Leaf Stoichiometry of Potentilla fruticosa Across Elevations in China's Qilian Mountains

As an individual plant species can develop its own leaf stoichiometry to adapt to environmental changes, this stoichiometry can provide critical information about a plant species' growth and its potential management in the ecosystem housing it. However, leaf stoichiometry is largely undocumented in regions with large environmental changes arising from differences in elevation. The leaf stoichiometry of Potentilla fruticosa L., a major alpine shrub playing an important role in supporting ecosystem functions and services in China's Qilian Mountains (Northeast Qinghai-Tibetan Plateau), was investigated at different elevations (2,400, 2,600, 2,800, 3,000, 3,200, 3,500, and 3,800 m). At each elevation, leaf elemental (C, N, and P) concentrations were measured in P. fruticosa leaves sampled from three plots (10 × 10 m), and edaphic properties were assessed in nine quadrats (1 × 1 m, three quadrats per plot). Temperature and precipitation were calculated using an empirical formula. Maximum and minimum leaf carbon (C) concentrations ([C] _leaf ) of 524 ± 5.88 and 403 ± 3.01 g kg^-1 were measured at 2,600 and 3,500 m, respectively. Leaf nitrogen (N) concentration ([N] _leaf ) showed a generally increasing trend with elevation and peaked at 3,500 m (27.33 ± 0.26 g kg^-1). Leaf phosphorus (P) concentration ([P] _leaf ) varied slightly from 2,400 to 3,200 m and then dropped to a minimum (0.60 ± 0.10 g kg^-1) at 3800 m. The [C] _leaf :[N] _leaf , [C] _leaf :[P] _leaf , and [N] _leaf :[P] _leaf varied little from 2,400 to 3,000 m but fluctuated somewhat at higher elevations. The main factors affecting P. fruticosa leaf stoichiometry were soil organic C, pH, and soil total P, and the main limiting element for the growth of P. fruticosa in the study area was P. In conclusion, changes in elevation affected leaf stoichiometry of P. fruticosa mainly due to altered soil properties, and addressing phosphorus limitation, especially at higher elevations mainly due to losses caused by high precipitation and sparse vegetation, is a key measure to promote P. fruticosa growth in this region.

The role of climate change and vegetation greening on the variation of terrestrial evapotranspiration in northwest China's Qilian Mountains

Terrestrial evapotranspiration (ET_a) reflects the complex interactions of climate, vegetation, soil and terrain and is a critical component in water and energy cycles. However, the manner in which climate change and vegetation greening influence ET_a remains poorly understood, especially in alpine regions. Drawing on the Global Land Evaporation Amsterdam Model (GLEAM) ET_a data, the interannual variability of ET_a and its ties to precipitation (P), potential evaporation (ET_p) and vegetation (NDVI) were analysed. The Budyko framework was implemented over the period of 1982 to 2015 to quantify the response of ET_a to climate change's direct (P and ET_p) and indirect (NDVI) impacts. The ET_a, P, ET_p and NDVI all showed significant increasing trends from 1981 to 2015 with rates of 1.52 mm yr^-1, 3.18 mm yr^-1, 0.89 mm yr^-1 and 4.0 × 10^-4 yr^-1, respectively. At the regional level, the positive contribution of increases in P and NDVI offset the negative contribution of ET_p to the change in ET_a (∆ET_a). The positive ∆ET_a between 1982 and 2001 was strongly linked with the concomitant increase in NDVI. Increases in vegetation contributing to a positive ∆ET_a differed among landscape types: for shrub, meadow and steppe they occurred during both periods, for alpine vegetation between 1982 and 2001, and for desert between 2002 and 2015. Climate change directly contributed to a rise in ET_a, with P as the dominant factor affecting forested lands during both periods, and alpine vegetation between 2002 and 2015. Moreover, ET_p was a dominant factor for the desert between 1982 and 2001, where the variation of P was not significant. The contributions of factors having an impact on ∆ET_a are modulated by both the sensitivity of impact factors acting on ET_a as well as the magnitudes of factor changes. The greening of vegetation can influence ET_a by increasing vegetation transpiration and rainfall interception in forest, brush and meadow landscapes. These findings can help in developing a better understanding of the interaction of ecosystems and hydrology in alpine regions.

Causality of climate, food production and conflict over the last two millennia in the Hexi Corridor, China

The relationship between climate and human society has frequently been investigated to ascertain whether climate variability can trigger social crises (e.g., migration and armed conflicts). In the current study, statistical methods (e.g., correlation analysis and Granger Causality Analysis) are used in a systematic analysis of the potential causality of climate variability on migration and armed conflicts. Specifically, the statistical methods are applied to determine the relationships between long-term fine-grained temperature and precipitation data and contemporary social conditions, gleaned from historical documents covering the last two millennia in China's Hexi Corridor. Results found the region's reconstructed temperature to be strongly coupled with precipitation dynamics, i.e., a warming climate was associated with a greater supply of moisture, whereas a cooling period was associated with more frequent drought. A prolonged cold period tended to coincide with societal instability, such as a shift from unification towards fragmentation. In contrast, a prolonged warm period coincided with rapid development, i.e., a shift from separation to unification. The statistical significance of the causality linkages between climate variability, bio-productivity, grain yield, migration and conflict suggests that climate variability is not the direct causative agent of these phenomena, but that climate reduced food production which gradually lead to migration and conflicts. A conceptual causal model developed through this study describes the causative pathway of climate variability impacts on migration and conflicts in the Hexi Corridor. Applied to current conditions, the model suggests that steady and proactive promotion of the nation's economic buffering capacity might best address the uncertainty brought on by a range of potential future climate scenarios and their potential impacts.

A real-time hourly water index for flood risk monitoring: Pilot studies in Brisbane, Australia, and Dobong Observatory, South Korea

A water resources index based on weight-accumulated precipitation over the passage of time in heavy rainfall events is used in this study for monitoring flood risk and peak danger, as well as to develop flood warnings. In this research, an hourly water resources index (WRI_hr) based on rainfall accumulations over the passage of time is proposed. WRI_hr is able to monitor flood risk by taking into account the hourly effective precipitation that accumulates the precipitation (P_hr) of both current and antecedent hours, while the contributions from the preceding hours is subjected to a time-dependent reduction function that addresses the depletion of water volume by various hydrological processes (e.g., discharge, runoff, evapotranspiration). By converting rainfall into a water resources index (WRI), the hourly precipitation over a 24-h period is redistributed to formulate a long-term water resources index (WRI_hr-L) that monitors flood status based on long-term (more than 1 year) fluctuations in P_hr and a short-term water resources index (WRI_D-hr-S) that considers shorter (D = 24-148 hourly) accumulations of the P_hr data. WRI was assessed for its potential in flood monitoring at two hydrologically diverse sites: Dobong (South Korea; August 1998) and Brisbane (Australia; December 2010-January 2011), and its applicability was verified using river water level (H) measurements at hydrological stations. The power spectrum density and spectral coherence of hourly rainfall, river water level, and the corresponding WRI showed good agreements, as did the low and high frequency wavelet components using the discrete wavelet transform algorithm. Importantly, WRI_24-hr-S computed over 24 hourly accumulation periods was able to mimic the risk of short-term (flash-style) floods caused by concentrated rainfall, whereas WRI_hr-L was more useful for flood risk assessment caused by an event over a long-term period. Dynamical changes in H were closely in-phase with the patterns of change noted in the WRI_hr over the respective temporal scale. We conclude that the proposed WRI was able to replicate the flood evolution over the passage of time and, therefore, could possibly aid in the early warning of water-related disasters, demonstrating its practicality for continuous monitoring of the flood risk when a sustained period of rainfall event is observed.

Very short-term reactive forecasting of the solar ultraviolet index using an extreme learning machine integrated with the solar zenith angle

Exposure to erythemally-effective solar ultraviolet radiation (UVR) that contributes to malignant keratinocyte cancers and associated health-risk is best mitigated through innovative decision-support systems, with global solar UV index (UVI) forecast necessary to inform real-time sun-protection behaviour recommendations. It follows that the UVI forecasting models are useful tools for such decision-making. In this study, a model for computationally-efficient data-driven forecasting of diffuse and global very short-term reactive (VSTR) (10-min lead-time) UVI, enhanced by drawing on the solar zenith angle (θ_s) data, was developed using an extreme learning machine (ELM) algorithm. An ELM algorithm typically serves to address complex and ill-defined forecasting problems. UV spectroradiometer situated in Toowoomba, Australia measured daily cycles (0500-1700h) of UVI over the austral summer period. After trialling activations functions based on sine, hard limit, logarithmic and tangent sigmoid and triangular and radial basis networks for best results, an optimal ELM architecture utilising logarithmic sigmoid equation in hidden layer, with lagged combinations of θ_s as the predictor data was developed. ELM's performance was evaluated using statistical metrics: correlation coefficient (r), Willmott's Index (WI), Nash-Sutcliffe efficiency coefficient (E_NS), root mean square error (RMSE), and mean absolute error (MAE) between observed and forecasted UVI. Using these metrics, the ELM model's performance was compared to that of existing methods: multivariate adaptive regression spline (MARS), M5 Model Tree, and a semi-empirical (Pro6UV) clear sky model. Based on RMSE and MAE values, the ELM model (0.255, 0.346, respectively) outperformed the MARS (0.310, 0.438) and M5 Model Tree (0.346, 0.466) models. Concurring with these metrics, the Willmott's Index for the ELM, MARS and M5 Model Tree models were 0.966, 0.942 and 0.934, respectively. About 57% of the ELM model's absolute errors were small in magnitude (±0.25), whereas the MARS and M5 Model Tree models generated 53% and 48% of such errors, respectively, indicating the latter models' errors to be distributed in larger magnitude error range. In terms of peak global UVI forecasting, with half the level of error, the ELM model outperformed MARS and M5 Model Tree. A comparison of the magnitude of hourly-cumulated errors of 10-min lead time forecasts for diffuse and global UVI highlighted ELM model's greater accuracy compared to MARS, M5 Model Tree or Pro6UV models. This confirmed the versatility of an ELM model drawing on θ_sdata for VSTR forecasting of UVI at near real-time horizon. When applied to the goal of enhancing expert systems, ELM-based accurate forecasts capable of reacting quickly to measured conditions can enhance real-time exposure advice for the public, mitigating the potential for solar UV-exposure-related disease.