Above-Ground Biomass Estimation in Oats Using UAV Remote Sensing and Machine Learning

Current strategies for phenotyping above-ground biomass in field breeding nurseries demand significant investment in both time and labor. Unmanned aerial vehicles (UAV) can be used to derive vegetation indices (VIs) with high throughput and could provide an efficient way to predict forage yield with high accuracy. The main objective of the study is to investigate the potential of UAV-based multispectral data and machine learning approaches in the estimation of oat biomass. UAV equipped with a multispectral sensor was flown over three experimental oat fields in Volga, South Shore, and Beresford, South Dakota, USA, throughout the pre- and post-heading growth phases of oats in 2019. A variety of vegetation indices (VIs) derived from UAV-based multispectral imagery were employed to build oat biomass estimation models using four machine-learning algorithms: partial least squares (PLS), support vector machine (SVM), Artificial neural network (ANN), and random forest (RF). The results showed that several VIs derived from the UAV collected images were significantly positively correlated with dry biomass for Volga and Beresford (r = 0.2–0.65), however, in South Shore, VIs were either not significantly or weakly correlated with biomass. For Beresford, approximately 70% of the variance was explained by PLS, RF, and SVM validation models using data collected during the post-heading phase. Likewise for Volga, validation models had lower coefficient of determination (R2 = 0.20–0.25) and higher error (RMSE = 700–800 kg/ha) than training models (R2 = 0.50–0.60; RMSE = 500–690 kg/ha). In South Shore, validation models were only able to explain approx. 15–20% of the variation in biomass, which is possibly due to the insignificant correlation values between VIs and biomass. Overall, this study indicates that airborne remote sensing with machine learning has potential for above-ground biomass estimation in oat breeding nurseries. The main limitation was inconsistent accuracy in model prediction across locations. Multiple-year spectral data, along with the inclusion of textural features like crop surface model (CSM) derived height and volumetric indicators, should be considered in future studies while estimating biophysical parameters like biomass.

Impacts of Agricultural Areas on Spatio-Temporal Variability of Daily Minimum Extreme Flows during the Transitional Seasons (Spring and Fall) in Southern Quebec

Several statistical methods were used to analyze the spatio-temporal variability of daily minimum extreme flows (DMEF) in 17 watersheds—divided into three homogenous hydroclimatic regions of southern Quebec—during the transitional seasons (spring and fall), during the 1930–2019 period. Regarding spatial variability, there was a clear difference between the south and north shores of the St. Lawrence River, south of 47° N. DMEF were lower in the more agricultural watersheds on the south shore during transitional seasons compared to those on the north shore. A correlation analysis showed that this difference in flows was mainly due to more agricultural areas ((larger area (>20%) on the south than on the north shore (<5%)). An analysis of the long-term trend of these flows showed that the DMEF of south-shore rivers have increased significantly since the 1960s, during the fall (October to December), due to an increase in rainfall and a reduction in cultivated land, which increased the infiltration in the region. Although there was little difference between the two shores in the spring (April to June), we observed a decrease in minimum extreme flows in half (50%) of the south-shore rivers located north of 47° N.

Invasive Fucus serratus (Fucaceae, Phaeophyceae) responds to climate change along the Atlantic Coast of Nova Scotia, Canada

The distribution and ecology of the invasive brown alga Fucus serratus along the 500 km Atlantic coast of Nova Scotia, Canada, has been poorly explored. We observed significant intertidal penetration at four sites in the southwestern part of the province, and then examined numerous sites along the Atlantic coast of Nova Scotia. Surveys of attached algae in intertidal and shallow subtidal zones and wrack show that F. serratus has become a dominant plant in the low to mid-intertidal zone and can be expected on headlands along the South Shore of Nova Scotia where it can occupy up to 40% of the intertidal zone with cover >75% and mean densities of up to 10 kg m−1. In this zone, F. serratus has replaced Chondrus crispus as the major canopy species, although C. crispus and Corallina officinalis remain primary understory species. At slightly higher elevations, F. serratus was common as an understory beneath Ascophyllum nodosum and Fucus vesiculosus. While geographic spread along the Atlantic coast might reflect the natural dispersal capacity of F. serratus, we hypothesize that the ecological extension into the intertidal zone may be facilitated by harvesting of A. nodosum and by climate change in an ocean-warming hotspot.

Does Anxiety Affect the Clock-Drawing Task in Patients in a Memory Clinic? A Clinical Correlation Study

Background: It is unknown if anxiety levels affect performance on clock drawing test (CDT) in memory clinic patients. Method: We performed a retrospective analysis of memory clinic patients in the south shore of Boston from 2010 to 2019. We correlated anxiety screen data (GAD7) to CDT scores, based on contour, numbers, and hands placement. Univariate analyses used Spearman correlation. A multivariate regression model analzed GAD7 to covariates of CDT, age, sex, and race. Hypothesis : We hypothesized a positive correlation between anxiety levels scored by the GAD7 and CDT. Results: 994 patients in the memory clinic between 2010-2020 had analyzable data. Patients were 58.6% female, 84.6 % White. Mean age was 70.1±14.4, CDT 1.84±1.04. CDT score correlated significantly to race (⍴=-0.16, p&lt; 0.001), age (⍴=-0.28, p&lt;0.001), gender (⍴=0.05, p=0.16), but not GAD7 (⍴=0.05, p=0.27). Multivariate model confirmed the lack of association of anxiety scores to CDT (□= 0.08, p=0.78). GAD7 scores correlated to female gender (□= -1.16, p=0.04). Conclusions: CDT scores were not affected by anxiety as measured on GAD7 scores. However, a positive correlation was shown on anxiety scores in females to CDT completion.

Is Immediate Recall test scores affected by Anxiety in a memory clinic population? A clinical correlation study

Background: It is unknown if anxiety affects performance on immediate recall testing (IR) in memory clinic patients. Method: We performed a retrospective analysis of memory clinic patients in the south shore of Boston from 2010 to 2019. We correlated anxiety screen data (GAD7) to IR scores. Univariate analyses used Spearman correlation. A multivariate regression model analyzed GAD7 to covariates of IR, age, sex, and race. Hypothesis: We hypothesized a positive correlation between anxiety levels scored by GAD7 and IR. Results: 994 patients in the memory clinic between 2010-2020 had analyzable data. Patients were 58.6% female, 84.6 % White. The mean age was 70.1±14.4, IR 6.62 ± 5.4, GAD7 5.5±5.71. On univariate analysis, IR correlated significantly to age (⍴ = 0.08, p = 0.01), gender (⍴ = 0.06, p = 0.046), and race (⍴= - 0.25, p &lt;0.001), but not to GAD7 (⍴=-0.07, p=0.14). The multivariate model confirmed the lack of association of anxiety scores to (□=-0.05, p=0.41) to GAD7 scores. IR task performance was significantly associated only to age (□= -0.04, p=0.03) and gender (□= -1.18, p=0.04) in the regression model. Conclusions: Immediate recall task performance was not significantly affected by anxiety measured by GAD7 scores in a memory clinic population. However, a negative correlation was shown on immediate recall scores in males and older subjects.

Does Anxiety Affect Performance on Attention Task (digit span forward) on the MOCA Test? A clinical correlation study

It is unknown if anxiety affects performance on Digit span forward (DSF) in memory clinic patients. We performed a retrospective chart review of memory clinic patients in the south shore of Boston from 2010 to 2020. We correlated anxiety screen data (GAD7) to Digit Span Forward (DSF) scores obtained from the MoCA. As the data were not normal, we performed univariate analyses with Spearman correlation. A multivariate regression model estimated the relationship of DSF to covariates of GAD7, age, sex, and race. We hypothesized a negative correlation between anxiety levels scored by GAD7 and DSF. H0: Digit span forward DSF ~ GAD7+Age+Sex+Race. A chart review found 965 patients attending the memory clinic between 2010 to 2020 had analyzable data. 433 patients with available DSF and 737 had available GAD7. The patients were 58.7% female and 84.7% caucasian. The mean age was 70.1±14.4, DSF 0.8±0.4 and GAD 5.6±5.7. DSF correlated significantly to race (⍴=-0.25, p=&lt;0.001), but not to gender (⍴=0.05, p=0.149), age (⍴=-0.04, p=0.3), or GAD7 (⍴=-0.018, p=0.71). There was no significant association of DSF to race, age, gender or GAD7 on the multivariate model. In memory clinic subjects there exists no correlation between anxiety levels scored by GAD7 and DSF performance.

Clinical Correlation of Cerebrospinal Fluid Total tau Levels and MMSE Score in a memory clinic

Tau protein levels in cerebrospinal fluid are a biomarker of Alzheimer’s disease. We correlated MMSE severity to CSF tau levels in a large memory clinic sample. We retrospectively analyzed data from patients attending a memory clinic in the south shore of Boston from 2010 to 2020, and had a lumbar puncture to obtain CSF Tau levels. We compiled cognitive screen data from MMSE scores. Univariate analyses used Spearman correlation as data were non-normal. A multivariate model was created including covariates of age, sex, and race. 965 patients attended the memory clinic from 2010 to 2020. 711 had available MMSE scores. 129 subjects had lumbar punctures and available CSF tau levels. Univariate analyses showed that cognition as measured by MMSE total was not correlated to total tau levels in the CSF (rho=-0.07, p&gt;0.05), but caucasian race was inversely associated with CSF tau levels (rho= -0.217, p&lt;0.05). In a multivariate model, tau levels in the CSF were not associated with MMSE, race, gender, or age. In a large memory clinic sample, CSF tau levels did not correlate to MMSE scores, age, race or gender.