Is deeper always better? Evaluating deep learning models for yield forecasting with small data.

Predicting crop yields, and especially anomalously low yields, is of special importance for food insecure countries. In this study, we investigate a flexible deep learning approach to forecast crop yield at the provincial administrative level based on deep 1D and 2D convolutional neural networks using limited data. This approach meets the operational requirements-public and global records of satellite data in an application ready format with near real time updates-and can be transferred to any country with reliable yield statistics. Three-dimensional histograms of normalized difference vegetation index (NDVI) and climate data are used as input to the 2D model, while simple administrative-level time series averages of NDVI and climate data to the 1D model. The best model architecture is automatically identified during efficient and extensive hyperparameter optimization. To demonstrate the relevance of this approach, we hindcast (2002-2018) the yields of Algeria's three main crops (barley, durum and soft wheat) and contrast the model's performance with machine learning algorithms and conventional benchmark models used in a previous study. Simple benchmarks such as peak NDVI remained challenging to outperform while machine learning models were superior to deep learning models for all forecasting months and all tested crops. We attribute the poor performance of deep learning to the small size of the dataset available.

High temporal resolution of leaf area data improves empirical estimation of grain yield.

Empirical yield estimation from satellite data has long lacked suitable combinations of spatial and temporal resolutions. Consequently, the selection of metrics, i.e., temporal descriptors that predict grain yield, has likely been driven by practicality and data availability rather than by systematic targetting of critically sensitive periods as suggested by knowledge of crop physiology. The current trend towards hyper-temporal data raises two questions: How does temporality affect the accuracy of empirical models? Which metrics achieve optimal performance? We followed an in silico approach based on crop modelling which can generate any observation frequency, explore a range of growing conditions and reduce the cost of measuring yields in situ. We simulated wheat crops across Australia and regressed six types of metrics derived from the resulting time series of Leaf Area Index (LAI) against wheat yields. Empirical models using advanced LAI metrics achieved national relevance and, contrary to simple metrics, did not benefit from the addition of weather information. This suggests that they already integrate most climatic effects on yield. Simple metrics remained the best choice when LAI data are sparse. As we progress into a data-rich era, our results support a shift towards metrics that truly harness the temporal dimension of LAI data.

Social capital and transaction costs in millet markets.

In sub-Saharan Africa, transaction costs are believed to be the most significant barrier that prevents smallholders and farmers from gaining access to markets and productive assets. In this study, we explore the impact of social capital on millet prices for three contrasted years in Senegal. Social capital is approximated using a unique data set on mobile phone communications between 9 million people allowing to simulate the business network between economic agents. Our approach is a spatial equilibrium model that integrates a diversified set of data. Local supply and demand were respectively derived from remotely sensed imagery and population density maps. The road network was used to establish market catchment areas, and transportation costs were derived from distances between markets. Results demonstrate that accounting for the social capital in the transaction costs explained 1-9% of the price variance depending on the year. The year-specific effect remains challenging to assess but could be related to a strengthening of risk aversion following a poor harvest.

A multi-disciplinary perspective on emergent and future innovations in peer review.

Peer review of research articles is a core part of our scholarly communication system. In spite of its importance, the status and purpose of peer review is often contested. What is its role in our modern digital research and communications infrastructure? Does it perform to the high standards with which it is generally regarded? Studies of peer review have shown that it is prone to bias and abuse in numerous dimensions, frequently unreliable, and can fail to detect even fraudulent research. With the advent of web technologies, we are now witnessing a phase of innovation and experimentation in our approaches to peer review. These developments prompted us to examine emerging models of peer review from a range of disciplines and venues, and to ask how they might address some of the issues with our current systems of peer review. We examine the functionality of a range of social Web platforms, and compare these with the traits underlying a viable peer review system: quality control, quantified performance metrics as engagement incentives, and certification and reputation. Ideally, any new systems will demonstrate that they out-perform and reduce the biases of existing models as much as possible. We conclude that there is considerable scope for new peer review initiatives to be developed, each with their own potential issues and advantages. We also propose a novel hybrid platform model that could, at least partially, resolve many of the socio-technical issues associated with peer review, and potentially disrupt the entire scholarly communication system. Success for any such development relies on reaching a critical threshold of research community engagement with both the process and the platform, and therefore cannot be achieved without a significant change of incentives in research environments.

A global reference database of crowdsourced cropland data collected using the Geo-Wiki platform.

A global reference data set on cropland was collected through a crowdsourcing campaign using the Geo-Wiki crowdsourcing tool. The campaign lasted three weeks, with over 80 participants from around the world reviewing almost 36,000 sample units, focussing on cropland identification. For quality assessment purposes, two additional data sets are provided. The first is a control set of 1,793 sample locations validated by students trained in satellite image interpretation. This data set was used to assess the quality of the crowd as the campaign progressed. The second data set contains 60 expert validations for additional evaluation of the quality of the contributions. All data sets are split into two parts: the first part shows all areas classified as cropland and the second part shows cropland average per location and user. After further processing, the data presented here might be suitable to validate and compare medium and high resolution cropland maps generated using remote sensing. These could also be used to train classification algorithms for developing new maps of land cover and cropland extent.

National-scale cropland mapping based on spectral-temporal features and outdated land cover information.

The lack of sufficient ground truth data has always constrained supervised learning, thereby hindering the generation of up-to-date satellite-derived thematic maps. This is all the more true for those applications requiring frequent updates over large areas such as cropland mapping. Therefore, we present a method enabling the automated production of spatially consistent cropland maps at the national scale, based on spectral-temporal features and outdated land cover information. Following an unsupervised approach, this method extracts reliable calibration pixels based on their labels in the outdated map and their spectral signatures. To ensure spatial consistency and coherence in the map, we first propose to generate seamless input images by normalizing the time series and deriving spectral-temporal features that target salient cropland characteristics. Second, we reduce the spatial variability of the class signatures by stratifying the country and by classifying each stratum independently. Finally, we remove speckle with a weighted majority filter accounting for per-pixel classification confidence. Capitalizing on a wall-to-wall validation data set, the method was tested in South Africa using a 16-year old land cover map and multi-sensor Landsat time series. The overall accuracy of the resulting cropland map reached 92%. A spatially explicit validation revealed large variations across the country and suggests that intensive grain-growing areas were better characterized than smallholder farming systems. Informative features in the classification process vary from one stratum to another but features targeting the minimum of vegetation as well as short-wave infrared features were consistently important throughout the country. Overall, the approach showed potential for routinely delivering consistent cropland maps over large areas as required for operational crop monitoring.

The academic, economic and societal impacts of Open Access: an evidence-based review.

Ongoing debates surrounding Open Access to the scholarly literature are multifaceted and complicated by disparate and often polarised viewpoints from engaged stakeholders. At the current stage, Open Access has become such a global issue that it is critical for all involved in scholarly publishing, including policymakers, publishers, research funders, governments, learned societies, librarians, and academic communities, to be well-informed on the history, benefits, and pitfalls of Open Access. In spite of this, there is a general lack of consensus regarding the potential pros and cons of Open Access at multiple levels. This review aims to be a resource for current knowledge on the impacts of Open Access by synthesizing important research in three major areas: academic, economic and societal. While there is clearly much scope for additional research, several key trends are identified, including a broad citation advantage for researchers who publish openly, as well as additional benefits to the non-academic dissemination of their work. The economic impact of Open Access is less well-understood, although it is clear that access to the research literature is key for innovative enterprises, and a range of governmental and non-governmental services. Furthermore, Open Access has the potential to save both publishers and research funders considerable amounts of financial resources, and can provide some economic benefits to traditionally subscription-based journals. The societal impact of Open Access is strong, in particular for advancing citizen science initiatives, and leveling the playing field for researchers in developing countries. Open Access supersedes all potential alternative modes of access to the scholarly literature through enabling unrestricted re-use, and long-term stability independent of financial constraints of traditional publishers that impede knowledge sharing. However, Open Access has the potential to become unsustainable for research communities if high-cost options are allowed to continue to prevail in a widely unregulated scholarly publishing market. Open Access remains only one of the multiple challenges that the scholarly publishing system is currently facing. Yet, it provides one foundation for increasing engagement with researchers regarding ethical standards of publishing and the broader implications of 'Open Research'.

Timely monitoring of Asian Migratory locust habitats in the Amudarya delta, Uzbekistan using time series of satellite remote sensing vegetation index.

The Asian Migratory locust (Locusta migratoria migratoria L.) is a pest that continuously threatens crops in the Amudarya River delta near the Aral Sea in Uzbekistan, Central Asia. Its development coincides with the growing period of its main food plant, a tall reed grass (Phragmites australis), which represents the predominant vegetation in the delta and which cover vast areas of the former Aral Sea, which is desiccating since the 1960s. Current locust survey methods and control practices would tremendously benefit from accurate and timely spatially explicit information on the potential locust habitat distribution. To that aim, satellite observation from the MODIS Terra/Aqua satellites and in-situ observations were combined to monitor potential locust habitats according to their corresponding risk of infestations along the growing season. A Random Forest (RF) algorithm was applied for classifying time series of MODIS enhanced vegetation index (EVI) from 2003 to 2014 at an 8-day interval. Based on an independent ground truth data set, classification accuracies of reeds posing a medium or high risk of locust infestation exceeded 89% on average. For the 12-year period covered in this study, an average of 7504 km2 (28% of the observed area) was flagged as potential locust habitat and 5% represents a permanent high risk of locust infestation. Results are instrumental for predicting potential locust outbreaks and developing well-targeted management plans. The method offers positive perspectives for locust management and treatment of infested sites because it is able to deliver risk maps in near real time, with an accuracy of 80% in April-May which coincides with both locust hatching and the first control surveys. Such maps could help in rapid decision-making regarding control interventions against the initial locust congregations, and thus the efficiency of survey teams and the chemical treatments could be increased, thus potentially reducing environmental pollution while avoiding areas where treatments are most likely to cause environmental degradation.