P or K in the RF. The chemical-organic fertilizer experiment is composed of CK, FP, RF, and replacement of one-third of N in RF with either commercial manure (CM), pig manure (PM) or rice straw (S). The sustainable yield index (SYI) and CV were calculated, and a grey linear model was established to quantitatively analyze long-term yield trend. 【Results】 In the chemical fertilization experiment, the average yields of both crops (peanut 3936 kg/hm2 and sweet potato 19070 kg/hm2), and their SYI (peanut 0.729 ± 0.019 and sweet potato 0.501 ± 0.028) were significantly higher in RF than in the other treatments. The SYI of both crops under RF + PM treatment (peanut 0.689 ± 0.013 and sweet potato 0.514 ± 0.029) were significantly higher than under RF and RF + S treatments, but similar with RF + CM treatment. According to the grey linear trend model, the average yield trend of peanut and sweet potato in RF treatment were 3780 kg/hm2 and 19408 kg/hm2, respectively, and they were significantly higher than those recorded in other treatments. The average yield trend of peanut under RF + PM (3492 kg/hm2) was significantly higher than those in RF, RF + CM and RF + S treatments. The average yield trend of sweet potato under RF + S (17567 kg/hm2) was significantly higher than those in RF and RF + CM, but not significantly different from that in RF + PM. 【Conclusions】 Recommended fertilizer application (RF) can improve the yield stability of peanut and sweet potato in a rotation system. Replacing one-third of N in the recommended fertilizer with pig manure for peanut, or with straw or pig manure for sweet potato could enhance the yield trend level and the sustainable yield index in the long run. -
Key words:
- latosolic red soil /
- long-term experiment /
- peanut /
- sweet potato /
- sustainable yield index /
- yield trend
-
表 1 花生–甘薯轮作制长期定位试验处理和施肥量 (kg/hm2)
Table 1. Treatments and fertilizer input (kg/hm2) in the two long-term peanut–sweet potato rotation experiments
长期定位试验
Long-term experiment处理编号
Treatment number处理
Treatment花生 Peanut 甘薯 Sweet potato N P2O5 K2O 有机肥 Manure N P2O5 K2O 有机肥 Manure 化肥
Chemical fertilizer1 CK 0 0 0 0 0 0 2 FP 90 45 75 225 45 150 3 RF 75 60 90 180 45 225 4 RF-N 0 60 90 0 45 225 5 RF-P 75 0 90 180 0 225 6 RF-K 75 60 0 180 45 0 化肥和有机肥配施
Chemical and organic fertilizer1 CK 0 0 0 0 0 0 0 0 2 RF 75 60 90 0 180 60 225 0 3 RF + CM 50 40 60 1995 120 40 150 4140 4 RF + PM 50 50 71 6585 120 35 177 15795 5 RF + S 50 57 38 2745 120 51 101 6600 注(Note):FP—农民习惯施肥 Farmer's practice; RF—推荐施肥 Recommended fertilization; RF-N—推荐施肥减氮 Without N in the recommended fertilization; RF-P—推荐施肥减磷 Without P in the recommended fertilization; RF-K—推荐施肥减钾 Without K in the recommended fertilization; CM—商品有机肥 Commercial organic fertilizer (N + P2O5 + K2O ≥ 5%, N 1.7%); RF + CM、RF + PM 和 RF + S 分别表示推荐施氮量的 1/3 由商品有机肥、猪粪和稻草替代 RF + CM, RF + PM and RF + S represent the replacement of one-third of recommended N input with commercial organic manure, pig manure, and rice straw, respectively. 表 2 定位试验不同年限轮作体系中各施肥处理花生和甘薯的产量
Table 2. Yield of peanut and sweet potato in two different fertilization experiments and different experimental years
长期试验
Experiment处理
Treatment花生 Peanut (kg/hm2, dry base) 甘薯 Sweet potato (kg/hm2, fresh base) 1~5年
1–5 years6~10年
6–10 years11~15年
11–15 years*平均
Mean1~5年
1–5 years6~10年
6–10 years11~15年
11–15 years*平均
Mean化肥**
Chemical fertilizer applicationCK 1503 ± 764 d 776 ± 351 d 984 ± 538 e 1101 ± 603 e 5649 ± 951 e 6858 ± 2570 e 7033 ± 1934 e 6092 ± 2161 e FP 3674 ± 324 ab 2974 ± 659 ab 3544 ± 277 b 3411 ± 536 b 15760 ± 1400 b 17262 ± 5070 b 16063 ± 3319 b 16178 ± 3356 b RF 4229 ± 305 a 3367 ± 674 a 4147 ± 399 a 3936 ± 608 a 18640 ± 1235 a 20073 ± 6198 a 18976 ± 3155 a 19070 ± 3796 a RF-N 2659 ± 623 c 1747 ± 767 c 2080 ± 481 d 2183 ± 695 d 10210 ± 1773 d 10503 ± 2639 d 10085 ± 1518 d 10020 ± 2101 d RF-P 3236 ± 492 bc 2432 ± 649 b 2791 ± 373 cd 2829 ± 580 c 13833 ± 2177 c 14091 ± 4193 c 14529 ± 2198 c 13667 ± 2889 c RF-K 3251 ± 392 b 2336 ± 601 bc 2870 ± 449 c 2844 ± 594 c 12989 ± 3017 cd 13196 ± 3897 c 12569 ± 2218 c 12798 ± 2894 c 化肥和有机肥配施***
Chemical and organic fertilizerCK 2151 ± 580 b 986 ± 375 c 1131 ± 200 c 1449 ± 926 c 5871 ± 1778 d 8214 ± 1826 b 7928 ± 3472 c 6701 ± 2139 c RF 3558 ± 523 a 3000 ± 584 b 3033 ± 1056 b 3223 ± 532 b 11490 ± 2409 c 16905 ± 4107 a 14640 ± 2077 b 14250 ± 3831 b RF + CM 3765 ± 536 a 3160 ± 624 ab 3217 ± 649 ab 3393 ± 567 ab 12457 ± 2921 bc 18299 ± 3804 a 15476 ± 1893 ab 15355 ± 3950 b RF + PM 4027 ± 478 a 3308 ± 536 a 3478 ± 675 a 3613 ± 540 a 13116 ± 2837 b 19337 ± 3722 a 19643 ± 3647 a 16681 ± 4563 ab RF + S 3646 ± 483 a 2927 ± 577 b 3226 ± 638 ab 3269 ± 546 b 14624 ± 2518 a 19464 ± 5762 a 17829 ± 1447 ab 17085 ± 4370 a 注(Note):*—化肥有机肥配合试验为第三阶段时间为 11~13 年; **—化肥长期地位试验花生产量方差分析的处理间 F 值和年际间 F 值分别为 213.9 和 26.8; 甘薯产量的处理间 F 值和年际间 F 值则分别为 161.8 和 21.7; ***—化肥有机肥配合长期定位试验的花生产量方差分析的处理间和年际间的 F 值分别为 132.8 和 23.4; 甘薯产量的处理间和年际间的 F 值则分别为 60.6 和 15.4; 同列数据后不同字母表示同一试验处理间差异达 5% 显著水平。*—The period for the chemical and organic fertilizer combination experiment was 11 to 13 years; **—In the chemical fertilization experiment, the inter-treatment and inter-annual F values for peanut yield were 213.9 and 26.8, and those of sweet potato yield were 161.8 and were 21.7, respectively; ***—In the chemical and organic fertilizer combination experiment, the inter-treatment and inter-annual F values for peanut yield were 132.8 and 23.4, and those of sweet potato were 60.6 and 15.4, respectively. Values followed by different small letters in each column indicate significant differences among treatments in the same experiment at 0.05 level. FP—农民习惯施肥 Farmer's practice; RF—推荐施肥 Recommended fertilization; RF-N—推荐施肥减氮 Without N in the recommended fertilization; RF-P—推荐施肥减磷 Without P in the recommended fertilization; RF-K—推荐施肥减钾 Without K in the recommended fertilization; CM—商品有机肥 Commercial organic fertilizer (N + P2O5 + K2O ≥ 5%, N 1.7%); RF + CM、RF + PM 和 RF + S分别表示推荐施氮量的 1/3 由商品有机肥、猪粪和稻草替代 RF + CM, RF + PM and RF + S represent the replacement of one-third of recommended N input with commercial organic manure, pig manure, and rice straw, respectively. 表 3 长期不同施肥处理对轮作体系中花生和甘薯产量可持续性指数的影响
Table 3. Effects of different fertilization patterns on sustainable yield index (SYI) of peanut and sweet potato
化肥长期定位试验 Chemical fertilizer experiment 化肥有机肥配施长期定位试验 Chemical and organic fertilizer experiment 处理 Treatment 花生 Peanut 甘薯 Sweet potato 处理 Treatment 花生 Peanut 甘薯 Sweet potato CK 0.101 ± 0.006 e 0.128 ± 0.020 e CK 0.117 ± 0.004 c 0.182 ± 0.003 d FP 0.631 ± 0.014 b 0.421 ± 0.016 b RF 0.605 ± 0.012 b 0.421 ± 0.011 c RF 0.729 ± 0.019 a 0.501 ± 0.028 a RF+CM 0.630 ± 0.023 ab 0.462 ± 0.011 b RF-N 0.324 ± 0.032 d 0.262 ± 0.008 d RF+PM 0.689 ± 0.013 a 0.492 ± 0.023 ab RF-P 0.491 ± 0.021 c 0.355 ± 0.022 c RF+S 0.609 ± 0.033 b 0.514 ± 0.029 a RF-K 0.492 ± 0.020 c 0.328 ± 0.013 c F值 F value 1380.4** 288.5** F值 F value 359.7** 216.5** 注(Note):FP—农民习惯施肥 Farmer's practice; RF—推荐施肥 Recommended fertilization; RF-N—推荐施肥减氮 Without N in the recommended fertilization; RF-P—推荐施肥减磷 Without P in the recommended fertilization; RF-K—推荐施肥减钾 Without K in the recommended fertilization; CM—商品有机肥 Commercial organic fertilizer (N + P2O5 + K2O ≥ 5%, N 1.7%); RF + CM、RF + PM和RF + S 分别表示推荐施氮量的 1/3 由商品有机肥、猪粪和稻草替代 RF + CM, RF + PM and RF + S represent the replacement of one-third of recommended N input with commercial organic manure, pig manure, and rice straw, respectively. 同列数据后不同字母表示同一试验处理间差异达 5% 显著水平 Values followed by different small letters in each column indicate significant differences among treatments at 0.05 level. 表 4 轮作系统中长期不同施肥处理的花生和甘薯产量趋势灰色线性模型
Table 4. Linear grey model of peanut and sweet potato yield trends as affected by fertilization patterns
定位试验
Experiment处理
Treatments花生 Peanut 甘薯 Sweet potato Y = a + bt F值
F valueR2 b的95%置信区间
95% confidence interval of bY = a + bt F值
F valueR2 b的95%置信区间
95% confidence interval of b化肥
Chemical fertilizer experimentCK Y = 2598 + 889t 1055.5** 0.988 829~948 Y = –2844 + 6472t 1263.7** 0.990 6078~6865 FP Y = 1095 + 3276t 6928.3** 0.998 3191~3361 Y = –2322 + 16477t 3258.2** 0.996 15853~17100 RF Y = 1096 + 3780t 5363.0** 0.998 3668~3891 Y = –3980 + 19408t 4234.4** 0.997 18763~20052 RF-N Y = 2715 + 1959t 2578.9** 0.995 1876~2042 Y = 1403 + 10072t 3567.3** 0.996 9708~10436 RF-P Y = 2162 + 2650t 5583.1** 0.998 2574~2727 Y = 1270 + 13678t 3089.7** 0.996 13147~14210 RF-K Y = 1993 + 2652t 3663.2** 0.996 2557~2747 Y = 2989 + 12637t 2934.5** 0.996 12133~13141 化肥有机肥配施
Chemical and organic fertilizer experimentCK Y = 4208 + 1159t 606.2** 0.982 1056~1263 Y = –1209 + 6818t 1153.8** 0.991 6376~7260 RF Y = 1990 + 3093t 8566.9** 0.999 3028~3158 Y = –7392 + 14807t 1728.1** 0.994 14023~15591 RF + CM Y = 1951 + 3276t 9789.1** 0.999 3203~3349 Y = –7286 + 15901t 1683.4** 0.994 15048~16754 RF + PM Y = 1905 + 3492t 9474.6** 0.999 3413~3571 Y = –10408 + 17177t 1252.5** 0.991 16109~18245 RF + S Y = 1581 + 3154t 10332.0** 0.999 3086~3222 Y = –7092 + 17567t 2231.6** 0.995 16749~18386 注(Note):一元回归方程中,Y 表示产量,t表示试验年限,b表示斜率,即为年际趋势产量。In the unitary regression equation, Y is the yield, t is the experimental years, b is the slope, indicating the inter-annual yield trend. FP—农民习惯施肥 Farmer's practice; RF—推荐施肥 Recommended fertilization; RF-N—推荐施肥减氮 Without N in the recommended fertilization; RF-P—推荐施肥减磷 Without P in the recommended fertilization; RF-K—推荐施肥减钾 Without K in the recommended fertilization; CM—商品有机肥 Commercial organic fertilizer (N + P2O5 + K2O ≥ 5%, N 1.7%); RF + CM、RF + PM 和 RF + S 分别表示推荐施氮量的 1/3 由商品有机肥、猪粪和稻草替代 RF + CM, RF + PM and RF + S represent the replacement of one-third of recommended N input with commercial organic manure, pig manure, and rice straw, respectively.
**—P < 0.01.表 5 长期不同施肥处理下轮作体系中花生和甘薯产量的变异系数和可持续性指数 (SYI)
Table 5. The coefficient of variation (CV) and sustainable yield index of peanut and sweet potato under different treatments in two long-term experiments
定位试验
Experiment处理
Treatment花生 Peanut 甘薯 Sweet potato CV (%) SYI CV (%) SYI 化肥
Chemicalfertilizer experimentCK 55.3 ± 2.0 a 0.175 ± 0.014 d 36.9 ± 9.6 a 0.339 ± 0.114 b FP 15.7 ± 0.7 d 0.711 ± 0.027 a 22.1 ± 2.0 b 0.500 ± 0.030 a RF 15.6 ± 1.3 d 0.730 ± 0.026 a 21.4 ± 2.1 b 0.501 ± 0.028 a RF-N 32.5 ± 1.6 b 0.407 ± 0.029 c 21.8 ± 3.2 b 0.579 ± 0.056 a RF-P 21.1 ± 1.3 c 0.579 ± 0.042 b 22.4 ± 2.3 b 0.523 ± 0.052 a RF-K 21.3 ± 2.2 c 0.598 ± 0.050 b 23.4 ± 1.7 b 0.519 ± 0.033 a F值 F value 565.0** 199.4** 9.8** 9.8** 化肥有机肥配施
Chemical and organic fertilizer experimentCK 64.1 ± 1.0 a 0.138 ± 0.004 b 33.2 ± 1.2 a 0.442 ± 0.030 b FP 16.7 ± 0.7 b 0.637 ± 0.018 a 27.3 ± 2.2 b 0.489 ± 0.020 ab RF + CM 17.6 ± 2.1 b 0.631 ± 0.034 a 26.1 ± 0.6 b 0.515 ± 0.005 a RF + PM 15.4 ± 1.2 b 0.670 ± 0.027 a 27.6 ± 2.1 b 0.507 ± 0.031 a RF + S 17.3 ± 1.4 b 0.643 ± 0.037 a 26.0 ± 2.1 b 0.513 ± 0.026 a F值F value 641.6** 10.2** 213.2** 5.6* 注(Note):FP—农民习惯施肥 Farmer's practice; RF—推荐施肥 Recommended fertilization; RF-N—推荐施肥减氮 Without N in the recommended fertilization; RF-P—推荐施肥减磷 Without P in the recommended fertilization; RF-K—推荐施肥减钾 Without K in the recommended fertilization; CM—商品有机肥Commercial organic fertilizer (N + P2O5 + K2O ≥ 5%,N 1.7%); RF + CM、RF + PM和RF + S分别表示推荐施氮量的1/3由商品有机肥、猪粪和稻草替代 RF + CM, RF + PM and RF + S represent the replacement of one-third of recommended N input with commercial organic manure, pig manure, and rice straw, respectively. 同列数据后不同小写字母表示同一定位试验处理间该指标差异显著达 P < 0.05 水平 Values followed by different small letters in each column indicate significant differences among treatments in the same experiment at 0.05 level. 表 6 一元线性回归对长期不同施肥处理花生和甘薯产量趋势的拟合精度
Table 6. Fitting accuracy of linear regression on the yield trendofpeanut and sweet potato under long-term fertilization
长期定位试验
Experiment处理
Treatment花生 Peanut 甘薯 Sweet potato F值 F value R2 F值 F value R2 化肥
Chemical fertilizer application experimentCK 1.9 0.127 0.0 0.000 FP 0.0 0.000 0.0 0.000 RF 0.1 0.004 0.1 0.007 RF-N 1.5 0.101 0.6 0.043 RF-P 1.3 0.091 0.2 0.013 RF-K 0.5 0.034 0.3 0.020 化肥有机肥配施
Chemical and organic fertilizer experimentCK 8.3* 0.429 0.3 0.003 FP 5.2* 0.320 0.8 0.064 RF + CM 4.9* 0.308 0.7 0.060 RF + PM 4.7 0.299 2.4 0.179 RF + S 2.0 0.156 0.7 0.056 注(Note):FP—农民习惯施肥 Farmer's practice; RF—推荐施肥Recommended fertilization; RF-N—推荐施肥减氮 Without N in the recommended fertilization; RF-P—推荐施肥减磷 Without P in the recommended fertilization; RF-K—推荐施肥减钾 Without K in the recommended fertilization; CM—商品有机肥 Commercial organic fertilizer (N + P2O5 + K2O ≥ 5%, N 1.7%); RF + CM、RF + PM 和 RF + S 分别表示推荐施氮量的 1/3 由商品有机肥、猪粪和稻草替代 RF + CM, RF + PM and RF + S represent the replacement of one-third of recommended N input with commercial organic manure, pig manure, and rice straw, respectively. -
[1] Calderini D F, Slafer G A. Changes in yield and yield stability in wheat during the 20th century[J]. Field Crops Research, 1998, 57: 335–347. doi: 10.1016/S0378-4290(98)00080-X [2] Verón S R, José M P, Slafer G A. Inter annual variability of wheat yield in the Argentine Pampas during the 20th century[J]. Agriculture Ecosystems and Environment, 2004, 103: 177–190. doi: 10.1016/j.agee.2003.10.001 [3] Ladha J K, Dawe D, Pathak H. How extensive are yield declines in long-term rice–wheat experiments in Asia[J]. Field Crops Research, 2003, 81(2/3): 159–180. [4] Manna M C, Swarup A, Wanjari R H, Mishra B, Shahi D K. Long-term fertilization, manure and liming effects on soil organic matter and crop yields[J]. Soil and Tillage Research, 2007, 94(2): 397–409. doi: 10.1016/j.still.2006.08.013 [5] 陈欢, 曹承富, 孔令聪, 等. 长期施肥下淮北砂姜黑土区小麦产量稳定性研究[J]. 中国农业科学, 2014, 47(13): 2580–2590. Chen H, Cao C F, Kong L C, et al. Study on wheat yield stability in Huaibei lime concretion black soil area based on long-term fertilization experiment[J]. Scientia Agricultura Sinica, 2014, 47(13): 2580–2590. doi: 10.3864/j.issn.0578-1752.2014.13.010 [6] 高洪军, 彭畅, 张秀芝, 等. 长期不同施肥对东北黑土区玉米产量稳定性的影响[J]. 中国农业科学, 2015, 48(23): 4790–4799. Gao H J, Pen C, Zhang X Z, et al. Effect of long-term different fertilization on maize yield stability in the northeast black soil region[J]. Scientia Agricultura Sinica, 2015, 48(23): 4790–4799. doi: 10.3864/j.issn.0578-1752.2015.23.020 [7] 张秀芝, 高洪军, 彭畅, 等. 长期有机培肥黑土有机碳、全氮及玉米产量稳定性的变化特征[J]. 植物营养与肥料学报, 2019, 25(9): 1473–1481. Zhang X J, Gao H J, Pen C, et al. Variation trend of soil organic carbon, total nitrogen and the stability of maize yield in black soil under long–term organic fertilization[J]. Journal of Plant Nutrition and Fertilizers, 2019, 25(9): 1473–1481. doi: 10.11674/zwyf.18390 [8] 门明新, 李新旺, 许皞. 长期施肥对华北平原潮土作物产量及稳定性的影响[J]. 中国农业科学, 2008, 41(8): 2339–2346. Men M X, Li X W, Xu H. Effects of long– term fertilization on crop yield and stability[J]. Scientia Agricultura Sinica, 2008, 41(8): 2339–2346. doi: 10.3864/j.issn.0578-1752.2008.08.017 [9] 李新旺, 陈亚恒, 王树涛, 等. 长期不同施肥下潮土养分生产力及其可持续性演变[J]. 土壤学报, 2010, 47(3): 555–562. Li X W, Chen Y H, Wan S T, et al. Soil nutrient productivity and sustainability in Fluvo–aquic soil under long-term fertilization[J]. Acta Pedologica Sinica, 2010, 47(3): 555–562. doi: 10.11766/trxb2010470323 [10] 魏猛, 张爱君, 诸葛玉平, 等. 长期施肥下甘薯产量稳定性及品质特性研究[J]. 西北农业学报, 2017, 26(4): 588–595. Wei M, Zhang A J, Zhuge Y P, et al. Study of different fertilization practices on sweet potato yield stability and quality characteristics[J]. Acta Agriculturae Boreali-occidentalis Sinica, 2017, 26(4): 588–595. doi: 10.7606/j.issn.1004-1389.2017.04.014 [11] 王婷, 丁宁平, 李利利, 等. 化肥与有机肥或秸秆配施提高陇东旱塬黑垆土上作物产量的稳定性和可持续性[J]. 植物营养与肥料学报, 2019, 25(11): 1817–1826. Wan T, Ding N P, Li L L, et al. Combining chemical fertilizer with organic manure or straw increase the yield stability and sustainability of maize and wheat in Loess Plateau of east Gansu Province[J]. Journal of Plant Nutrition and Fertilizers, 2019, 25(11): 1817–1826. doi: 10.11674/zwyf.19190 [12] 刘彦伶, 李渝, 张雅蓉, 等. 长期氮磷钾肥配施对贵州黄壤玉米产量和土壤养分可持续性的影响[J]. 应用生态学报, 2017, 28(11): 3581–3588. Liu Y L, Li Y, Zhang Y R, et al. Effect of long-term application of NPK fertilizer on maize yield and yellow soil nutrients sustainability in Guizhou, China[J]. Chinese Journal of Applied Ecology, 2017, 28(11): 3581–3588. [13] 马力, 杨林章, 沈明星, 等. 基于长期定位试验的典型稻麦轮作区作物产量稳定性研究[J]. 农业工程学报, 2011, 27(4): 117–124. Ma L, Yang L Z, Shen M X, et al. Study on crop yield stability in a typical region of rice–wheat rotation based on long-term fertilization experiment[J]. Chinese Society of Agricultural Engineering, 2011, 27(4): 117–124. doi: 10.3969/j.issn.1002-6819.2011.04.020 [14] 廖育林, 郑圣先, 聂军, 等. 长期施用化肥和稻草对红壤水稻土肥力和生产力持续性的影响[J]. 中国农业科学, 2009, 42(10): 3541–3550. Liao Y L, Zheng S X, Nie J, et al. Effects of long-term application of fertilizer and rice straw on soil fertility and sustainability of a reddish paddy soil productivity[J]. Scientia Agricultura Sinica, 2009, 42(10): 3541–3550. doi: 10.3864/j.issn.0578-1752.2009.10.0020 [15] 黄欠如, 胡锋, 李辉信, 等. 红壤性水稻土施肥的产量效应及与气候、地力的关系[J]. 土壤学报, 2006, 43(6): 926–929. Huang Q R, Hu F, Li H X, et al. Crop yield response to fertilization and its relations with climate and soil fertility in red paddy soil[J]. Acta Pedologica Sinica, 2006, 43(6): 926–929. doi: 10.3321/j.issn:0564-3929.2006.06.007 [16] 吴焕焕, 徐明岗, 吕家珑. 长期不同施肥条件下红壤水稻产量可持续性特征[J]. 西北农林科技大学学报(自然科学版), 2014, 42(7): 163–168. Wu H H, Xu M G, Lu J L, et al. Yield sustainability of paddy soil rice with long-term fertilization[J]. Journal of Northwest A& F University(Natural Science Edition), 2014, 42(7): 163–168. [17] 冀建华, 侯红乾, 刘益仁, 等. 长期施肥对双季稻产量变化趋势、稳定性和可持续性的影响[J]. 土壤学报, 2015, 52(3): 607–618. Ji J H, Hou H Q, Liu Y R, et al. Effect of long-term fertilization on yield variation trend, yield stability and sustainability in the double cropping rice system[J]. Acta Pedologica Sinica,, 2015, 52(3): 607–618. [18] 李忠芳, 徐明岗, 逄焕成, 等. 中国南方潴育性水稻土产量演变及其肥力驱动因素分析[J]. 土壤学报, 2014, 51(5): 953–962. Li Z F, Xu M G, Pang H C, et al. The rice yield trend in three hydragric paddy soil in south China and its fertility factors[J]. Acta Pedologica Sinica, 2014, 51(5): 953–962. [19] 李忠芳, 徐明岗, 张会民, 等. 长期施肥下作物产量演变特征的研究进展[J]. 西南农业学报, 2012, 25(6): 2387–2302. Li Z F, Xu M G, Zhang H M, et al. Summarize of crop yield dynamic under long–term fertilization[J]. Southwest China Journal of Agricultural Science, 2012, 25(6): 2387–2302. doi: 10.3969/j.issn.1001-4829.2012.06.087 [20] 鲁如坤. 土壤农业化学分析方法[M]. 北京: 中国农业科技出版社, 2000.146–196. Lu R K. Soil Agrochemical Methods[M]. Beijing: China Agricultural Science and Technology Press, 2000.146–196. [21] 章明清, 林琼, 杨杰, 等. 花生–甘薯轮作制中磷钾肥施肥模型研究[J]. 土壤通报, 2004, 35(6): 758–762. Zhang M Q, Lin Q, Yang J, et al. Phosphate and potassium fertilizer application model in peanut and sweet potato rotation[J]. Chinese Journal of soil Science, 2004, 35(6): 758–762. doi: 10.3321/j.issn:0564-3945.2004.06.019 [22] Piepho H. Methods for comparing the yield stability of cropping systems[J]. Journal of Agronomy and Crop Science, 1998, 180(4): 193–213. doi: 10.1111/j.1439-037X.1998.tb00526.x [23] Singh P R, Das S K, Bhaskara R U M. Towards sustainable dryland agricultural practices[M]. Hyderabad, India: Central Research Institute for Dryland Agriculture, 1990.106. [24] Manna M C, Swarup A, Wanjari P H, et al. Long-term effect of fertilizer and manure application on soil organic carbon storage, soil quality and yield sustainability under sub-humid and semi-arid tropical[J]. India Field Crops Research, 2005, 93: 264–280. doi: 10.1016/j.fcr.2004.10.006 [25] 刘思峰, 杨英杰, 吴利丰, 等. 灰色系统理论及其应用(第七版)[M]. 北京: 科学出版社, 2018. Liu S F, Yang Y J, Wu L F, et al. Grey system theory and its application (7th Edition)[M]. Beijing: Science Press, 2018. [26] 薛毅, 陈立萍. 实用数据分析与MATLAB软件[M]. 北京: 北京工业大学出版社, 2015.189–274, 377–394. Xue Y, Chen L P. Practical data analysis and MATLAB software[M]. Beijing: Beijing University of Technology Press, 2015.189–274, 377–394. [27] William S, Mohamed H G. Sustainable crop production intensification in the Senegal and Niger River basins of francophone West Africa[J]. International Journal of Agricultural Sustainability, 2011, 9(1): 171–185. doi: 10.3763/ijas.2010.0559 [28] 李忠芳, 徐明岗, 张会民, 等. 长期不同施肥模式对我国玉米产量可持续性的影响[J]. 玉米科学, 2009, 17(6): 82–87. Li Z F, Xu M G, Zhang H M, et al. Effects of long-term different fertilizations on sustainability of maize yield in China[J]. Journal of Maize Sciences, 2009, 17(6): 82–87. [29] 杨生茂, 李凤民, 索东让, 等. 长期施肥对绿洲农田土壤生产力及土壤硝态氮积累的影响[J]. 中国农业科学, 2005, 38(10): 2043–2052. Yang S M, Li F M, Suo D R, et al. Effect of long-term fertilization on soil productivity and nitrate accumulation in Gansu Oasis[J]. Scientia Agricultura Sinica, 2005, 38(10): 2043–2052. doi: 10.3321/j.issn:0578-1752.2005.10.015 [30] 李秀英, 李燕婷, 赵秉强, 等. 褐潮土长期定位不同施肥制度土壤生产功能演化研究[J]. 作物学报, 2006, 32(5): 683–689. Li X Y, Li Y T, Zhao B Q, et al. The dynamics of crop yields under different fertilization systems in drab fluvo-aquic soil[J]. Acta Agronomica Sinica, 2006, 32(5): 683–689. doi: 10.3321/j.issn:0496-3490.2006.05.010 [31] 李忠芳, 徐明岗, 张会民, 等. 长期施肥和不同生态条件下我国作物产量可持续性特征[J]. 应用生态学报, 2010, 21(5): 1264–1269. Li Z F, Xu M G, Zhang H M, et al. Sustainability of crop yields in China under long-term fertilization and different ecologicalconditions[J]. Chinese Journal of Applied Ecology, 2010, 21(5): 1264–1269. [32] Manna M C, Swarup A, Wanjari R H, et al. Long-term effect of fertilizer and manure application on soil organic carbon storage, soil quality and yield sustainability under sub-humid and semi-arid tropical India[J]. Field Crops Research, 2005, 93: 264–280. doi: 10.1016/j.fcr.2004.10.006 [33] 宋永林, 唐华俊, 李小平. 长期施肥对作物产量及褐潮土有机质变化的影响研究[J]. 华北农学报, 2007, 22(增刊ent): 100–105. Song Y L, Tang H J, Li X P. The effects of long-term fertilization on crop yield and aqui-cinnamon soil organic matter[J]. Acta Agriculturae Boreali-Sinica, 2007, 22(Supplment): 100–105. [34] 李忠芳, 徐明岗, 张会民, 等. 长期施肥下中国主要粮食作物产量的变化[J]. 中国农业科学, 2009, 42(7): 2407–2414. Li Z F, Xu M G, Zhang H M, et al. Grain yield trends of different food crop under long-term fertilization in China[J]. Scientia Agricultura Sinica, 2009, 42(7): 2407–2414. doi: 10.3864/j.issn.0578-1752.2009.07.020 [35] 宇万太, 赵鑫, 张璐, 马强. 长期施肥对作物产量的贡献[J]. 生态学杂志, 2007, 26(12): 2040–2044. Yu W T, Zhao X, Zhang L, Ma Q. Contribution of long-term fertilization to crop yield[J]. Chinese Journal of Ecology, 2007, 26(12): 2040–2044. [36] 冯耀宗. 人工生态系统稳定性概念及其指标[J]. 生态学杂志, 2002, 21(5): 58–60. Feng Y Z. Concept and criteria of stability for managed ecosystem[J]. Chinese Journal of Ecology, 2002, 21(5): 58–60. doi: 10.3321/j.issn:1000-4890.2002.05.013 -