基于Sentinel-2影像的不同耕作方式下玉米关键生育期长势差异分析

Growth Differences Analysis of Maize at Key Growth Stages under Different Tillage Practices Based on Sentinel-2 Imagery

摘要:

玉米长势差异分析是评估不同耕作方式对作物生长影响的重要手段。以莫旗试验区为例，基于Sentinel-2遥感监测参数反演裸土期秸秆覆盖度，并据此划分免耕、少耕和传统耕作3种类型;在玉米关键生育期（拔节期、抽雄期和乳熟期）反演叶面积指数（LAI）、植被覆盖度（FVC）、地上生物量（AGB）、植株含氮量（PNC）、叶绿素含量（CAB）和冠层含水量（CWC）6项指标，采用变异系数法构建玉米综合长势指数（CGI），进而从空间、均值和平均变化速率方面开展不同耕作方式下玉米长势差异分析。结果表明：拔节期CGI整体偏低（均值0.101），抽雄期升至峰值（均值0.435），乳熟期略有下降（均值0.394），整体变化呈先升后缓降趋势，且不同地块间表现出明显的空间异质性。不同耕作方式下玉米长势呈现“前期传统耕作占优，后期免耕占优”的阶段性特征。拔节期至抽雄期传统耕作增幅最大（0.365），变化速率（0.010）高于免耕（0.009），主要得益于LAI与CWC的较高贡献;抽雄期至乳熟期免耕降幅最小（0.038），变化速率（0）明显优于少耕（-0.001）和传统耕作（-0.002），主要归因于CAB对衰减的抑制能力，从而形成“前期传统耕作驱动强、后期免耕稳长能力优”的阶段性差异格局。研究结果可为玉米智慧种植决策与精准田间管理提供关键长势参数和科学依据，支撑区域尺度保护性耕作实施效果的定量评估。

Abstract:

Analyzing maize growth differences is a critical approach for evaluating the impacts of varied tillage practices on crop development. Taking Moqi experimental zone as study area, the crop residue cover during the bare-soil period was firstly retrieved based on Sentinel-2 remote sensing parameters, and accordingly the fields were classified into three tillage types: no-till, reduced-till, and conventional tillage. During key maize growth stages (jointing, tasseling, and milk stages), six biophysical indicators, i.e., leaf area index (LAI), fractional vegetation cover (FVC), above-ground biomass (AGB), plant nitrogen content (PNC), chlorophyll a/b (CAB), and canopy water content (CWC) were retrieved. A comprehensive growth index (CGI) was constructed by using the coefficient of variation method. Subsequently, spatial patterns, mean values, and average change rates were analyzed to compare maize growth performance across tillage types. Results showed that CGI was relatively low at the jointing stage (mean: 0.101), peaked at the tasseling stage (mean: 0.435), and slightly declined at the milk stage (mean: 0.394), exhibiting a trend of initial increase followed by a gradual decrease, with pronounced spatial heterogeneity among plots. Under different tillage regimes, maize growth exhibited a stage-dependent pattern characterized by "conventional tillage dominance in the early stage and no-till dominance in the later stage." From jointing to tasseling, conventional tillage showed the greatest increment (0.365) and a higher change rate (0.010) than no-till (0.009), mainly attributed to the higher contributions of LAI and CWC. From tasseling to milk stage, no-till presented the smallest decline (0.038) with a change rate of 0, markedly outperforming reduced-till (-0.001) and conventional tillage (-0.002), which was primarily due to the suppressing effect of CAB on growth deterioration. These findings revealed a distinct phased differentiation pattern: conventional tillage drived stronger early-stage growth, whereas no-till exhibited superior stability in maintaining growth during the later stage. The research result can provide key growth parameters and scientific evidence for smart decision-making in maize cultivation and precision field management, which can support the quantitative assessment of conservation tillage effectiveness at regional scale.