Search

Identifying and characterizing vascular plants in time and space is required in various disciplines, e.g. in forestry, conservation and agriculture. Remote sensing emerged as a key technology revealing both spatial and temporal vegetation patterns. Harnessing the ever growing streams of remote sensing data for the increasing demands on vegetation assessments and monitoring requires efficient, accurate and flexible methods for data analysis. In this respect, the use of deep learning methods is trend-setting, enabling high predictive accuracy, while learning the relevant data features independently in an end-to-end fashion. Very recently, a series of studies have demonstrated that the deep learning method of Convolutional Neural Networks (CNN) is very effective to represent spatial patterns enabling to extract a wide array of vegetation properties from remote sensing imagery. This review introduces the principles of CNN and distils why they are particularly suitable for vegetation remote sensing. The main part synthesizes current trends and developments, including considerations about spectral resolution, spatial grain, different sensors types, modes of reference data generation, sources of existing reference data, as well as CNN approaches and architectures. The literature review showed that CNN can be applied to various problems, including the detection of individual plants or the pixel-wise segmentation of vegetation classes, while numerous studies have evinced that CNN outperform shallow machine learning methods. Several studies suggest that the ability of CNN to exploit spatial patterns particularly facilitates the value of very high spatial resolution data. The modularity in the common deep learning frameworks allows a high flexibility for the adaptation of architectures, whereby especially multi-modal or multi-temporal applications can benefit. An increasing availability of techniques for visualizing features learned by CNNs will not only contribute to interpret but to learn from such models and improve our understanding of remotely sensed signals of vegetation. Although CNN has not been around for long, it seems obvious that they will usher in a new era of vegetation remote sensing.

Ecosystem restoration and reforestation often operate at large scales, whereas monitoring practices are usually limited to spatially restricted field measurements that are (i) time- and labour-intensive, and (ii) unable to accurately quantify restoration success over hundreds to thousands of hectares. Recent advances in remote sensing technologies paired with deep learning algorithms provide an unprecedented opportunity for monitoring changes in vegetation cover at spatial and temporal scales. Such data can feed directly into adaptive management practices and provide insights into restoration and regeneration dynamics. Here, we demonstrate that convolutional neural network (CNN) segmentation algorithms can accurately classify the canopy cover of

Xylella fastidiosa (Xf) is a harmful plant pathogenic bacterium, able to infect over 500 plant species worldwide. Successful eradication and containment strategies for harmful pathogens require large-scale monitoring techniques for the detection of infected hosts, even when they do not display visual symptoms. Although a previous study using airborne hyperspectral and thermal imagery has shown promising results for the early detection of Xf-infected olive (Olea europaea) trees, further work is needed when adopting these techniques for large scale monitoring using multispectral cameras on board airborne platforms and satellites. We used hyperspectral and thermal imagery collected during a two-year airborne campaign in a Xf-infected area in southern Italy to assess the performance of spectrally constrained machine-learning algorithms for this task. The algorithms were used to assess multispectral bandsets, selected from the original hyperspectral imagery, that were compatible with large-scale monitoring from unmanned platforms and manned aircraft. In addition, the contribution of solar–induced chlorophyll fluorescence (SIF) and the temperature-based Crop Water Stress Index (CWSI) retrieved from hyperspectral and thermal imaging, respectively, were evaluated to quantify their relative importance in the algorithms used to detect Xf infection. The detection performance using support vector machine algorithms decreased from ∼80% (kappa, κ = 0.42) when using the original full hyperspectral dataset including SIF and CWSI to ∼74% (κ = 0.36) when the optimal set of six spectral bands most sensitive to Xf infection were used in addition to the CWSI thermal indicator. When neither SIF nor CWSI were used, the detection yielded less than 70% accuracy (decreasing κ to very low performance, 0.29), revealing that tree temperature was more important than chlorophyll fluorescence for the Xf detection. This work demonstrates that large-scale Xf monitoring can be supported using airborne platforms carrying multispectral and thermal cameras with a limited number of spectral bands (e.g., six to 12 bands with 10 nm bandwidths) as long as they are carefully selected by their sensitivity to the Xf symptoms. More precisely, the blue (bands between 400 and 450 nm to derive the NPQI index) and thermal (to derive CWSI from tree temperature) were the most critical spectral regions for their sensitivity to Xf symptoms in olive., Data collection was partially supported by the European Union’s Horizon 2020 research and innovation program through grants to the POnTE (Pest Organisms threatening Europe; grant 635646 from European Union’s Horizon 2020 Framework Research Programme) and XF-ACTORS (Xylella fastidiosa Active Containment Through a Multidisciplinary-Oriented Research Strategy; grant 727987 from European Union’s Horizon 2020 Framework Research Programme) projects. A. Hornero was supported by a research fellowship DTC GEO 29 “Detection of global photosynthesis and forest health from space” from the Science Doctoral Training Centre (Swansea University, UK).

Plant functional traits (‘traits’) are essential for assessing biodiversity and ecosystem processes, but cumbersome to measure. To facilitate trait measurements, we test if traits can be predicted through visible morphological features by coupling heterogeneous photographs from citizen science (iNaturalist) with trait observations (TRY database) through Convolutional Neural Networks (CNN). Our results show that image features suffice to predict several traits representing the main axes of plant functioning. The accuracy is enhanced when using CNN ensembles and incorporating prior knowledge on trait plasticity and climate. Our results suggest that these models generalise across growth forms, taxa and biomes around the globe. We highlight the applicability of this approach by producing global trait maps that reflect known macroecological patterns. These findings demonstrate the potential of Big Data derived from professional and citizen science in concert with CNN as powerful tools for an efficient and automated assessment of Earth’s plant functional diversity.

Plant functional traits play a key role in the assessment of ecosystem processes and properties. Optical remote sensing is ascribed a high potential in capturing those traits and their spatiotemporal patterns. In vegetation remote sensing, reflectance-based retrieval methods are either statistical (relying on empirical observations) or physically-based (based on inversions of a radiative transfer model, RTM). Both trait retrieval approaches remain poorly investigated regarding phenology. However, within the phenology of a plant, its leaf constituents, canopy structure, and the presence of phenology-related organs (i.e., flowers or inflorescence) vary considerably – and so does its reflectance. We, therefore, addressed the question of how plant phenology affects the predictive performance of both statistical and RTM-based methods and how this effect differs between traits. For a complete growing season, we weekly measured traits of 45 herbaceous plant species together with hyperspectral canopy reflectance (ASD FieldSpec III). Plants were grown in an experimental setup. The investigated traits comprised Leaf Area Index (LAI) and the leaf traits chlorophyll, anthocyanins, carotenoids, equivalent water thickness, and leaf mass per area. We compared the predictive performances of PLSR models and three variants of PROSAIL inversions based on (1) all observations and based on (2) a phenological subset where flowering plants were excluded and only those observations most suitable for modeling were kept. Our results show that both statistical and RTM-based trait retrievals were largely affected by phenology. For carotenoids for example, R2 decreased from 0.58 at non-flowering canopies to 0.25 at 100% flowering canopies. Temporal trends were diverse. LAI and equivalent water thickness were best estimated earlier in the growing season; chlorophyll and carotenoids towards senescence. PLSR models showed generally higher bias than the PROSAIL-based retrieval approaches. Lookup-table inversion of PROSAIL in combination with a continuous wavelet transformation of reflectance showed highest accuracies. We found RTM-based retrieval not to be as accurate and transferable as previously indicated. Our results suggest that phenology is essential for accurate retrieval of plant functional traits and varies depending on the studied species and functional traits, respectively.

Many water quality parameters such as concentrations of suspended matter, nutrients and algae directly or indirectly change the electromagnetic reflectance and transmission properties of surface water bodies. Optical measurement approaches have shown great potential to partially substitute water sampling and laboratory analyses, but are obstructed by limited flexibility or high maintenance demands. In order to overcome these problems and to bridge the gap between in situ and remote sensing measurements, the use of close-range, above-surface reflectance measurements in the VIS-NIR domain to measure water quality parameters in surface water bodies was investigated. Remote sensing reflectance in a 1 m³ water tank with increasing, known concentrations of suspended solids was measured. A partial least squares model was trained to predict concentrations from reflectance curves, which performed well, considering the wide range of concentrations and illumination conditions (R²cal = 0.96, R²val = 0.97). The approach was then transferred to the field and further parameters were tested. Using a semi-autonomous spectrometer mounted to a boom stand on a motor boat, we traced substance concentrations in close intervals along a longitudinal gradient from inflow to dam in a drinking water reservoir in Brazil. The method is suitable for parameters directly influencing the reflection properties of the water body (e.g. suspended solids (R²cal = 0.93), chlorophyll-a (R²cal = 0.74)), or for parameters closely related to those (e.g. total phosphorus (R²cal = 0.97)). For chemical oxygen demand, the method is not well suited (R²cal = 0.14, R²val = 0.45). Once calibrated to the local conditions, the spectrometer can be used stationary or on moving platforms to map and monitor surface waters. The integration of the procedure into acoustic and imaging techniques is further investigated.

Plant pathogens cause significant losses to agricultural yields and increasingly threaten food security1, ecosystem integrity and societies in general2–5. Xylella fastidiosa is one of the most dangerous plant bacteria worldwide, causing several diseases with profound impacts on agriculture and the environment6. Primarily occurring in the Americas, its recent discovery in Asia and Europe demonstrates that X. fastidiosa’s geographic range has broadened considerably, positioning it as a reemerging global threat that has caused socioeconomic and cultural damage7,8. X. fastidiosa can infect more than 350 plant species worldwide9, and early detection is critical for its eradication8. In this article, we show that changes in plant functional traits retrieved from airborne imaging spectroscopy and thermography can reveal X. fastidiosa infection in olive trees before symptoms are visible. We obtained accuracies of disease detection, confirmed by quantitative polymerase chain reaction, exceeding 80% when high-resolution fluorescence quantified by three-dimensional simulations and thermal stress indicators were coupled with photosynthetic traits sensitive to rapid pigment dynamics and degradation. Moreover, we found that the visually asymptomatic trees originally scored as affected by spectral plant-trait alterations, developed X. fastidiosa symptoms at almost double the rate of the asymptomatic trees classified as not affected by remote sensing. We demonstrate that spectral plant-trait alterations caused by X. fastidiosa infection are detectable previsually at the landscape scale, a critical requirement to help eradicate some of the most devastating plant diseases worldwide.

The estimation of various forest inventory attributes from high spatial resolution airborne remote sensing data has been widely examined and proved to be successful at the experimental level. Nevertheless, the operational use of these data in automated procedures to support forest inventories and forest management is still limited to a small number of cases. The reasons for this are high data costs, limited availability of remote sensing data over large areas and resistance from practitioners. In this review the main aim is to stimulate debate about spaceborne very high resolution stereo-imagery (VHRSI) as an alternative to airborne remote sensing data by presenting: (1) a case study on the retrieval of stand density, aboveground biomass and tree species using a set of easy-to-calculate variables obtained from VHRSI data combined with image processing and nonparametric classification and modelling approaches; and (2) the results of an expert opinion survey on the potential of VHRSI as compared with Light Detection and Ranging (LiDAR), hyperspectral and airborne digital imagery to derive a range of forest inventory attributes. In the case study, stand density was estimated with r² = 0.71 and RMSE = 156 trees (rel./norm. RMSE = 24.9 per cent/12.4 per cent), biomass with r² = 0.64 and RMSE of 36.7 t/ha (rel./norm. RMSE = 20.0 per cent/12.8 per cent) while tree species classifications with five species reached overall accuracies of 84.2 per cent (kappa = 0.81). These results were comparable to earlier studies in the same test site, obtained with more expensive airborne acquisitions. Expert opinions were more diverse for VHRSI and aerial photographs (Shannon index values of 0.94 and 0.97) than for LiDAR and hyperspectral data (Shannon index values 0.69 and 0.88). In our opinion, this reflects the current state-of-the-art in the application of VHRSI for automatically retrieving forest inventory attributes. The number of studies using these data is still limited, and the full potential of these datasets is not yet completely explored. Compared with LiDAR and hyperspectral data, which both mostly received high scores for forest inventory products matching the sensor systems’ strengths, VHRSI and aerial photographs received more homogeneous scores indicating their potential as multi-purpose instruments to collect forest inventory information. In summary, considering the simpler acquisition, reasonable price and the comparably easy data format and handling of VHRSI compared with other sensor types, we recommend further research on the application of these data for supporting operational forest inventories.

With consumer-grade unmanned aerial vehicles (UAVs) on the rise, which enable easy, time-flexible, and cost-effective acquisition of very high-resolution RGB data, the mapping of forest tree species using solely RGB imagery is of high interest, as it does not rely on sophisticated sensors, does not require extensive calibration and preprocessing and, therefore, enables the application by a wide audience. In combination with convolutional neural networks (CNNs), which particularly exploit spatial patterns and, therefore, highly benfit from very high-resolution remote sensing, this offers great potential for accurately mapping forest tree species.Here, we present the findings of our recent study, in which we used very high-resolution RGB imagery from UAVs in combination with CNNs for the mapping of forest tree species. In this study, we used multicopter UAVs to obtain very high-resolution (In this contribution, we will give an outlook on how the combination of UAV imagery and CNNs can be integrated with multitemporal satellite imagery (Sentinel-1 and Sentinel-2) in order to extrapolate the UAV-based tree species maps to larger areas.

With the advent of Sentinel-2, it is now possible to generate large-scale chlorophyll content maps with unprecedented spatial and temporal resolution, suitable for monitoring ecological processes such as vegetative stress and/or decline. However methodological gaps exist for adapting this technology to heterogeneous natural vegetation and for transferring it among vegetation species or plan functional types. In this study, we investigated the use of Sentinel-2A imagery for estimating needle chlorophyll (Ca+b) in a sparse pine forest undergoing significant needle loss and tree mortality. Sentinel-2A scenes were acquired under two extreme viewing geometries (June vs. December 2016) coincident with the acquisition of high-spatial resolution hyperspectral imagery, and field measurements of needle chlorophyll content and crown leaf area index. Using the high-resolution hyperspectral scenes acquired over 61 validation sites we found the CI chlorophyll index R750/R710 and Macc index (which uses spectral bands centered at 680 nm, 710 nm and 780 nm) had the strongest relationship with needle chlorophyll content from individual tree crowns (r2 = 0.61 and r2 = 0.59, respectively; p 0.7 for June and >0.4 for December; p

The Tibetan Plateau is a unique, biodiverse ecosystem with an important role in the climate and hydrological system of Asia. Its vegetation supports important functions including fodder provision, erosion prevention and water retention. Assessing vegetation trends of the Tibetan Plateau is crucial to understand effects of recent climate and land-use changes. Most existing vegetation trend products covering the entire Tibetan Plateau have a coarse spatial grain and cover short temporal ranges. This hampers their applicability in studies conducted at local scales where land-use decisions take place and at time scales where climate changes become apparent. Here, we present vegetation trend products for the entire Tibetan Plateau at a spatial resolution of 30 m for the time period 1990–2018. These products include results of a modified Mann-Kendall trend test applied to annual Landsat-based NDVI mosaics, composed from all satellite observations acquired during the vegetation periods as well as NDVI difference images. These data can be valuable to many researchers including for example wildlife ecologists, rangeland experts and climate change researchers., Design Type(s)modeling and simulation objective • data transformation objective • time series designMeasurement Type(s)vegetation layerTechnology Type(s)computational modeling techniqueFactor Type(s)geographic location • temporal_intervalSample Characteristic(s)Tibetan Plateau • plateau Machine-accessible metadata file describing the reported data (ISA-Tab format)

Spaceborne sensors allow for wide-scale assessments of forest ecosystems. Combining the products of multiple sensors is hypothesized to improve the estimation of forest biomass. We applied interferometric (Tandem-X) and photogrammetric (WorldView-2) based predictors, e.g. canopy height models, in combination with hyperspectral predictors (EO1-Hyperion) by using 4 different machine learning algorithms for biomass estimation in temperate forest stands near Karlsruhe, Germany. An iterative model selection procedure was used to identify the optimal combination of predictors. The most accurate model (Random Forest) reached a r 2 of 0.73 with a RMSE of 14.9% (29.4 t/ha). Further results revealed that the predictive accuracy depended highly on the statistical model and the area size of the field samples. We conclude that a fusion of canopy height and spectral information allows for accurate estimations of forest biomass from space.

We used spectral, textural and photogrammetric information from very-high resolution (VHR) stereo satellite data (Pleiades and WorldView-2) to estimate forest biomass across two test sites located in Chile and Germany. We compared Random Forest model performances of different predictor sets (spectral, textural, and photogrammetric), forest inventory designs and filter sizes (texture information). Best model performances were obtained with photogrammetric combined with either textural or spectral information and smaller, but more field plots. Stereo-VHR images showed a great potential for canopy height model (CHM) generation and could be an adequate alternative to LiDAR and InSAR techniques.

Land cover is of fundamental importance to many environmental applications and serves as critical baseline information for many large scale models e.g. in developing future scenarios of land use and climate change. Although there is an ongoing movement towards the development of higher resolution global land cover maps, medium resolution land cover products (e.g. GLC2000 and MODIS) are still very useful for modelling and assessment purposes. However, the current land cover products are not accurate enough for many applications so we need to develop approaches that can take existing land covers maps and produce a better overall product in a hybrid approach. This paper uses geographically weighted regression (GWR) and crowdsourced validation data from Geo-Wiki to create two hybrid global land cover maps that use medium resolution land cover products as an input. Two different methods were used: (a) the GWR was used to determine the best land cover product at each location; (b) the GWR was only used to determine the best land cover at those locations where all three land cover maps disagree, using the agreement of the land cover maps to determine land cover at the other cells. The results show that the hybrid land cover map developed using the first method resulted in a lower overall disagreement than the individual global land cover maps. The hybrid map produced by the second method was also better when compared to the GLC2000 and GlobCover but worse or similar in performance to the MODIS land cover product depending upon the metrics considered. The reason for this may be due to the use of the GLC2000 in the development of GlobCover, which may have resulted in areas where both maps agree with one another but not with MODIS, and where MODIS may in fact better represent land cover in those situations. These results serve to demonstrate that spatial analysis methods can be used to improve medium resolution global land cover information with existing products.

Photosynthesis is essential for life on earth as it, inter alia, influences the composition of the atmosphere and is the driving mechanism of primary production. Photosynthesis is particularly controlled by leaf pigments such as chlorophyll, carotenoids or anthocyanins. Incoming solar radiation is mainly captured by chlorophyll, whereas plant organs are also protected from excess radiation by carotenoids and anthocyanins. Current and upcoming optical earth observation sensors are sensitive to these radiative processes and thus feature a high potential for mapping the spatial and temporal variation of these photosynthetic pigments. In the context of remote sensing, leaf pigments are either quantified as leaf area-based content [μg/cm2] or as leaf mass-based concentration [g/g or %]. However, these two metrics are fundamentally different, and until now there has been neither an in-depth discussion nor a consensus on which metric to choose. This is notable considering the amount of studies that do not explicitly differentiate between pigment content and concentration. We therefore seek to outline the differences between both metrics and thus show that the remote sensing of leaf pigment concentration [%] is unsubstantial. This is due to the fact that, firstly, pigment concentration is likely to primarily reflect variation in leaf mass per area and not pigments itself. Second, the radiative transfer in plant leaves is especially determined by the absolute content of pigments in a leaf and not its relative concentration to other leaf constituents. And third, as a ratio, pigment concentration is an ambiguous metric, which further complicates the quantification of leaf pigments at the canopy scale. Given these issues related to the use of chlorophyll concentration, we thus conclude that remote sensing of leaf pigments should be primarily performed on an area basis [μg/cm2]., The project was funded by the German Aerospace Center (DLR) on behalf of the Federal Ministry of Economics and Technology (BMWi), FKZ50EE 1347.

For reasons of documentation, management and certification there is a high interest in efficient inventories of palm plantations on the single plant level. Recent developments in unmanned aerial vehicle (UAV) technology facilitate spatial and temporal flexible acquisition of high resolution 3D data. Common single tree detection approaches are based on Very High Resolution (VHR) satellite or Airborne Laser Scanning (ALS) data. However, VHR data is often limited to clouds and does commonly not allow for height measurements. VHR and in particualar ALS data are characterized by high relatively high acquisition costs. Sperlich et al. (2013) already demonstrated the high potential of UAV-based photogrammetric point clouds for single tree detection using pouring algorithms. This approach was adjusted and improved for an application on palm plantation. The 9.4ha test site on Tarawa, Kiribati, comprised densely scattered growing palms, as well as abundant undergrowth and trees. Using a standard consumer grade camera mounted on an octocopter two flight campaigns at 70m and 100m altitude were performed to evaluate the effect Ground Sampling Distance (GSD) and image overlap. To avoid comission errors and improve the terrain interpolation the point clouds were classified based on the geometric characteristics of the classes, i.e. (1) palm, (2) other vegetation (3) and ground. The mapping accuracy amounts for 86.1 % for the entire study area and 98.2 % for dense growing palm stands. We conclude that this flexible and automatic approach has high capabilities for operational use.

In both ecology and forestry, there is a high demand for structural information of forest stands. Forest structures, due to their heterogeneity and density, are often difficult to assess. Hence, a variety of technologies are being applied to account for this "difficult to come by" information. Common techniques are aerial images or ground- and airborne-Lidar. In the present study we evaluate the potential use of unmanned aerial vehicles (UAVs) as a platform for tree stem detection in open stands. A flight campaign over a test site near Freiburg, Germany covering a target area of 120 × 75 [m2] was conducted. The dominant tree species of the site is oak (quercus robur) with almost no understory growth. Over 1000 images with a tilt angle of 45° were shot. The flight pattern applied consisted of two antipodal staggered flight routes at a height of 55 [m] above the ground. We used a Panasonic G3 consumer camera equipped with a 14–42 [mm] standard lens and a 16.6 megapixel sensor. The data collection took place in leaf-off state in April 2013. The area was prepared with artificial ground control points for transformation of the structure-from-motion (SFM) point cloud into real world coordinates. After processing, the results were compared with a terrestrial laser scanner (TLS) point cloud of the same area. In the 0.9 [ha] test area, 102 individual trees above 7 [cm] diameter at breast height were located on in the TLS-cloud. We chose the software CMVS/PMVS-2 since its algorithms are developed with focus on dense reconstruction. The processing chain for the UAV-acquired images consists of six steps: a. cleaning the data: removing of blurry, under- or over exposed and off-site images; b. applying the SIFT operator [Lowe, 2004]; c. image matching; d. bundle adjustment; e. clustering; and f. dense reconstruction. In total, 73 stems were considered as reconstructed and located within one meter of the reference trees. In general stems were far less accurate and complete as in the TLS-point cloud. Only few stems were considered to be fully reconstructed. From the comparison of reconstruction achievement with respect to height above ground, we can state that reconstruction accuracy decreased in the crown layer of the stand. In addition we were cutting 50 [cm] slices in z-direction and applied a robust cylinder fit to the stem slices. Radii of the TLS-cloud and the SFM-cloud surprisingly correlated well with a Pearson's correlation coefficient of r = 0.696. This first study showed promising results for UAV-based forest structure modelling. Yet, there is a demand for additional research with regard to vegetation stages, flight pattern, processing setup and the utilisation of spectral information.