Description: To improve our understanding of the role of phytoplankton for marine ecosystems and global biogeochemical cycles, information on the global distribution of major phytoplankton groups is essential. Although algorithms have been developed to assess phytoplankton diversity from space for over two decades, so far the application of these data sets has been limited. This scientific roadmap identifies user needs, summarizes the current state of the art, and pinpoints major gaps in long-term objectives to deliver space-derived phytoplankton diversity data that meets the user requirements. These major gaps in using ocean color to estimate phytoplankton community structure were identified as: (a) the mismatch between satellite, in situ and model data on phytoplankton composition, (b) the lack of quantitative uncertainty estimates provided with satellite data, (c) the spectral limitation of current sensors to enable the full exploitation of backscattered sunlight, and (d) the very limited applicability of satellite algorithms determining phytoplankton composition for regional, especially coastal or inland, waters. Recommendation for actions include but are not limited to: (i) an increased communication and round-robin exercises among and within the related expert groups, (ii) the launching of higher spectrally and spatially resolved sensors, (iii) the development of algorithms that exploit hyperspectral information, and of (iv) techniques to merge and synergistically use the various streams of continuous information on phytoplankton diversity from various satellite sensors' and in situ data to ensure long-term monitoring of phytoplankton composition.

Description: Oxygenic photosynthesis is responsible for virtually all of the biochemical production of organic matter, and regulates atmospheric carbon dioxide and oxygen concentrations, which profoundly affects both climate and biogeochemical cycles. Observations of biomass, health and productivity of ocean and land ecosystems are crucial for monitoring changes in the Earth system. Here, space-borne hyperspectral data, mainly of the SCIAMACHY instrument (onboard ENVISAT), have been investigated in terms of application to globalobservations of marine and terrestrial primary producers. This study focuses primarily on retrieving inelastic processes from the natural waters: CDOM and chlorophyll (chl) a fluorescence. Originally, the chl a fluorescence retrieval was developed in its red peak and for the ocean application only. However, it was further extended to the far-red peak of chl a fluorescence, and subsequently applied to terrestrial scenes. All retrievals are based on the Differential Optical Absorption Spectroscopy (DOAS), and involve fitting spectral features of the filling in of the Fraunhofer lines by fluorescence processes. The reference spectra of chl a and CDOM fluorescence, used in the DOAS fits, were calculated with the ocean-atmosphere coupled radiative transfer model SCIATRAN. Furthermore, a simple algorithm to retrieve a chl proxy of terrestrial vegetation was developed. The retrievals were developed with the use of simulated radiances, and subsequently applied to SCIAMACHY data. Although the original aim of the SCIAMACHY instrument was to monitor atmospheric composition, its unique spectral characteristics (namely broad spectral range from 240 nm to 2380 nm, and high resolution of 0.2 nm to 1.5 nm) have enabled previously other novel retrievals to be developed. These included observations of inelastic processes (rotational and vibrational Raman scattering), and marine phytoplank- ton. In this thesis, the chl a fluorescence and chl proxy retrievals were applied to eight years of the SCIAMACHY data (2004-2011). In addition to presenting yearly composites and monthly climatologies of the obtained results, monthly averages were applied to study the seasonality of both, marine phytoplankton and terrestrial vegetation. Modeling studies of CDOM fluorescence, followed by preliminary retrievals applied to SCIAMACHY data, have not been successful in retrieving CDOM fluorescence from hyperspectral satellite data. On the other hand, the obtained chl a fluorescence results showed good spatial agreement with other datasets. Marine observations of the red peak of chl a fluorescence captured successfully the phytoplankton seasonal cycles and interannual variability for two studied regions: a subregion of the Indian Ocean near Madagascar, and the equatorial Pacific. Good agreement with multispectral ocean color products (MODIS nFLH and MODIS Chl a) was obtained. Response of phytoplankton to climate fluctuations, as expressed by Multivariate ENSO Index, was observed for the equatorial Pacific. In case of land observations, all retrieved parameters (red and far-red chl a fluorescence, and chl proxy) followed the seasonal cycles of vegetation for five regions representing different biomes worldwide (croplands in the North America, evergreen needleaf forest in Euroasia, evergreen broadleaf forest in Central Africa, woody savannas in Central Africa and savannas in Southern Africa). However, the three SCIAMACHY datasets did not show exactly the same seasonal pattern, and their relationship varied over time and among biomes. This proves that the retrieved parameters do not carry the same information on vegetation, and hence suggests that they all should be used simultaneously for observations of vegetation dynamics. The calculated ratio and the difference of the two peaks of chl a fluorescence, followed the increase of chl content and canopy development, which supports previous findings by in situ measurements and models.The red and far-red chl a fluorescence and chl proxy algorithms have enabled simultaneous retrievals of multiple parameters of marine phytoplankton and terrestrial vegetation. While the application of the SCIAMACHY results is still constrained (mainly by the noisiness of the results and spatio-temporal resolution of the satellite measurements), the developed retrievals and their successful application to studies of ocean and land phenology have advanced the prospects of observations of phytoplankton and terrestrial vegetation with hyperspectral satellite instruments.

Description: SCIATRAN is a comprehensive software package which is designed to model radiative transfer processes in the terrestrial atmosphere and ocean in the spectral range from the ultraviolet to the thermal infrared (0.18–40 μm). It accounts for multiple scattering processes, polarization, thermal emission and ocean–atmosphere coupling. The main goal of this paper is to present a recently developed version of SCIATRAN which takes into account accurately inelastic radiative processes in both the atmosphere and the ocean. In the scalar version of the coupled ocean–atmosphere radiative transfer solver presented by Rozanov et al. [61] we have implemented the simulation of the rotational Raman scattering, vibrational Raman scattering, chlorophyll and colored dissolved organic matter fluorescence. In this paper we discuss and explain the numerical methods used in SCIATRAN to solve the scalar radiative transfer equation including trans-spectral processes, and demonstrate how some selected radiative transfer problems are solved using the SCIATRAN package. In addition we present selected comparisons of SCIATRAN simulations with those published benchmark results, independent radiative transfer models, and various measurements from satellite, ground-based, and ship-borne instruments. The extended SCIATRAN software package along with a detailed User's Guide is made available for scientists and students, who are undertaking their own research typically at universities, via the web page of the Institute of Environmental Physics (IUP), University of Bremen: http://www.iup.physik.uni-bremen.de.

Description: Satellite remote sensing has been widely used in the last decades for agricultural applications, both for assessing vegetation condition and for subsequent yield prediction. Existing remote sensing-based methods to estimate gross primary productivity (GPP), which is an important variable to indicate crop photosynthetic function and stress, typically rely on empirical or semi-empirical approaches, which tend to over-simplify photosynthetic mechanisms. In this work, we take advantage of all parallel developments in mechanistic photosynthesis modeling and satellite data availability for an advanced monitoring of crop productivity. In particular, we combine process-based modeling with the soil-canopy energy balance radiative transfer model (SCOPE) with Sentinel-2 and Landsat 8 optical remote sensing data and machine learning methods in order to estimate crop GPP. With this approach, we by-pass the need for an intermediate step to retrieve the set of vegetation biophysical parameters needed to accurately model photosynthesis, while still accounting for the complex processes of the original physically-based model. Several implementations of the machine learning models are tested and validated using simulated and flux tower-based GPP data. Our final neural network model is able to estimate GPP at the tested flux tower sites with r2 of 0.92 and RMSE of 1.38 gC d−1 m−2, which outperforms empirical models based on vegetation indices. The first test of applicability of this model to Landsat 8 data showed good results (r2 of 0.82 and RMSE of 1.97 gC d−1 m−2), which suggests that our approach can be further applied to other sensors. Modeling and testing is restricted to C3 crops in this study, but can be extended to C4 crops by producing a new training dataset with SCOPE that accounts for the different photosynthetic pathways. Our model successfully estimates GPP across a variety of C3 crop types and environmental conditions even though it does not use any local information from the corresponding sites. This highlights its potential to map crop productivity from new satellite sensors at a global scale with the help of current Earth observation cloud computing platforms.

Description: SCIATRAN is a comprehensive software package which is designed to model radiative transfer processes in the terrestrial atmosphere and ocean in the spectral range from the ultraviolet to the thermal infrared (0.18–40 μm). It accounts for multiple scattering processes, polarization, thermal emission and ocean–atmosphere coupling. The main goal of this paper is to present a recently developed version of SCIATRAN which takes into account accurately inelastic radiative processes in both the atmosphere and the ocean. In the scalar version of the coupled ocean–atmosphere radiative transfer solver presented by Rozanov et al. [61] we have implemented the simulation of the rotational Raman scattering, vibrational Raman scattering, chlorophyll and colored dissolved organic matter fluorescence. In this paper we discuss and explain the numerical methods used in SCIATRAN to solve the scalar radiative transfer equation including trans-spectral processes, and demonstrate how some selected radiative transfer problems are solved using the SCIATRAN package. In addition we present selected comparisons of SCIATRAN simulations with those published benchmark results, independent radiative transfer models, and various measurements from satellite, ground-based, and ship-borne instruments. The extended SCIATRAN software package along with a detailed User's Guide is made available for scientists and students, who are undertaking their own research typically at universities, via the web page of the Institute of Environmental Physics (IUP), University of Bremen: http://www.iup.physik.uni-bremen.de.

Description: Agriculture monitoring and yield estimation are important for food security, both regionally and globally.Understanding the year-to-year variability in crop yield and its relationship to meteorological conditions isparticularly important for several regions where yield is highly dependent on changing environmental factors.For example, wheat yields in India have been steadily increasing since the 1960s and 1970s due to benefits ofthe Green Revolution, but in recent years the wheat yield has been unstable as major crop yield losses wereattributed to unfavorable meteorological conditions. Modeling the effects of various environmental variables canbe challenging, as their impact on the final yield is complex and varies depending on their intensity and thecrop growth stage at which they occur (e.g., moderate rainfall is beneficial for crops, but extensive and untimelyrainfalls can lead to huge yield losses).In this work, we exploit interconnections between meteorological conditions and satellite data on vegeta-tion during the whole growing season, and their simultaneous impact on wheat yield in the Wheat Belt inIndia. We use GLDAS 2.1 data as the meteorological input and MODIS data for the vegetation remote sensinginput. Adding satellite information on crop is crucial for yield estimation, as it carries information on both cropphenology, as well as the crop response to the meteorological conditions. We apply machine learning algorithms(e.g., convolutional neural networks, CNNs) that can model non-linear processes and can extract importantfeatures in the multivariate time series automatically from data, without prior knowledge or human effort in featuredesign. By doing so, we do not force assumptions on which time is the most important for the final crop yieldand we can include in the analysis the whole time series of multiple input variables at a high temporal resolution.Furthermore, we analyze the CNNs in terms of important features and crucial time windows for yield estimation,which shows that they the vary across space and time. By combining meteorological and satellite vegetation datawith CNNs this work may help to disentangle the complex interactions between the features in the time series ofthe input data and the wheat yield.

Description: Remote sensing data analysis retrieves spatial-temporal information about the Earth‘s surface from remotely sensed optical and radar images. For this purpose accurate and efficient classification or parameter quantification techniques must be used. Consequently, there exists a long tradition in remote sensing to employ methods and techniques from the field of machine learning. They can be regarded as „universal function approximators“ that are able to link any data in order to derive connections, conclusions and predictions efficiently using different learning strategies. In the following, current research topics of the Remote Sensing section of the GFZ are presented, in which different forms of machine learning are used.