Water storage in the stems of woody plants contributes to their responses to short-term water shortages. To estimate the contribution of water storage to the daily water budget of trees, time lags of sap flow between different positions of trunk are used as a proxy of stem water storage. In lianas, another large group of woody species, it has rarely been studied whether stored water functions in their daily water use, despite their increasing roles in the carbon and water dynamics of tropical forests caused by their increasing abundance. We hypothesized that lianas would exhibit large time lags due to their extremely long stems, wide vessels and large volume of parenchyma in the stem. We examined time lags in sap flow, diel changes of stem volumetric water content (VWC) and biophysical properties of sapwood of 19 lianas and 26 co-occurring trees from 27 species in 4 forests (karst, tropical seasonal, flood plain and savanna) during a wet season. The plants varied in height/length from <5 to >60 m. The results showed that lianas had significantly higher saturated water content (SWC) and much lower wood density than trees. Seven of 19 liana individuals had no time lags; in contrast, only 3 of 26 tree individuals had no time lags. In general, lianas had shorter time lags than trees in our data set, but this difference was not significant for our most conservative analyses. Across trees and lianas, time lag duration increased with diurnal maximum changeable VWC but was independent of the body size, path length, wood density and SWC. The results suggest that in most lianas, internal stem water storage contributes little to daily water budget, while trees may rely more on stored water in the stem.

As modern complex computer systems become increasingly sophisticated and data-intensive, people have long recognized the importance of precisely and unambiguously specifying them with formal methods. This paper advocates the use of Object-Z, a formal specification language, in the description of complex systems. Object-Z is an extension to the Z language to facilitate specification in an object-oriented style. First, we informally describe the syntax and semantics of Object-Z, highlighting those features that facilitate decomposing a large system into a collection of underlying objects and thus separating concerns. Then, we demonstrate the use of Object-Z by presenting a case study of a petrol supply system, illustrating how the system runs by communicating the constituent objects. Finally, we discuss several issues we encountered in this exercise, which may serve as feedback to the development of Object-Z.

Using solid phase extraction and high performance liquid chromatography (HPLC), this paper studies the adsorption capability of activated carbon to phenol. The phenol wastewater is simulated with raw water and pure water, respectively. In detail, we examine impact factors including adsorption time, activated carbon dosage, equilibrium concentration of phenol. Using the adsorption principles from the aspect of kinetics, our study shows that adsorption of powder activated carbon to phenol conforms to the Fran Delhi Freundrich adsorption model.

In regressing demographic data, neighborhood information needs to be taken into account, which makes conventional linear regression models inappropriate. In this paper, to incorporate neighborhood information without incurring extra computational cost, we investigate various possibilities to reformulate input data. Experiments show that augmenting each site with its neighborhood information gives the best results.

Classification on geospatial data is different from classical classification in that spatial context must be taken into account. In particular, the validation criterion functions should incorporate both classification accuracy and spatial accuracy. However, direct combination of the two accuracies is cumbersome, due to their different subjects and scales. To circumvent this difficulty, we develop two new criterion functions that indirectly incorporate spatial accuracy into classification accuracy-based functions. Next, we formally introduce a set of ideal properties that an appropriate criterion function should satisfy, giving a more meaningful interpretation to the relative significance coefficient in the weighted scheme. Finally, we compare the proposed new criterion functions with existing ones on a large data set for 1980 US presidential election.