Scholarlink

Aims <p>Flooded rice rotations in the Everglades Agricultural Area (EAA) are widely used to slow subsidence of organic Histosols, yet it remains unclear how much soil health is driven by (i) flood depth within the operational range, versus (ii) sampling time, interannual variability, and (iii) soil quality index (SQI) framework choice.</p> Methods <p>We conducted a two-year field experiment (2019–2020) testing four continuous flood depths (5, 10, 15, and 20&#xa0;cm). Nine soil indicators were measured at pre-plant and post-harvest. Treatment effects were evaluated using factorial analysis of variance (ANOVA) with year, sampling time, and flood depth as fixed factors and plot as the experimental unit; principal component analysis (PCA) identified minimum datasets (MDS); and structural equation modeling (SEM) quantified direct and indirect pathways. We compared ten SQI approaches (CASH, FSHA, and eight TDS/MDS indices differing by linear versus nonlinear scoring and weighted versus unweighted aggregation) and assessed sensitivity and yield linkages.</p> Results <p>Sampling time significantly affected all indicators, whereas flood depth influenced only bulk density, maximum water holding capacity, and soil protein (p &lt; 0.05). Bulk density declined and soil protein increased markedly from pre-plant to post-harvest in both years. PCA indicated year-specific dominant controls: active carbon and total Kjeldahl N drove variability in 2019, with soil protein emerging as an additional key contributor in 2020. SEM showed flood-depth effects were weaker than indicator-indicator linkages. SQI behavior diverged strongly by framework: MDS-based indices were most responsive (up to 754% change relative to the 2019 pre-plant baseline), while CASH/FSHA were least sensitive (3.65–4.93%) and were positively related to yield (R<sup>2</sup> = 0.79, <i>p</i> = 0.003); most TDS/MDS indices showed negative yield-ΔSQI relationships.</p> Conclusions <p>In EAA flooded rice systems, soil health inference depends critically on sampling time, interannual variability, and SQI construction. Flood depth (5–20&#xa0;cm) is a secondary driver over a two-year window, affecting only select physical/biological properties. SQI choice should therefore be purpose-driven, balancing sensitivity to change against functional relevance (e.g., yield tracking).</p> Graphical Abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Flood-depth effects on rice soil health depend on sampling time, year, and soil quality index framework

  • 期刊论文

  • 发表日期：

    2026年4月23日

查看全文
  • Yuchuan Fan,
  • Naba R. Amgain,
  • Abul Rabbany,
  • Matthew VanWeelden,
  • Jehangir H. Bhadha

摘要

Aims

Flooded rice rotations in the Everglades Agricultural Area (EAA) are widely used to slow subsidence of organic Histosols, yet it remains unclear how much soil health is driven by (i) flood depth within the operational range, versus (ii) sampling time, interannual variability, and (iii) soil quality index (SQI) framework choice.

Methods

We conducted a two-year field experiment (2019–2020) testing four continuous flood depths (5, 10, 15, and 20 cm). Nine soil indicators were measured at pre-plant and post-harvest. Treatment effects were evaluated using factorial analysis of variance (ANOVA) with year, sampling time, and flood depth as fixed factors and plot as the experimental unit; principal component analysis (PCA) identified minimum datasets (MDS); and structural equation modeling (SEM) quantified direct and indirect pathways. We compared ten SQI approaches (CASH, FSHA, and eight TDS/MDS indices differing by linear versus nonlinear scoring and weighted versus unweighted aggregation) and assessed sensitivity and yield linkages.

Results

Sampling time significantly affected all indicators, whereas flood depth influenced only bulk density, maximum water holding capacity, and soil protein (p < 0.05). Bulk density declined and soil protein increased markedly from pre-plant to post-harvest in both years. PCA indicated year-specific dominant controls: active carbon and total Kjeldahl N drove variability in 2019, with soil protein emerging as an additional key contributor in 2020. SEM showed flood-depth effects were weaker than indicator-indicator linkages. SQI behavior diverged strongly by framework: MDS-based indices were most responsive (up to 754% change relative to the 2019 pre-plant baseline), while CASH/FSHA were least sensitive (3.65–4.93%) and were positively related to yield (R2 = 0.79, p = 0.003); most TDS/MDS indices showed negative yield-ΔSQI relationships.

Conclusions

In EAA flooded rice systems, soil health inference depends critically on sampling time, interannual variability, and SQI construction. Flood depth (5–20 cm) is a secondary driver over a two-year window, affecting only select physical/biological properties. SQI choice should therefore be purpose-driven, balancing sensitivity to change against functional relevance (e.g., yield tracking).

Graphical Abstract