Deuterium tracing for the estimation of transpiration from trees Part 3. Measurements of transpiration from Eucalyptus plantation, India

Measurements of transpiration from individual trees of Eucalyptus from plantations at four different sites in Karnataka, Southern India, are presented. These show large (as much as tenfold) differences in the transpiration between premonsoon and postmonsoon periods, a reflection of the effects of soil-moisture stress in the premonsoon periods. For trees with diameters at breast height (DBH) less than 10 cm the transpiration rate of individual trees is proportional to the square of the DBH. For trees which are not experiencing soil-water stress the daily transpiration rate of individual trees, q, is well represented by the relation: q = (6.6 ± 0.3)g (m³ day⁻¹ where g (m²) is the tree basal area. On a unit ground area basis the transpiration rate, expressed as a depth per day is given by the relation: E_t = (0.66 ± 0.03)G (mm day⁻¹ where g(m²ha⁻¹) is the total basal area per hectare. For all the sites studied, although there is evidence for the ‘mining’ of soil water as roots penetrate deeper depths in the soil each year, there is no evidence for direct abstraction from the water table.     

The impact of land use change on water resources in sub-Saharan Africa: a modelling study of Lake Malawi

A modelling study to investigate the effects of land use change from natural forest to agricultural land on large-scale catchment runoff in southern Africa is described. The evaporative component of the model considers the catchment to be composed of one of three surface types—forest, agricultural land or water surface. Values of the model parameters for the forest and agricultural lands were obtained from experimental studies carried out in the dry zone of India. Estimates of average monthly potential evaporation, together with measurements of monthly rainfall, were used in the model to predict the monthly levels of Lake Malawi. These were compared with observed levels. From 1896 to 1967 the major fluctuations in lake level, both seasonally and annually, are well described by this model (excepting the period from 1935 to 1945, immediately following the time when there was no outflow from the lake) using a value of 64% for the forest coverage of the catchment. The overall agreement between prediction and observation indicates that variations in rainfall alone, without changes in either evaporative demand or in the hydraulic regime of the lake, are sufficient to explain lake level changes. For the more recent period (1954–1994), model predictions of lake level which take into account a decrease in forest cover of 13% over the period 1967–1990 (consistent with the actual decrease in forest cover for this period) agree well with observations both annually and seasonally. Without this decrease in forest cover, the model predicted that the lake level would have been about 1 m lower than that observed during the southern African drought of 1992. The model, in conjunction with real-time rainfall data obtained from land-based gauges, radar or satellite observations, can be used for real-time water resource management applications such as the operation of barrages regulating the flow from Lake Malawi or for the issuing of flood or drought warnings.     

Geochemical implications of induced changes in C13 fractionation by blue-green algae

The photosynthetic fractionation of carbon isotopes by blue-green algae in laboratory culture is dependent in a non-linear fashion on the CO₂ concentration in the feed gas. For the three species tested, the minimum fractionation occurred at a CO₂ concentration of 0.2% in air and was approximately zero for the two marine species tested. Enrichment of C¹² in the reduced carbon is not an inevitable result of photosynthetic carbon fixation. Temperature and pH had no detectable effect on fractionation. The maximum fractionation observed in the laboratory cultures or in recent blue-green algal mats was 18‰. Differences in the isotope ratio of coexisting oxidized and reduced carbon in Precambrian stromatolites are as great as 31‰. Present carbon isotopic evidence is not consistent with the idea that blue-green algae were major contributors to the organic matter in Precambrian sediments.     

Hydrological impact of Eucalyptus plantation in India

A brief review is given of the results of hydrological studies carried out in southern India on the effects of plantations of Eucalyptus and other fast growing exotic tree species on water resources, erosion and soil nutrients at sites of differing rainfall and soil depth in Karnataka. New results on the impacts of the plantations on raindrop-induced erosion and preliminary results from studies that are aimed at improving the water use efficiency of the plantations are also presented.

The erosion studies indicate that soil detachment resulting from net rainfall beneath Eucalyptus camaldulensis will be greater than beneath Pinus caribaea but less than that beneath the indigenous species, Tectona grandis.

The water use efficiency studies, which make use of a controlled environment facility, have confirmed that in the dry zone of southern India water availability is the principal limiting factor on growth although, when water is available, nutrient limitations become important. Removal of both water and nutrient stress results typically in a fivefold increase in volume growth for the first year of growth.     

Effects of dredging and open-water disposal on benthic macroinvertebrates in a South Carolina estuary

Approximately 28,475 m³ of muddy sediments were dredged from a shoal in a South Carolina estuarine system and released near the surface at a nearby site having high tidal current velocities. Effects at the dredged sites included decreased macrofaunal abundance and changes in species composition. These effects appeared to be short term, with substantial recovery occurring within 3 months. Rapid recovery was primarily attributed to immigration through slumping of channel wall sediments similar to those dredged. Detrimental effects on benthic macrofauna in the area of open water disposal were minimal. Most differences noted in community structure between collection dates were attributed to sampling and seasonal variability. The absence of a major longterm disruption to the benthos in the disposal area was probably due to (1) strong tidal currents, which rapidly dispersed the moderate amount of mud sediments released; (2) surface disposal, permitting wider dispersal; and (3) disposal during late autumn, a period of low faunal recruitment.     

Drowned shelf-edge deltas, barrier islands and related features along the outer continental shelf north of the head of De Soto Canyon, NE Gulf of Mexico

Four drowned shelf-edge delta complexes, two drowned shelf deltas, three drowned barrier islands, large areas of “hardground”, and fields of bedforms on the mid and outer continental shelf and uppermost slope north of the head of De Soto Canyon, NE Gulf of Mexico were mapped with high-resolution multibeam echosounder. Deltas formed not during the last eustatic low stand, but during one or more interstadials when eustatic sea levels were only 60 to 80 m below present sea level. The barrier islands and deltas must have been cemented prior to rapid falls of eustatic sea level that occurred during global glaciations. Cementation is necessary to have preserved the barrier islands from erosion and subsequent destruction by the rapid sea-level rise during the last deglaciation. The preservation of the relict bathymetry is so good that features that superficially resemble trough blowouts are found in association with one of the relict barrier islands. Asymmetric bedforms on the midshelf in water depths of 50 to 60 m indicate transport directions to the SW but asymmetric bedforms in water depths of the upper slope between 100 and 120 m on the S and SE flanks of the drowned shelf-edge deltas indicate a different current direction, a separate flow that is a continuation of a SW-flowing current that was previously found on the upper slope off NW Florida.     

An assessment of coefficient accuracy in linear regression models with spatially varying coefficients

The realization in the statistical and geographical sciences that a relationship between an explanatory variable and a response variable in a linear regression model is not always constant across a study area has led to the development of regression models that allow for spatially varying coefficients. Two competing models of this type are geographically weighted regression (GWR) and Bayesian regression models with spatially varying coefficient processes (SVCP). In the application of these spatially varying coefficient models, marginal inference on the regression coefficient spatial processes is typically of primary interest. In light of this fact, there is a need to assess the validity of such marginal inferences, since these inferences may be misleading in the presence of explanatory variable collinearity. In this paper, we present the results of a simulation study designed to evaluate the sensitivity of the spatially varying coefficients in the competing models to various levels of collinearity. The simulation study results show that the Bayesian regression model produces more accurate inferences on the regression coefficients than does GWR. In addition, the Bayesian regression model is overall fairly robust in terms of marginal coefficient inference to moderate levels of collinearity, and degrades less substantially than GWR with strong collinearity.     

Ultraprecise Absolute Time Synchronization for Distributed Acquisition Systems

In this paper, we describe an algorithm built on top of a precision time protocol (PTP) implementation that allows for synchronization, syntonization and absolute time referencing to coordinated universal time (UTC), including the estimation of timestamp uncertainty. We call this the software grandmaster (SWGM) algorithm, since it provides similar services to a PTP grandmaster clock. We show that SWGM allows timestamps to be coordinated between multiple participants in a distributed measurement system with typical performance of plusmn 86 ns [root mean square (rms)] over commodity switched Ethernet connections using hardware PTP and hardware-derived timestamps. We further show that when software-derived timestamps are used the uncertainty in the timestamps is primarily driven by the latency of the system calls to read the PTP hardware, and may be on the order of 15-25 s (rms) depending on process priority, hardware bus speed, and host processor clock rate. We also show that SWGM is robust against dropped network packets up to approximately 60% loss of packets.