Self-Supervised Learning of Satellite-Derived Vegetation Indices for Clustering and Visualization of Vegetation Types.

Vegetation indices are commonly used techniques for the retrieval of biophysical and chemical attributes of vegetation. This paper presents the potential of an Autoencoders (AEs) and Convolutional Autoencoders (CAEs)-based self-supervised learning approach for the decorrelation and dimensionality reduction of high-dimensional vegetation indices derived from satellite observations. This research was implemented in Mt. Zao and its base in northeast Japan with a cool temperate climate by collecting the ground truth points belonging to 16 vegetation types (including some non-vegetation classes) in 2018. Monthly median composites of 16 vegetation indices were generated by processing all Sentinel-2 scenes available for the study area from 2017 to 2019. The performance of AEs and CAEs-based compressed images for the clustering and visualization of vegetation types was quantitatively assessed by computing the bootstrap resampling-based confidence interval. The AEs and CAEs-based compressed images with three features showed around 4% and 9% improvements in the confidence intervals respectively over the classical method. CAEs using convolutional neural networks showed better feature extraction and dimensionality reduction capacity than the AEs. The class-wise performance analysis also showed the superiority of the CAEs. This research highlights the potential of AEs and CAEs for attaining a fine clustering and visualization of vegetation types.

Vegetation Structure Index (VSI): Retrieving Vegetation Structural Information from Multi-Angular Satellite Remote Sensing.

Utilization of the Bidirectional Reflectance Distribution Function (BRDF) model parameters obtained from the multi-angular remote sensing is one of the approaches for the retrieval of vegetation structural information. In this research, the potential of multi-angular vegetation indices, formulated by the combination of multi-spectral reflectance from different view angles, for the retrieval of forest above-ground biomass was assessed in the New England region. The multi-angular vegetation indices were generated by the simulation of the Moderate Resolution Imaging Spectroradiometer (MODIS) BRDF/Albedo Model Parameters Product (MCD43A1 Version 6)-based BRDF parameters. The effects of the seasonal (spring, summer, autumn, and winter) composites of the multi-angular vegetation indices on the above-ground biomass, the angular relationship of the spectral reflectance with above-ground biomass, and the interrelationships between the multi-angular vegetation indices were analyzed. Among the existing multi-angular vegetation indices, only the Nadir BRDF-adjusted NDVI and Hot-spot incorporated NDVI showed significant relationship (more than 50%) with the above-ground biomass. The Vegetation Structure Index (VSI), newly proposed in the research, performed in the most efficient way and explained 64% variation of the above-ground biomass, suggesting that the right choice of the spectral channel and observation geometry should be considered for improving the estimates of the above-ground biomass. In addition, the right choice of seasonal data (summer) was found to be important for estimating the forest biomass, while other seasonal data were either insensitive or pointless. The promising results shown by the VSI suggest that it could be an appropriate candidate for monitoring vegetation structure from the multi-angular satellite remote sensing.

Response and resilience of Asian agrifood systems to COVID-19: An assessment across twenty-five countries and four regional farming and food systems.

Context: The COVID-19 pandemic has been affecting health and economies across the world, although the nature of direct and indirect effects on Asian agrifood systems and food security has not yet been well understood. Objectives: This paper assesses the initial responses of major farming and food systems to COVID-19 in 25 Asian countries, and considers the implications for resilience, food and nutrition security and recovery policies by the governments. Methods: A conceptual systems model was specified including key pathways linking the direct and indirect effects of COVID-19 to the resilience and performance of the four principal Asian farming and food systems, viz, lowland rice based; irrigated wheat based; hill mixed; and dryland mixed systems. Based on this framework, a systematic survey of 2504 key informants (4% policy makers, 6% researchers or University staff, 6% extension workers, 65% farmers, and 19% others) in 20 Asian countries was conducted and the results assessed and analysed. Results and conclusion: The principal Asian farming and food systems were moderately resilient to COVID-19, reinforced by government policies in many countries that prioritized food availability and affordability. Rural livelihoods and food security were affected primarily because of disruptions to local labour markets (especially for off-farm work), farm produce markets (notably for perishable foods) and input supply chains (i.e., seeds and fertilisers). The overall effects on system performance were most severe in the irrigated wheat based system and least severe in the hill mixed system, associated in the latter case with greater resilience and diversification and less dependence on external inputs and long market chains. Farming and food systems' resilience and sustainability are critical considerations for recovery policies and programmes, especially in relation to economic performance that initially recovered more slowly than productivity, natural resources status and social capital. Overall, the resilience of Asian farming and food systems was strong because of inherent systems characteristics reinforced by public policies that prioritized staple food production and distribution as well as complementary welfare programmes. With the substantial risks to plant- and animal-sourced food supplies from future zoonoses and the institutional vulnerabilities revealed by COVID-19, efforts to improve resilience should be central to recovery programmes. Significance: This study was the first Asia-wide systems assessment of the effects of COVID-19 on agriculture and food systems, differentiating the effects of the pandemic across the four principal regional farming and food systems in the region.

Multi-dimensional evaluation of response to salt stress in wheat.

Soil salinity is a major threat to crop production worldwide. The global climate change is further accelerating the process of soil salinization, particularly in dry areas of the world. Increasing genetic variability of currently used wheat varieties by introgression of exotic alleles/genes from related progenitors' species in breeding programs is an efficient approach to overcome limitations due to the absence of valuable genetic diversity in elite cultivars. Synthetic hexaploid wheat (SHW) is widely regarded as donor of favourable exotic alleles to improve tolerance against biotic and abiotic stresses such as salinity stress. In this study, synthetic backcross lines (SBLs) winter wheat population "Z86", derived from crosses involving synthetic hexaploid wheat Syn86L with German elite winter wheat cultivar Zentos, was evaluated for salinity tolerance at different developmental stages under controlled and field conditions in three growing seasons. High genetic variability was detected across the SBLs and their parents at various growth stages under controlled as well as under salt stress field trials. Greater performance of Zentos over Syn86L was detected at germination stage across all salt treatments and with respect to shoot dry weight (SDW) and root dry weight (RDW) at seedling stage. Whereas for the root length (RL) and the shoot length (SL) Syn86L surpassed the elite cultivar and most of the progenies. Our experiments revealed for almost all traits that some genotypes among the SBLs showed higher performance than their parents. Furthermore, positive transgressive segregations were detected among the SBLs for germination at high salinity levels, as well as for RDW and SDW at seedling stage. Therefore, the studied Z86 population is a suitable population for assessment of salinity stress on morphological and physiological traits at different plant growth stages. The identified SBLs provide a valuable source for genetic gain through recombination of superior alleles that can be directly applied in breeding programs for efficiently breeding cultivars with improved salinity tolerance and desired agronomic traits.

Genetic and transcriptional variations in NRAMP-2 and OPAQUE1 genes are associated with salt stress response in wheat.

KEY MESSAGE: SNP alleles on chromosomes 4BL and 6AL are associated with sensitivity to salt tolerance in wheat and upon validation can be exploited in the development of salt-tolerant wheat varieties. The dissection of the genetic and molecular components of salt stress response offers strong opportunities toward understanding and improving salt tolerance in crops. In this study, GWAS was employed to identify a total of 106 SNP loci (R2 = 0.12-63.44%) linked to salt stress response in wheat using leaf chlorophyll fluorescence, grain quality and shoot ionic (Na+ and K+ ions) attributes. Among them, 14 SNP loci individually conferred pleiotropic effects on multiple independent salinity tolerance traits including loci at 99.04 cM (R2 ≥ 14.7%) and 68.45 cM (R2 ≥ 4.10%) on chromosomes 6AL and 4BL, respectively, that influenced shoot Na+-uptake, shoot K+/Na+ ratio, and specific energy fluxes for absorption (ABS/RC) and dissipation (DIo/RC). Analysis of the open reading frame (ORF) containing the SNP markers revealed that they are orthologous to genes involved in photosynthesis and plant stress (salt) response. Further transcript abundance and qRT-PCR analyses indicated that the genes are mostly up-regulated in salt-tolerant and down-regulated in salt-sensitive wheat genotypes including NRAMP-2 and OPAQUE1 genes on 4BL and 6AL, respectively. Both genes showed highest differential expression between contrasting genotypes when expressions of all the genes within their genetic intervals were analyzed. Possible cis-acting regulatory elements and coding sequence variation that may be involved in salt stress response were also identified in both genes. This study identified genetic and molecular components of salt stress response that are associated with Na+-uptake, shoot Na+/K+ ratio, ABS/RC, DIo/RC, and grain quality traits and upon functional validation would facilitate the development of gene-specific markers that could be deployed to improve salinity tolerance in wheat.

Allelic variations and differential expressions detected at quantitative trait loci for salt stress tolerance in wheat.

The increasing salinization of agricultural lands is a threat to global wheat production. Understanding of the mechanistic basis of salt tolerance (ST) is essential for developing breeding and selection strategies that would allow for increased wheat production under saline conditions to meet the increasing global demand. We used a set that consists of 150 internationally derived winter and facultative wheat cultivars genotyped with a 90K SNP chip and phenotyped for ST across three growth stages and for ionic (leaf K+ and Na+ contents) traits to dissect the genetic architecture regulating ST in wheat. Genome-wide association mapping revealed 187 Single Nucleotide Polymorphism (SNPs) (R2 = 3.00-30.67%), representing 37 quantitative trait loci (QTL), significantly associated with the ST traits. Of these, four QTL on 1BS, 2AL, 2BS and 3AL were associated with ST across the three growth stages and with the ionic traits. Novel QTL were also detected on 1BS and 1DL. Candidate genes linked to these polymorphisms were uncovered, and expression analyses were performed and validated on them under saline and non-saline conditions using transcriptomics and qRT-PCR data. Expressed sequence comparisons in contrasting ST wheat genotypes identified several non-synonymous/missense mutation sites that are contributory to the ST trait variations, indicating the biological relevance of these polymorphisms that can be exploited in breeding for ST in wheat.

A Machine Learning and Cross-Validation Approach for the Discrimination of Vegetation Physiognomic Types Using Satellite Based Multispectral and Multitemporal Data.

This paper presents the performance and evaluation of a number of machine learning classifiers for the discrimination between the vegetation physiognomic classes using the satellite based time-series of the surface reflectance data. Discrimination of six vegetation physiognomic classes, Evergreen Coniferous Forest, Evergreen Broadleaf Forest, Deciduous Coniferous Forest, Deciduous Broadleaf Forest, Shrubs, and Herbs, was dealt with in the research. Rich-feature data were prepared from time-series of the satellite data for the discrimination and cross-validation of the vegetation physiognomic types using machine learning approach. A set of machine learning experiments comprised of a number of supervised classifiers with different model parameters was conducted to assess how the discrimination of vegetation physiognomic classes varies with classifiers, input features, and ground truth data size. The performance of each experiment was evaluated by using the 10-fold cross-validation method. Experiment using the Random Forests classifier provided highest overall accuracy (0.81) and kappa coefficient (0.78). However, accuracy metrics did not vary much with experiments. Accuracy metrics were found to be very sensitive to input features and size of ground truth data. The results obtained in the research are expected to be useful for improving the vegetation physiognomic mapping in Japan.

Earthquake Damage Visualization (EDV) Technique for the Rapid Detection of Earthquake-Induced Damages Using SAR Data.

The damage of buildings and manmade structures, where most of human activities occur, is the major cause of casualties of from earthquakes. In this paper, an improved technique, Earthquake Damage Visualization (EDV) is presented for the rapid detection of earthquake damage using the Synthetic Aperture Radar (SAR) data. The EDV is based on the pre-seismic and co-seismic coherence change method. The normalized difference between the pre-seismic and co-seismic coherences, and vice versa, are used to calculate the forward (from pre-seismic to co-seismic) and backward (from co-seismic to pre-seismic) change parameters, respectively. The backward change parameter is added to visualize the retrospective changes caused by factors other than the earthquake. The third change-free parameter uses the average values of the pre-seismic and co-seismic coherence maps. These three change parameters were ultimately merged into the EDV as an RGB (Red, Green, and Blue) composite imagery. The EDV could visualize the earthquake damage efficiently using Horizontal transmit and Horizontal receive (HH), and Horizontal transmit and Vertical receive (HV) polarizations data from the Advanced Land Observing Satellite-2 (ALOS-2). Its performance was evaluated in the Kathmandu Valley, which was hit severely by the 2015 Nepal Earthquake. The cross-validation results showed that the EDV is more sensitive to the damaged buildings than the existing method. The EDV could be used for building damage detection in other earthquakes as well.