We present the first nonlinear results on the problem of non-rotating thermal convection in an internally heated full sphere. A nonlinear stability analysis by the energy method yields that, at least for no-slip boundary conditions, the critical Rayleigh numbers for linear stability and nonlinear stability coincide. We then explore different ranges of the parameter regime using direct numerical simulations. We first report on the system behaviour for a fixed Prandtl number of unity and both stress-free and no-slip boundary conditions up to very high forcing, reaching Rayleigh number $Ra=2\times 10^{12}$, approximately 250 million times the critical value ($Ra_c$) for the onset of convection under no-slip conditions. For both boundary conditions, we observe a scaling for the advective heat transfer measured by the Nusselt number $Nu$ close to $Nu \sim Ra^{1/4}$. This is consistent with a scaling prediction that we formulate analogously to the classical scaling in Rayleigh–Bénard convection. We then investigate the Prandtl number dependence at low to intermediate forcing for stress-free boundary conditions in the ranges $0.1 \leq Pr \leq 30$ and $Ra_c=3091\leq Ra \leq 3\times 10^5 \approx 100Ra_c$. We find five distinct dynamical regimes depending on the Prandtl number, describe each regime individually and issue heuristic interpretations of the system behaviour where possible.

The architect Hans Döllgast (1891–1974) has steadily gained in international recognition. His works of postwar reconstruction in Munich have been heralded as original contributions to modern architecture that resist historiographic classifications, such as modernist vs regionalist, avant garde vs traditionalist, or internationalist vs nationalist. Döllgast was also a revered pedagogue and prolific author, and his varied writings have yet to receive much scholarly attention. Döllgast’s books and essays present a significant body of sources that shed light on the complexity of architectural discourse in the formative years of modern architecture in Germany. This article considers Döllgast’s study of farmhouse ‘parlours’, entitled Alte und neue Bauernstuben (‘Old and New Farmhouse Parlours’) was first published in 1937. It was both his most popular book and the one that critics and historians have paid least attention to. Though it may appear antiquarian at first glance, it is in fact both critical and contemporary in spirit. Döllgast’s study sheds light upon his mature thinking about the relevance of the vernacular for the modern house. It also serves to question a general assumption in the existing literature that Döllgast only engaged with tenets of modern architecture after the war, having been a regionalist aloof from the discourse of the modern movement prior to the war. Scholars have shown that the loaded motif of the vernacular was never the sole preserve of anti-modernist conservatives and played a significant, if ambivalent, role within modernist discourse, from the late Wilhelmine period to postwar West Germany. While it reflects these wider trends, Alte und neue Bauernstuben also eschews alignment with the dominant strands of architectural discourse of its time by charting an independent-minded path in the context of imposed totalitarian uniformity. Döllgast’s text thus stands out in modern architectural discourse less for adducing the farmhouse as such, than for developing such a close, multifaceted reading of a particular vernacular interior, while alluding to more than elaborating its relevance for contemporary architecture. Ultimately, Döllgast’s study served him to develop a practical phenomenology of dwelling.

Assistive forces transmitted from wearable robots to the robot’s users are often defined by controllers that rely on the accurate estimation of the human posture. The compliant nature of the human–robot interface can negatively affect the robot’s ability to estimate the posture. In this article, we present a novel algorithm that uses machine learning to correct these errors in posture estimation. For that, we recorded motion capture data and robot performance data from a group of participants (n = 8; 4 females) who walked on a treadmill while wearing a wearable robot, the Myosuit. Participants walked on level ground at various gait speeds and levels of support from the Myosuit. We used optical motion capture data to measure the relative displacement between the person and the Myosuit. We then combined this data with data derived from the robot to train a model, using a grading boosting algorithm (XGBoost), that corrected for the mechanical compliance errors in posture estimation. For the Myosuit controller, we were particularly interested in the angle of the thigh segment. Using our algorithm, the estimated thigh segment’s angle RMS error was reduced from 6.3° (2.3°) to 2.5° (1.0°), mean (standard deviation). The average maximum error was reduced from 13.1° (4.9°) to 5.9° (2.1°). These improvements in posture estimation were observed for all of the considered assistance force levels and walking speeds. This suggests that ML-based algorithms provide a promising opportunity to be used in combination with wearable-robot sensors for an accurate user posture estimation.

Current dam discharge patterns in Noxon Rapids Reservoir reduce concentration and exposure times (CET) of herbicides used for aquatic plant management. Herbicide applications during periods of low dam discharge may increase herbicide CETs and improve efficacy. Applications of rhodamine WT dye were monitored under peak (736 to 765 m3 s−1) and minimum (1.4 to 2.8 m3 s−1) dam discharge patterns to quantify water-exchange processes. Whole-plot dye half-life under minimal discharge was 33 h, a 15-fold increase compared with the dye treatment during peak discharge. Triclopyr concentrations measured during minimum discharge within the treated plot ranged from 214 ± 25 to 1,243 ± 36 µg L−1 from 0 to 48 h after treatment (HAT), respectively. Endothall concentrations measured during minimum discharge in the same plot ranged from 164 ± 78 to 2,195 ± 1,043 µg L−1 from 0 to 48 HAT, respectively. Eurasian watermilfoil (Myriophyllum spicatum L.) occurrence in the treatment plot was 66%, 8%, and 14% during pretreatment, 5 wk after treatment (WAT), and 52 WAT, respectively. Myriophyllum spicatum occurrence in the nontreated plot was 68%, 71%, and 83% during pretreatment, 5 WAT, and 52 WAT, respectively. Curlyleaf pondweed (Potamogeton crispus L.) occurrence in the treatment plot was 29%, 0%, and 97% during pretreatment, 5 WAT, and 52 WAT, respectively. Potamogeton crispus increased from 24% to 83% at 0 WAT to 52 WAT, respectively, in the nontreated plot. Native species richness declined from 3.3 species per point to 2.1 in the treatment plot in the year of treatment but returned to pretreatment numbers by 52 WAT. Native species richness did not change during the study in the nontreated reference plot. Herbicide applications during periods of low flow can increase CETs and improve control, whereas applications during times of high-water flow would shorten CETs and could result in reduced treatment efficacy.