Seagrass condition, resilience and ecosystem services are affected by the below-ground tissues (BGr) but these are rarely monitored. In this study we compiled historical data across northern Australia to investigate biomass allocation strategies in 13 tropical seagrass species. There was sufficient data to undertake statistical analysis for five species: Cymodocea serrulata, Halophila ovalis, Halodule uninervis, Thalassia hemprichii, and Zostera muelleri. The response of below-ground biomass (BGr) to above-ground biomass (AGr) and other environmental and seagrass community composition predictor variables were assessed using Generalized Linear Models. Environmental data included: region, season, sediment type, water depth, proximity to land-based sources of pollution, and a light stress index. Seagrass community data included: species diversity and dominant species class (colonising, opportunistic or persistant) based on biomass. The predictor variables explained 84–97% of variance in BGr on the log-scale depending on the species. Multi-species meadows showed a greater investment into BGr than mono-specific meadows and when dominated by opportunistic or persistent seagrass species. This greater investment into BGr is likely to enhance their resistance to disturbances if carbohydrate storage reserves also increase with biomass. Region was very important for the estimation of BGr from AGr in four species (not in C. serrulata). No temporally changing environmental features were included in the models, therefore, they cannot be used to predict local-scale responses of BGr to environmental change. We used a case study for Cairns Harbour to predict BGr by applying the models to AGr measured at 362 sites in 2017. This case study demonstrates how the model can be used to estimate BGr when only AGr is measured. However, the general approach can be applied broadly with suitable calibration data for model development providing a more complete assessment of seagrass resources and their potential to provide ecosystem services.

海草状况，复原力和生态系统服务受地下组织（BGr）的影响，但很少对其进行监测。在这项研究中，我们汇总了澳大利亚北部的历史数据，以调查13种热带海草物种的生物量分配策略。有足够的数据为五个品种进行统计分析：丝粉藻属石楠，喜盐草，二药藻，Thalassia hemprichii和大叶muelleri。使用广义线性模型评估了地下生物量（BGr）对地上生物量（AGr）以及其他环境和海草群落组成预测变量的响应。环境数据包括：地区，季节，沉积物类型，水深，与陆地污染源的距离以及轻度压力指数。海草群落数据包括：基于生物量的物种多样性和优势物种类别（殖民化，机会主义或持久性）。预测变量解释了对数尺度上BGr变异的84–97％，具体取决于物种。与单种草甸相比，多物种草甸对BGr的投资更大，并且以机会性或持久性海草物种为主导。如果碳水化合物的储藏量也随着生物量的增加而增加，对BGr的更多投资可能会增强其抗干扰能力。区域对于从四个物种的AGr估算BGr非常重要（不是C. serrulata）。模型中没有包含随时间变化的环境特征，因此，它们不能用于预测BGr对环境变化的局部响应。我们使用凯恩斯港口的案例研究，通过将模型应用于2017年在362个站点上测得的AGr来预测BGr。该案例研究演示了仅测量AGr时如何使用该模型估算BGr。但是，一般方法可通过适当的校准数据广泛应用于模型开发，从而提供对海草资源及其提供生态系统服务潜力的更完整评估。