Observing and documenting shape has fueled biological understanding as the shape of biomolecules, cells, tissues, and organisms arise from the effects of genetics, development, and the environment. The vision of Topological Data Analysis (TDA), that data is shape and shape is data, will be relevant as biology transitions into a data-driven era where meaningful interpretation of large datasets is a limiting factor. We focus first on quantifying the morphology of X-ray CT scans of barley spikes and seeds using topological descriptors based on the Euler Characteristic Transform. We then successfully train a support vector machine to distinguish and classify 28 different varieties of barley based solely on the 3D shape of their grains. This shape characterization will allow us later to link genotype with phenotype, furthering our understanding on how the physical shape is genetically coded in DNA.
Regional crop production estimates are important in both public and private sectors to ensure the adequacy of a food supply and aid policymakers and farmers in managing harvest, storage, import/export, transportation, and anticipate market fluctuations. Food security will be progressively challenged by population growth and climate change. Thus, the prediction of accurate regional crop yield is essential for national food security and the sustainable development of the Indian agriculture sector. In this study, we have selected Punjab, the highest wheat yielding state in India. The district-wise wheat yield data were available for the year 2000 – 2019. We have used several covariates for crop health viz. normalized difference vegetation index (NDVI), leaf area index (LAI), fraction of absorbed photosynthetically active radiation (fAPAR); meteorological indicators viz. land surface temperature (LST), and evapotranspiration (ET); and surface characteristics viz. protrusion coefficient (PC). These indicators were generated at 250 m spatial resolution from the MODIS data using Google Earth Engine. The whole data was divided into two groups for training (2000 – 2009, 2011, 2013, 2014, 2016 - 2019) and testing (2010, 2012, 2015), which were randomly selected. This study uses the random forest (RF) regression method to create a wheat yield prediction model. We created several combinations of covariates and found that fAPAR and ET are highly correlated with NDVI and do not have much influence on the model’s prediction accuracy. Hence, only four out of six covariates were selected for final training. The coefficient of determination between district-level yield vs. (NDVI/LAI/PC/LST) was 0.37/0.31/0.15/0.13 respectively. We used randomized search cross-validation as well as grid search cross-validation for hyper-parameter tuning. Furthermore, we used mean absolute error (MAE) and accuracy as quality metrics. The MAE for training was 0.1870 t/Ha with 95.81% accuracy, whereas the MAE on test data was obtained as 0.4293 t/Ha with 90.02% accuracy. The results of this study are within acceptable error limits of the published research articles. Overall, this study demonstrates that covariates derived from coarse resolution satellite data can predict district-level crop yield with reasonable accuracy.
Ensuring the stability of food production is essential for adapting food systems to rising climate variability. Food production instability is determined by changes in yield, planted area, and the ratio between planted and harvested area. Yet most research has focused on evaluating and improving the response of crop yields to climate fluctuations, and there remains a poor understanding of whether and to what extent changes in planted and harvested areas affect crop production stability. Here we use the example of corn in the United States, the country’s most widely produced crop, to evaluate the relative importance of disruptions in yield, harvested area, and the planted: harvested area ratio in contributing to production instability. We apply a new time-series shock detection approach to data covering 2511 counties from the years 1970 to 2020. We find that disruptions in yield, ratio, and planted area explain 34%, 32%, and 20% of total production instabilities, respectively. Considering multiple shocks could happen simultaneously, 48% of the production fluctuations coincided with area (either ratio or planted area) instabilities. In terms of shocks (negative disruptions), the proportion of production shocks occurring concurrently with area shocks rises to 54%, and with yield shocks rise to 45%. The greater impact on production shocks confirms the risk of area shocks to production fluctuation and food security. Based on correlation analysis between the county level ratio shocks and the frequencies of six natural disasters, we show that ratio shocks are significantly correlated with the occurrence of flood, drought, and hail (P<0.001). These findings suggest that fluctuations in planted and harvested area may determine production instability more frequently than yield and that decisions about cropping patterns can thus play a crucial role in stabilizing food production in the face of climate variability.
ABSTRACT Counting maize tassels in field conditions is predominantly done manually. Recently, computer-vision based methods have been utilized to detect tassels from images captured by UAV transects or poled-mounted cameras , , . Once tassels are detected, deep-learning based local regression methods, Tasselnet, have been used to estimate in-field tassel counts . However, field images are mostly captured over a period of time. Consequently, the input images in the foregoing Tasselnet technique are not independent but often form unequal sequences of correlated images. As such, the temporal sequence of images offers information about the growth trajectory of the plants. We propose a hybrid model that (a) utilizes convolutional neural network-based tassel localization in images, and (b) drives the local count of tassels utilizing the plant growth trajectory learned from the time-series of images. The resulting model can also handle important auxiliary information, obtained from in-field sensors (for example: soil moisture, air temperature etc.), that impacts plant growth and tassel counts. We implement our methodology on benchmark dataset  and compare our results with the SOTA Tasselnet . Our initial results suggest that our technique is computationally viable and can produce accurate point estimates of tassel counts along with interval estimates capturing the precision of our estimates. Keywords: Computer vision, Convolutional neural networks, Deep learning, Maize tassels, Time-series. REFERENCES  Shi, Y., Alzadjali, A., Alali, M., Veeranampalayam-Sivakumar, A. N., Deogun, J., Scott, S., & Schnable, J. (2021). Maize tassel detection from UAV imagery using deep learning. Dryad. https://doi.org/10.5061/dryad.r2280gbcg.  Mirnezami, S. V., Srinivasan, S., Zhou, Y., Schnable, P. S., & Ganapathysubramanian, B. (2021). Detection of the Progression of Anthesis in Field-Grown Maize Tassels: A Case Study. Plant Phenomics, 2021, 4238701. doi:10.34133/2021/4238701.  Shete, S., Srinivasan, S., & Gonsalves, T. A. (2020). TasselGAN: An Application of the Generative Adversarial Model for Creating Field-Based Maize Tassel Data. Plant Phenomics, 2020, 8309605. doi:10.34133/2020/8309605.  Lu, H., Cao, Z., Xiao, Y., Zhuang, B., & Shen, C. (2017). TasselNet: counting maize tassels in the wild via local counts regression network. Plant Methods, 13(1), 79. doi:10.1186/s13007-017-0224-0.  Xiong, H., Cao, Z., Lu, H., Madec, S., Liu, L., & Shen, C. (2019). TasselNetv2: in-field counting of wheat spikes with context-augmented local regression networks. Plant Methods, 15(1), 150. doi:10.1186/s13007-019-0537-2.
Understanding onset of droughts and its potential linkage to resulting responses like severity (deficit volume) is crucial for providing timely information related to drought sectors including the cultivation planning and monitoring crop productivity. Using high-quality daily observed streamflow records from 82 medium-to-large sized catchments over (tropical) peninsular India, we show that the variability in onset timing drives the severity of hydrological droughts. The strength of onset timing-severity relationships using observed records indicate seasonality of rainfall and catchment characteristics mainly modulate hydrological drought responses in peninsular India, which is not readily apparent from land-surface model simulations. The observed trend for mean onset of drought depicts delayed occurrence for more than half of the catchments. Around one-third of the catchments shows a stronger non-linear significant dependency (>0.7) between severity and onset of drought. The findings of the study highlight the need for accounting feedback between drought onset and severity and their concurrent changes for seasonal-to-sub-seasonal predictability of droughts; and contributes to discussions on building resilience to extreme droughts in a changing climate.
High crop yield variation between years, impacted for example by extreme weather shocks and by other shocks on the food production system, can have substantial effect on food production. This, in turn introduces vulnerabilities within global food system. To mitigate the effects of these shocks there is a clear need for understanding how different adaptive capacity measures link to the crop yield variability. While existing literature provides many local scale studies on this linkage, no comprehensive global assessment yet exists. We assessed reported crop yield variation for wheat, maize, soybean and rice for time period 1981-2009 by measuring both yield loss risk (variation in negative yield anomalies considering all years) and changes in yields during only dry shock and hot shock years. We used machine learning algorithm XGBoost to assess globally the explanatory power of selected gridded anthropogenic indicators (i.e., adaptive capacity measures; such as Human Development Index, irrigation infrastructure, fertilizer use) on yield variation on 0.5 degree resolution, within climatically similar regions to rule out the role of average climate conditions. We found that the anthropogenic indicators explained 40-60% of yield loss risk variation whereas the indicators provided noticeably lower (5-20%) explanatory power during shock years. On continental scale, especially in Europe and Africa the indicators explained high proportion of the yield loss risk variation (up to around 80%). Assessing crop production vulnerabilities on global scale provides supporting knowledge to target specific adaptation measures, thus contributing to global food security.
Managing landscapes to increase agricultural productivity and environmental stewardship requires spatially distributed models that can integrate data and operate at spatial and temporal scales that are intervention-relevant. This paper presents Cycles-L, a landscape-scale, coupled agroecosystem hydrologic modeling system. Cycles-L couples a 3-D land surface hydrologic model, Flux-PIHM, with a 1-D agroecosystem model, Cycles. Cycles-L takes the landscape and hydrology structure from Flux-PIHM and most agroecosystem processes from Cycles. Consequently, Cycles-L can simulate landscape level processes affected by topography, soil heterogeneity, and management practices, owing to its physically-based hydrologic component and ability to simulate horizontal and vertical transport of mineral nitrogen (N) with water. The model was tested at a 730-ha agricultural experimental watershed within the Mahantango Creek watershed in Pennsylvania. Cycles-L simulated well stream water discharge and N exports (Nash-Sutcliffe coefficient 0.55 and 0.58, respectively), and grain crop yield (root mean square error 1.01 Mg ha−1), despite some uncertainty in the accuracy of survey-based input data. Cycles-L outputs are as good if not better than those obtained with the uncoupled Flux-PIHM (water discharge) and Cycles (crop yield) models. Model predicted spatial patterns of N fluxes clearly show the combined control of crop management and topography. Cycles-L spatial and temporal resolution fills a gap in the availability of analytical models at an operational scale relevant to evaluate costly strategic and tactical interventions in silico, and can become a core component of tools for applications in precision agriculture, precision conservation, and artificial intelligence-based decision support systems.
An open source computer algorithm, the Surface Energy Balance Algorithm for Land-Improved (SEBALI), was designed to estimate actual evapotranspiration (ET) at a basin level. In this study, we build on later versions of SEBALI/SEBALIGEE to estimate ET at a 30-m resolution for any scale application using advanced machine learning approaches (SEBALIGEE v2). We evaluate the monthly ET estimated from the new algorithm across several fluxnet sites in US, China, Italy, Belgium, Germany, and France, yielding an Absolute Mean Error (AME) of 0.41 mm/day versus 0.48 mm/day in the original SEBALIGEE. Analyses of the ET in the US indicate that the annual wheat ET decreases significantly between 2013 and 2021 (p < 0.05), accompanied by a significant air temperature increase. Net solar radiation is found to be the most influencing factor on ET of corn and soybeans with R2 values of ~0.72.
This presentation talks about the effect of portable ground control points (GCPs) on the accuracy of unmanned aerial vehicle (UAV)-based remote sensing data in predicting plant health. 6 GCPs equipped with GPS receivers were spaced around the experimental plots of citrus and strawberry. UAVs equipped with multispectral sensors were then used to collect the remote sensing data of citrus and strawberry plants. The remote sensing data was used to calculate various vegetation indices including normalized difference vegetation index (NDVI), Green NDVI, and soil adjusted vegetation indices (SAVI). These indices were compared with the data obtained from proximal sensors that include Handheld Spectroradiometer and Chlorophyll Meter. Correlation between various vegetation indices, chlorophyll content, and spectroradiometer data will be shown and discussed. A significantly higher correlation coefficients were obtained between the remote sensing and proximal sensor data when the GCPs were used. Increased accuracy of remote sensing data is important for the widespread adoption of UAV-based remote sensing technology for precision agriculture so that the technology can be used for site-specific input management by taking into account the infield variability.
This study was conducted to assess effect of tillage and mulch on soil erosion control in typical agroecological conditions of Benin. In addition, it involved also the assessment of soil moisture and runoff. The experiment was conducted on two sites in Central Benin during the short rain season of 2018. The effect of three tillage practices (contour ridging: CR; slope ridging: SR and no-tillage: NT) and three mulch doses (0 t.ha-1; 3 t.ha-1; and 7 t.ha-1) on soil erosion under maize was investigated at small experimental plots (21 m2). The 7Be method was used to assess the erosion rates, runoff was measured by total collection and soil moisture content was determined by thermo-gravimetric method. The results showed a signiﬁcant decrease in runoﬀ coeﬃcient and soil loss while increase soil moisture under no-tillage and contour ridges compared to slope ridges. This effect was pronounced with greatest. 3 and 7 tha-1. Highest runoff coefficient and soil loss and the lowest soil moisture were observed under slope ridging without mulch (i.e. SR0M). The 7Be measurement showed high soil losses under SR0M (-10.19 t ha-1) at Dan and under NT0M (-7.36 t ha-1) at Za-zounmè. The treatments NT7M (0.80 t ha-1); SR7M (0.69 t ha-1); IR3M (2.07 t ha-1) and CR7M (4.05 t ha-1) showed deposition at Dan while SR7M (0.23 t ha-1) and CR7M (3.93 t ha-1) showed deposition at Za-zounmè. This study revealed useful information to be taken into consideration when developing soil and water conservation management strategies in Benin.
Underground monitoring of root morphology and their interactions with the environment is critical to understand the overall performance of a plant. Such understanding allows plant breeders to develop plants that are resilient to the adverse effects of climate change and potentially even improve yields for food, fuel, and fiber. We propose and experimentally demonstrate the use of fiber Bragg grating-based fiber optic sensors as a non-destructive technology to measure width and depth of a root-like object and monitor the change in groundwater level as an indicator for soil-water content. Low-cost and continuous remote monitoring analyzed the spectral shift induced optical power change in the fiber optic sensors.
We present several methods for improving plant reconstruction from multiple 3D observations. Producing 3D data useful for plant phenotyping requires proximal sensing (e.g. line scanner, depth camera) at multiple incident angles (φ) and often with multiple passes. These resulting individual point clouds must then be assembled into a single point cloud for analysis. Our interest in improving the registration of individual plants is focused specifically on observations made within field settings which present additional challenges over laboratory 3D scans, where background, overlap and light conditions can be controlled. To develop these methods, we use several season’s worth of data from the University of Arizona’s Field Scanalyzer located in Maricopa, Arizona. Our approach prioritizes: (1) plant completeness, (2) noise reduction, (3) temporal similarity and (4) computational efficiency. The first priority is accomplished simply by prioritizing individual point clouds that contain the majority of the individual plant. 3D field scanning can result in component point clouds that are from near-identical φ and cover the same portions of the individual plant. This results in both additional noise and uncertainties due to small georeferencing errors and plant movement between scans. Thus, we remove the data that is furthest in time with non-unique φ in order to achieve priorities 2 and 3. Our method results in small scene reconstruction which has low memory and computational demands. In order to improve registration further, we investigate iterative closest point (ICP) registration fitting using weights defined by crop height distributions and semantic segmentation point labeling.
Previous crop yield improvements have been largely due to the implementation of new management strategies, mechanization, and application of emerging technologies. While these approaches have led to stable, linear improvements, increases in crop yields are currently plateauing. The use and improvement of rapid, automated, and accurate phenomic selection methods leveraging high-resolution data collected throughout a growing season could help identify stress-adaptive traits to meet the growing global food demand. As the capacity of phenomics to generate larger and higher dimensional data sets improves, there is an urgent need to develop and implement robust and scalable data processing pipelines for rapid turnaround of processed results. Current phenomics processing pipelines lack modularity and the ability to exploit the distributed computational infrastructure required for machine learning (ML)-based workloads. To address these challenges, we developed PhytoOracle (PO), a suite of modular, scalable pipelines that aim to improve data processing efficiency for plant science research. PO integrates open-source frameworks for distributed task management on local, cloud, or high-performance computing (HPC) systems. Each pipeline component is available as a standalone container which can be independently deployed or linked into a pipeline. Additionally, researchers can swap between available containers or integrate new ones suited to their specific research. PO extracts phenotype trait values such as volume, height, canopy temperature, and maximum quantum efficiency (F v /F m) of photosystem II from data captured in field settings, enabling the study of phenotypic variation for elucidation of the genetic components of quantitative traits.
Wildfire smoke frequently blankets the U.S. throughout the agricultural growing season, and this will likely increase with climate change. Studies of smoke impacts have largely focused on air quality and human health; however, understanding smoke’s impact on photosynthetically active radiation (PAR) is essential for predicting how smoke affects plant growth. We compare surface shortwave irradiance and diffuse fraction (DF) on smoke-impacted and smoke-free days from 2006-2020 using data from multifilter rotating shadowband radiometers at ten U.S. Department of Agriculture (USDA) UV-B Monitoring and Research Program stations and smoke plume locations from operational satellite products. On average, 20% of growing season days are smoke-impacted, but smoke prevalence increases over time (r = 0.60, p < 0.05). Smoke presence peaks in the mid- to late growing season (i.e., July, August), particularly over the northern Rocky Mountains, Great Plains, and Midwest. We find an increase in the distribution of PAR DF on smoke-impacted days, with larger increases at lower cloud fractions. On clear-sky days, daily average PAR DF increases by 10 percentage points when smoke is present. Spectral analysis of clear-sky days shows smoke increases DF (average: +45%) and decreases total irradiance (average: -6%) across all six wavelengths measured from 368-870 nm. Optical depth measurements from ground and satellite observations both indicate that spectral DF increases and total spectral irradiance decreases with increasing smoke plume optical depth (i.e., plume thickness). Our analysis provides a foundation for understanding smoke’s impact on PAR, which carries implications for agricultural crop productivity under a changing climate.
Soil property and litter quality are two key factors that control soil organic matter decomposition. Under climate change, it remains unclear how the changes of soil microbial community and litter quality affect soil organic carbon decomposition, although significant changes of these two factors have been reported intensively. This limits our ability to model the dynamics of terrestrial soil carbon in a changing climate. Using a long-term Free Air CO2 Enrichment facility equipped with warming, we investigated the effect of soil property and litter quality change on the decomposition rate of soil organic matter. Results showed that significant change of litter quality was observed under elevated CO2 and warming. Elevated CO2 decreased the concentration of N of rice and wheat straw, while warming decreased the concentration of N and K in wheat straw. However, these changes in plant litter quality did not lead to a shift in soil organic matter decomposition. The legacy effect of long-term elevated CO2 and warming on soil properties dominated the decomposition rate of soil organic matter. Elevated CO2 suppressed soil organic matter decomposition mainly by increasing phosphorous availability and lowering soil C/N, fungi/bacteria ratio, and N-acetyl-glucosaminidase activity; while warming or elevated CO2 plus warming had no effect on soil organic matter decomposition. Our results demonstrated that the change of soil properties other than litter quality control the decomposition of soil organic carbon; and soil property change should be taken into consideration in model developing when predicting terrestrial soil carbon dynamics under elevated atmospheric CO2 and warming.
The COVID-19 pandemic has increased the risk of global public health and has the potential to cause severe food and water insecurity due to economic recession during lockdown for people living in low-middle income countries like Bangladesh where capital resources are scarce. There is growing evidence that household food and water insecurity has been associated with poor psychological outcomes. The objective of this study was to determine the association between household food and water insecurity with mental health and whether these differed among urban-rural households. A cross-sectional online survey was conducted with 545 participants immediately after the COVID-19 lockdown period in Bangladesh (August 1-September 30, 2020). Household food and water security were determined using a 9-item Household Food Insecurity Access Scale (HFIAS) (score range 0-27) and a 12-item Household Water Insecurity Experiences (HWISE) scale (score range 0-36), respectively. The Perceived Stress Scale (PSS) was used to evaluate mental health. Multivariable logistic regression examined the association between household food and water insecurity with perceived stress, adjusting socioeconomic characteristics. An urban-rural stratified analysis was also performed. About 72.84% (397) respondents reported high stress and more than 70% of households suffered from food and water insecurity during the lockdown period. After adjusting covariates, logistic regression model results show that food insecurity was associated with a 1.07-point increase in high perceived stress (OR=1.07, 95% CI=1.01-1.11, p<0.01) while water insecurity was associated with 1.03 times greater odds of high perceived stress (OR=1.03, 95% CI=0.93-1.23, p<0.05). In stratified analysis, only food insecurity was associated with high perceived stress in the urban household (OR=1.08, 95% CI=1.00-1.11, p<0.05). However, none of the household insecurity was associated with perceived stress in rural households. Interventions that promote equal access to resources for low-income individuals will likely to be more effective to alleviate economic burden of pandemic.
We present EuropeAgriDB v1.0, a dataset of crop production and nitrogen (N) flows in European cropland 1961–2019. The dataset covers 26 present-day countries, detailing the cropland N harvests in 17 crop categories as well as cropland N inputs in synthetic fertilizers, manure, symbiotic fixation, and atmospheric deposition. The study builds on established methods but goes beyond previous research by combining data from FAOSTAT, Eurostat, and a range of national data sources. A key contribution is the comprehensive and detailed coverage of crop production, in particular fodder crops such as temporary grassland, green maize, and forage legumes. For these crops, we have combined the Eurostat crop production statistics database with a range of national databases, statistical yearbooks, and other sources. For other arable and permanent crops, we use the FAOSTAT database which apart from fodder crops offers the longest and most complete time series of crop production. Our crop production dataset, divided into 17 crop categories, provides a solid basis for understanding how crop mix and productivity have varied over time. A second key contribution is the detailed estimation of synthetic N fertilizer application to cropland and permanent grassland. We have assembled a comprehensive dataset based on a wide range of data sources and devised a rigorous method to process it. The result, we believe, is to date the most comprehensive and consistent estimate of the allocation of synthetic N fertilizer between cropland and permanent grassland in Europe. In summary, EuropeAgriDB v1.0 is a detailed, complete, and consistent dataset which will be useful both to understand Europe’s recent agricultural history and to make informed decisions about its future. This is particularly relevant in the current context of the EU Farm to Fork strategy, which requires major reduction in N inputs and surpluses and therefore the best possible quantification.
With over 700 million km2 Siberia is the largest expanse of the northern boreal forest—deciduous-needleleaf larch. Temperatures are increasing across this region, but the consequences to carbon balances are not well understood for larch forests. We present flux measurements from a larch forest near the southern edge of Central-Siberia where permafrost degradation and ecosystem shifting are already observed. Results indicate net carbon exchanges are influenced by the seasonality of permafrost active layers, temperature and humidity, and soil water availability. During periods when surface soils are fully thawed, larch forest is a significant carbon sink. During the spring-thaw and fall-freeze transition, there is a weak signal of carbon uptake at mid-day. Net carbon exchanges are near-zero when the soil is fully frozen from the surface down to the permafrost. We fit an empirical ecosystem functional model to quantify the dependence of larch-forest carbon balance on climatic drivers. The model provides a basis for ecosystem carbon budgets over time and space. Larch differs from boreal evergreens by having higher maximum productivity and lower respiration, leading to an increased carbon sink. Comparison to previous measurements from another northern larch site suggests climate change will result in an increased forest carbon sink if the southern larch subtype replaces the northern subtype. Observations of carbon fluxes in Siberian larch are still too sparse to adequately determine age dependence, inter-annual variability, and spatial heterogeneity though they suggest that boreal larch accounts for a larger fraction of global carbon uptake than has been previously recognized.