Triassic-Jurassic climate in continental high-latitude Asia was dominated by obliquity-paced variations (Junggar Basin, Ürümqi, China).

Empirical constraints on orbital gravitational solutions for the Solar System can be derived from the Earth's geological record of past climates. Lithologically based paleoclimate data from the thick, coal-bearing, fluvial-lacustrine sequences of the Junggar Basin of Northwestern China (paleolatitude ∼60°) show that climate variability of the warm and glacier-free high latitudes of the latest Triassic-Early Jurassic (∼198-202 Ma) Pangea was strongly paced by obliquity-dominated (∼40 ky) orbital cyclicity, based on an age model using the 405-ky cycle of eccentricity. In contrast, coeval low-latitude continental climate was much more strongly paced by climatic precession, with virtually no hint of obliquity. Although this previously unknown obliquity dominance at high latitude is not necessarily unexpected in a high CO2 world, these data deviate substantially from published orbital solutions in period and amplitude for eccentricity cycles greater than 405 ky, consistent with chaotic diffusion of the Solar System. In contrast, there are indications that the Earth-Mars orbital resonance was in today's 2-to-1 ratio of eccentricity to inclination. These empirical data underscore the need for temporally comprehensive, highly reliable data, as well as new gravitational solutions fitting those data.

Global attraction and stability for Cohen-Grossberg neural networks with delays.

We consider a class of Cohen-Grossberg neural networks with delays. We prove the existence and global asymptotic stability of an equilibrium point and estimate the region of existence. Furthermore, we show that the trajectories of the neural networks with positive initial data will stay in the positive region if the amplification function satisfies a divergent condition. We also establish the existence of a globally attracting compact set for more general networks. We estimate this compact set explicitly in terms of the network parameters from physiological and biological models. Our results can be applied to neural networks with a wide range of activation functions which are neither bounded nor globally Lipschitz continuous such as the Lotka-Volterra model. We also give some examples and simulations.