Retrieval of the phycocyanin concentration (PC), a characteristic pigment of, and proxy for, cyanobacteria biomass, from hyperspectral satellite remote sensing measurements is challenging due to uncertainties in the remote sensing reflectance (∆R_rs) resulting from atmospheric correction and instrument radiometric noise. Although several individual algorithms have been proven to capture local variations in cyanobacteria biomass in specific regions, their performance has not been assessed on hyperspectral images from satellite sensors. Our work leverages a machine-learning model, Mixture Density Networks (MDNs), trained on a large (N = 939) dataset of collocated in situ chlorophyll-a concentrations (Chla), PCs, and remote sensing reflectance (R_rs) measurements to estimate PC from all relevant spectral bands. The performance of the developed model is demonstrated via PC maps produced from select images of the Hyperspectral Imager for the Coastal Ocean (HICO) and Italian Space Agency's PRecursore IperSpettrale della Missione Applicativa (PRISMA) using a matchup dataset. As input to the MDN, we incorporate a combination of widely used band ratios (BRs) and line heights (LHs) taken from existing multispectral algorithms, that have been proven for both Chla and PC estimation, as well as novel BRs and LHs to increase the overall cyanobacteria biomass estimation accuracy and reduce the sensitivity to ∆R_rs. When trained on a random half of the dataset, the MDN achieves uncertainties of 44.3%, which is less than half of the uncertainties of all viable optimized multispectral PC algorithms. The MDN is notably better than multispectral algorithms at preventing overestimation on low (<10 mg m⁻³) PC. Visibly, HICO and PRISMA PC maps show the wider dynamic range that can be represented by the MDN. The available in situ and satellite-derived R_rs matchups and measured in situ PC demonstrate the robustness of the MDN for estimating low (<10 mg m⁻³) PC and the reduced impact of ∆R_rs on medium-to-high in situ PC (>10 mg m⁻³). According to our extensive assessments, the developed model is anticipated to enable practical PC products from PRISMA and HICO, therefore the model is promising for planned hyperspectral missions, such as the Plankton Aerosol and Cloud Ecosystem (PACE). This advancement will enhance the complementary roles of hyperspectral radiometry from satellite and low-altitude platforms for quantifying and monitoring cyanobacteria harmful algal blooms at both large and local spatial scales.

由于大气校正和仪器辐射噪声导致遥感反射率 (∆R _rs ) 的不确定性，从高光谱卫星遥感测量中检索藻蓝蛋白浓度 (PC)，这是一种蓝藻生物量的特征色素和代表，具有挑战性. 尽管几种单独的算法已被证明可以捕获特定区域蓝藻生物量的局部变化，但尚未在卫星传感器的高光谱图像上评估它们的性能。我们的工作利用了机器学习模型，即混合密度网络 (MDN)，该模型在大型 ( N  = 939) 的原位叶绿素a浓度 (Chl a)、PC 和遥感反射率 (R _rs ) 测量，以从所有相关光谱带估计 PC。所开发模型的性能通过使用匹配数据集从沿海高光谱成像仪 (HICO) 和意大利航天局的 PRecursore IperSpettrale della Missione Applicativa (PRISMA) 的精选图像生成的 PC 地图来证明。作为 MDN 的输入，我们结合了广泛使用的波段比 (BR) 和线高 (LH) 的组合，这些算法取自现有的多光谱算法，这些算法已被证明可用于 Chl a和 PC 估计，以及新颖的 BR 和 LH，以提高整体蓝藻生物量估计精度并降低对 ∆R _rs的敏感性. 在随机一半的数据集上训练时，MDN 的不确定性为 44.3%，这不到所有可行的优化多光谱 PC 算法的不确定性的一半。MDN 在防止对低 (<10 mg m ^-3 ) PC 的高估方面明显优于多光谱算法。显然，HICO 和 PRISMA PC 地图显示了 MDN 可以表示的更宽的动态范围。可用的原位和卫星衍生的 R _rs匹配以及原位PC测量证明了 MDN 在估算低 (<10 mg m ^-3 ) PC 方面的稳健性以及 ∆R _rs对中高原位的影响降低PC (>10 mg m ^-3）。根据我们的广泛评估，开发的模型有望实现 PRISMA 和 HICO 的实用 PC 产品，因此该模型有望用于计划中的高光谱任务，例如浮游生物气溶胶和云生态系统 (PACE)。这一进步将增强来自卫星和低空平台的高光谱辐射测量在大型和局部空间尺度上量化和监测蓝藻有害藻华的互补作用。