研究目的
The primary objectives of this study therefore were to investigate: (1) whether temporal patterns of phenologic development phases facilitate species differentiation on base of spectral signatures, applying the identical sample design and instrumentation as Wolf et al. [13]; (2) whether annual water temperature oscillations affect species-specific growth of submersed macrophytes; and (3) if native and invasive species have different tolerance to an increase in water temperature.
研究成果
The study confirmed that in-situ measured spectral variations within the growing season can be linked to phenologic stages of macrophyte populations. An interrelation between macrophyte growth and water temperature was demonstrated for the indigenous species Chara spp. and P. perfoliatus, but not for the invasive species E. nuttallii. The reflectance models developed proved to be able to mitigate gaps in in-situ data collection and deliver simulated spectra for each day and all sun positions of possible optical satellite data takes. The inversion of the models for Chara spp. and P. perfoliatus with the aim of macrophyte classification on species level is successful.
研究不足
The prediction limits of the reflectance models are directly related to the daily and seasonal distribution of in-situ measurements and the specific environmental frame conditions of the respective year. Short-term events like draughts, floods or turbidity after an intense rainfall cannot directly be represented by the models. The accuracy of the classification results depends on the number of spectra of the spectral database.
1：Experimental Design and Method Selection:
The study involved systematic collection of remote sensing reflectance spectra of submerged macrophytes and sediment during the growing seasons of 2011 and 2015 at Lake Starnberg, Germany, using a hyperspectral underwater spectroradiometer. The methodology included the development of reflectance models based on spectral libraries and the use of these models to analyze spectral signatures in relation to plant phenology and water temperature.
2：Sample Selection and Data Sources:
Measurements were taken at the same study site and with the same measurement setup in both years to cover different phenologic stages of three macrophyte species: Chara spp., Elodea nuttallii, and Potamogeton perfoliatus.
3：List of Experimental Equipment and Materials:
The measurement setup consisted of three submersible RAMSES spectroradiometers (ACC-VIS and ARC-VIS; spectral range: 320 nm to 950 nm; TriOS Mess- und Datentechnik GmbH, Rastede, Germany) and an underwater camera system (Canon PowerShot G10, Canon, Tokyo, Japan).
4：Experimental Procedures and Operational Workflow:
Field campaigns were conducted approximately every 3 weeks under cloud-free conditions. Measurements at each single day were designed to start in the morning hours and to last until late afternoon. Each dataset consisted of 20 replicates within 3 min at the same fixed place and depths.
5：Data Analysis Methods:
The data processing chain was developed with Python. For each dataset, the remote sensing reflectance spectra Rrs(b) and Rrs(0) of the two depths (depth b and 0) were calculated for 20 measurements of Ed and Lu. The spectra were smoothed by Savitzky–Golay filter of length 5. The reflectance models were developed using R, with linear interpolation of irregular gridded data applied to cover the complete vegetation season and differing sun heights.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
