Search

Enough of silos: develop a joint scientific agenda to understand the intertwined global crises of the Earth system.

ABSTRACTBody size is one of the most important individual traits, determining various other life-history traits, including fitness. Both evolutionary and ecological factors shape the body size in arthropods, but the relative contribution of abiotic drivers acting at different spatial scales has been little investigated. We aimed to identify the importance of two broad-scale variables (study region and elevation) in shaping body size of the free-running and locally abundant wolf spider Pardosa palustris(Linnaeus1758), in contrast to the fine-scaled variable topographic position. Therefore, we set up transects along environmental gradients in the arctic-alpine ecosystems of Norway, which we analyzed using a random forest approach to identify the relative importance of topographic position, elevation, and study region on body size of P. palustris. Our approach revealed that research region was the best explanatory variable, followed by elevation and topographic position. Differences in body size were most likely a consequence of the pronounced differences in season length and the ability of P. palustristo avoid local unfavorable environmental conditions due to its high mobility.

Biodiversity monitoring is an almost inconceivable challenge at the scale of the entire Earth. The current (and soon to be flown) generation of spaceborne and airborne optical sensors (i.e., imaging spectrometers) can collect detailed information at unprecedented spatial, temporal, and spectral resolutions. These new data streams are preceded by a revolution in modeling and analytics that can utilize the richness of these datasets to measure a wide range of plant traits, community composition, and ecosystem functions. At the heart of this framework for monitoring plant biodiversity is the idea of remotely identifying species by making use of the ‘spectral species’ concept. In theory, the spectral species concept can be defined as a species characterized by a unique spectral signature and thus remotely detectable within pixel units of a spectral image. In reality, depending on spatial resolution, pixels may contain several species which renders species‐specific assignment of spectral information more challenging. The aim of this paper is to review the spectral species concept and relate it to underlying ecological principles, while also discussing the complexities, challenges and opportunities to apply this concept given current and future scientific advances in remote sensing. Biodiversity monitoring based on field data is almost inconceivable at the scale of the entire Earth. Over the past decades, remote sensing has opened possibilities for Earth observation from air and space, allowing us to monitor ecological change, primarily expressed by changes in vegetation cover, distribution, and functioning, which can be subsequently linked to drivers of change in space and time, from local to global scale. Recently, the spectral species concept—an algorithm that clusterizes pixels from spectral images having a similar spectral signal (referred to as ‘spectral species’)—has brought attention. The aim of this paper is to review the ecological functioning principles of the spectral species concept and to refine its definition by a better linkage with field observations of plant species distribution data (i.e., presence‐absence data) available from vegetation surveys. Remote sensing has opened possibilities for Earth observation from air and space, allowing us to monitor ecological changeBiodiversity monitoring based on field data is almost inconceivable at the scale of the entire EarthThe spectral species concept, relating field to remotely sensed data, can open new ways to measure diversity from space Remote sensing has opened possibilities for Earth observation from air and space, allowing us to monitor ecological change Biodiversity monitoring based on field data is almost inconceivable at the scale of the entire Earth The spectral species concept, relating field to remotely sensed data, can open new ways to measure diversity from space

Accurate information on the spatial distribution of plant species and communities is in high demand for various fields of application, such as nature conservation, forestry, and agriculture. A series of studies has shown that Convolutional Neural Networks (CNNs) accurately predict plant species and communities in high-resolution remote sensing data, in particular with data at the centimeter scale acquired with Unoccupied Aerial Vehicles (UAV). However, such tasks often require ample training data, which is commonly generated in the field via geocoded in-situ observations or labeling remote sensing data through visual interpretation. Both approaches are laborious and can present a critical bottleneck for CNN applications. An alternative source of training data is given by using knowledge on the appearance of plants in the form of plant photographs from citizen science projects such as the iNaturalist database. Such crowd-sourced plant photographs typically exhibit very different perspectives and great heterogeneity in various aspects, yet the sheer volume of data could reveal great potential for application to bird’s eye views from remote sensing platforms. Here, we explore the potential of transfer learning from such a crowd-sourced data treasure to the remote sensing context. Therefore, we investigate firstly, if we can use crowd-sourced plant photographs for CNN training and subsequent mapping of plant species in high-resolution remote sensing imagery. Secondly, we test if the predictive performance can be increased by a priori selecting photographs that share a more similar perspective to the remote sensing data. We used two case studies to test our proposed approach with multiple RGB orthoimages acquired from UAV with the target plant species Fallopia japonicaand Portulacaria afrarespectively. Our results demonstrate that CNN models trained with heterogeneous, crowd-sourced plant photographs can indeed predict the target species in UAV orthoimages with surprising accuracy. Filtering the crowd-sourced photographs used for training by acquisition properties increased the predictive performance. This study demonstrates that citizen science data can effectively anticipate a common bottleneck for vegetation assessments and provides an example on how we can effectively harness the ever-increasing availability of crowd-sourced and big data for remote sensing applications.

AbstractDifferentiations in reproductive traits along climatic gradients can be substantial for a species to spread along a wide spatial range. We compared the reproductive effort allocated to first egg sacs of five species of the genus Pardosa: P. palustris(Linnaeus 1758), P. amentata(Clerck 1757), P. lugubris(Walckenaer 1802), P. hyperborea(Thorell 1872), and P. riparia(C. L. Koch 1833) along three elevation transects in central Norway. We tested whether population differences are consistent among the three transects, respectively along the elevational gradient. We assumed that the harsh environments of alpine areas would lead to adaptations in reproductive traits resulting in larger eggs but smaller clutches at higher elevations. The results show that female size and egg number were positively correlated among all species. However, no clear elevation-related trend was found. Other traits did not change consistently between species and along the elevational gradient. We assume that local microclimatic impacts on spider fitness are a crucial but poorly understood factor. Without further knowledge about adaptation and phenotypic plasticity in ectotherms, modeling of possible future reproduction biology might remain flawed.

Deep learning and particularly Convolutional Neural Networks (CNN) in concert with remote sensing are becoming standard analytical tools in the geosciences. A series of studies has presented the seemingly outstanding performance of CNN for predictive modelling. However, the predictive performance of such models is commonly estimated using random cross-validation, which does not account for spatial autocorrelation between training and validation data. Independent of the analytical method, such spatial dependence will inevitably inflate the estimated model performance. This problem is ignored in most CNN-related studies and suggests a flaw in their validation procedure. Here, we demonstrate how neglecting spatial autocorrelation during cross-validation leads to an optimistic model performance assessment, using the example of a tree species segmentation problem in multiple, spatially distributed drone image acquisitions. We evaluated CNN-based predictions with test data sampled from 1) randomly sampled hold-outs and 2) spatially blocked hold-outs. Assuming that a block cross-validation provides a realistic model performance, a validation with randomly sampled holdouts overestimated the model performance by up to 28%. Smaller training sample size increased this optimism. Spatial autocorrelation among observations was significantly higher within than between different remote sensing acquisitions. Thus, model performance should be tested with spatial cross-validation strategies and multiple independent remote sensing acquisitions. Otherwise, the estimated performance of any geospatial deep learning method is likely to be overestimated.