滨海湿地植被地上生物量遥感反演

盐沼植被生物量是滨海湿地生态系统碳循环研究的重要参数,是湿地生态系统健康评价、资源可持续利用的关键指标,开展盐沼植被地上生物量监测方法研究具有重要意义。目前,遥感技术在湿地生物量监测领域已经得到广泛应用,但反演方法仍以统计模型为主,模型构建需要实测数据支撑,时空拓展性不强。选择江苏盐城丹顶鹤保护区为研究区,基于冠层辐射(PROSAIL)传输模型,通过局部和全局敏感性分析,对模型参数本地化,构建了互花米草地上生物量半经验反演模型,应用于Landsat 8 OLI遥感影像,获得了互花米草地上生物量的时空分布。研究结果表明,利用PROSAIL模型模拟互花米草冠层反射率,叶面积指数(LAI)、叶片干物质含量(Cm)、叶倾角分布参数(LIDF)、等效水厚度(Cw)、叶绿素含量(Cab)、叶片结构参数(N)为高敏感性参数,类胡萝卜素含量(Car)、土壤参数(P_soil)为低敏感性参数;利用不同时刻的遥感影像反演了地上生物量,遥感反演结果与实测数据对比,拟合度R²为0.83,均方根误差(RMSE)为0.43kg/m²,平均相对误差(MRE)为15.7%,精度较高,模型具有较好的时空普适性。研究发展了盐沼植被地上生物量遥感反演方法,解决了以往过于依赖现场实测数据构建反演模型的局限性,该方法可以为研究滨海湿地生态系统碳循环以及准确估算其碳汇潜力提供技术支持。

Inversion of aboveground biomass of saltmarshes in coastal wetland using remote sensing

The biomass of saltmarsh vegetation serves as a crucial parameter in investigating the carbon cycling within coastal wetland ecosystems. It represents a pivotal indicator for assessing the ecological well-being of wetland ecosystems and ensuring the sustainable utilization of their resources. Hence, undertaking research on monitoring methods for the aboveground biomass of saltmarsh vegetation is of utmost significance. Currently, remote sensing technology has been widely applied in the field of wetland biomass monitoring. However, the biomass retrieval methods primarily rely on statistical models, which require support from field measurements during the model construction process, thereby limiting their temporal and spatial scalability. To address this issue, the Yancheng Wetland National Nature Reserve of Rare Birds in Jiangsu Province was selected as the study area for this research. By utilizing the PROSAIL vegetation radiative transfer model, the input parameters of the model were localized through local sensitivity analysis and global sensitivity analysis. Consequently, a semi-empirical inversion model was developed to estimate the aboveground biomass of Spartina alterniflora. This model was subsequently applied to Landsat 8 OLI satellite imagery, allowing for the retrieval of the spatiotemporal distribution of aboveground biomass for Spartina alterniflora. The research findings indicate that in the simulation of canopy reflectance for Spartina alterniflora using the PROSAIL model, parameters such as leaf area index (LAI), leaf dry matter content (Cm), leaf inclination distribution function (LIDF), Leaf equivalent water thinness (Cw), leaf chlorophyll content (Cab), and leaf mesophyll structure (N) exhibit high sensitivity. On the other hand, parameters carotenoid content (Car) and the soil brightness factor (P_soil) demonstrate low sensitivity. The aboveground biomass of vegetation was estimated for the corresponding time periods using remote sensing imagery captured at different time points. The remote sensing inversion results were compared with the field measurements, resulting in a coefficient of determination (R²) of 0.83, a root mean square error (RMSE) of 0.43 kg/m², and a mean relative error (MRE) of 15.7%. These findings demonstrate a high level of accuracy and precision in the model''s estimation. Furthermore, the results affirm the model''s good spatiotemporal applicability, indicating its suitability across different locations and time periods. This study has developed a remote sensing-based method for estimating aboveground biomass of coastal saltmarsh vegetation, addressing the limitations of previous approaches that heavily relied on field measurements for biomass inversion modeling. The proposed method provides technical support for studying carbon cycling in coastal wetland ecosystems and accurately estimating their carbon sequestration potential.