Scholarlink

<p>This study developed locally calibrated allometric models to estimate aboveground biomass (AGB) and belowground biomass (BGB) of <i>Cocos nucifera</i> (coconut palm) in Tanzania as functions of diameter at breast height (D) and total height (H). Destructive sampling was conducted on 46 and 29 for AGB and BGB, respectively, across two contrasting coastal districts: Mkuranga, located along the Indian Ocean, and Kisarawe, situated farther inland. Four log–log transformed and one&#xa0;untransformed linear models were fitted to estimate AGB and BGB using a mixed-effects approach, with site included as a random effect. The model performance was assessed using R<sup>2</sup>, RMSE, AIC, and Mean Prediction Error (PE%). Leave-one-out cross-validation (LOOCV) was applied to evaluate predictive reliability given the limited sample size. Model results showed moderate explanatory power, with R<sup>2</sup> ranging from 0.62 to 0.75 for AGB and 0.52 to 0.53 for BGB. Contrary to conventional findings for dicotyledonous trees, H consistently outperformed D as the strongest predictor for both AGB and BGB. This pattern reflects the monocot growth form of <i>C. nucifera</i>, where D stabilises early while biomass accumulates primarily through vertical growth. Models combining H and D did not significantly improve predictive accuracy and often produced insignificant coefficients. LOOCV confirmed that H-only models yielded the lowest bias and most consistent predictive performance. The findings further indicate that applying generic approaches, such as pantropical tree allometric models or volume-based methods using form factors and average wood density, may lead to biased AGB estimates for <i>C. nucifera</i>. The study provides the allometric equations for <i>C. nucifera</i> in Tanzania, offering an essential tool for biomass quantification in coconut-based agroforestry systems and carbon accounting frameworks. These models represent a methodological advance over generic defaults and support a more accurate assessment of carbon stocks in coastal agricultural landscapes.</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

Allometric models for estimating above- and belowground biomass of Cocos nucifera L. at Kisarawe and Mkuranga Districts in Tanzania

  • 期刊论文

  • 发表日期：

    2026年6月25日

查看全文
  • Wilson Ancelm Mugasha,
  • Eliakimu Zahabu

摘要

This study developed locally calibrated allometric models to estimate aboveground biomass (AGB) and belowground biomass (BGB) of Cocos nucifera (coconut palm) in Tanzania as functions of diameter at breast height (D) and total height (H). Destructive sampling was conducted on 46 and 29 for AGB and BGB, respectively, across two contrasting coastal districts: Mkuranga, located along the Indian Ocean, and Kisarawe, situated farther inland. Four log–log transformed and one untransformed linear models were fitted to estimate AGB and BGB using a mixed-effects approach, with site included as a random effect. The model performance was assessed using R2, RMSE, AIC, and Mean Prediction Error (PE%). Leave-one-out cross-validation (LOOCV) was applied to evaluate predictive reliability given the limited sample size. Model results showed moderate explanatory power, with R2 ranging from 0.62 to 0.75 for AGB and 0.52 to 0.53 for BGB. Contrary to conventional findings for dicotyledonous trees, H consistently outperformed D as the strongest predictor for both AGB and BGB. This pattern reflects the monocot growth form of C. nucifera, where D stabilises early while biomass accumulates primarily through vertical growth. Models combining H and D did not significantly improve predictive accuracy and often produced insignificant coefficients. LOOCV confirmed that H-only models yielded the lowest bias and most consistent predictive performance. The findings further indicate that applying generic approaches, such as pantropical tree allometric models or volume-based methods using form factors and average wood density, may lead to biased AGB estimates for C. nucifera. The study provides the allometric equations for C. nucifera in Tanzania, offering an essential tool for biomass quantification in coconut-based agroforestry systems and carbon accounting frameworks. These models represent a methodological advance over generic defaults and support a more accurate assessment of carbon stocks in coastal agricultural landscapes.