A comparison of MAGIC and paleolimnological predictions of preindustrial pH for 55 Swedish lakes

Two fundamentally different approaches to define reference conditions for acidification assessments are hydrogeochemical modeling and paleolimnological reconstructions. Both methods have been applied to calculate the preindustrial chemistry for 55 Swedish lakes in two independent studies. This paper investigates whether these methods give similar reconstructions of the preindustrial pH for these lakes. Special focus has been attached to the importance of total organic carbon concentrations and CO2 partial pressure (pCO2) in the conversion from ANC to pH in the hydrogeochemical modeling. With a uniform pCO2 value for all the lakes of 0.63 matm, the mean absolute difference between pH from the hydrogeochemical model and the paleolimnological pH was +0.23 units (mean absolute difference 0.36 units). If instead a lake specific preindustrial pCO2 is assumed, equal to contemporary pCO2, the mean difference in the predicted preindustrial pH between the two methods was reduced to +0.03 units (mean absolute difference 0.22 units). Statistical analyses indicated that with a lake specific pCO2, the difference between the reconstructions is smaller than 0.13 pH-units at a 95% level of significance. The results of this study build confidence in the reliability of both methods, providing that lake-specific estimates of pC02 are used.     

Spatial variation of streamwater chemistry in two Swedish boreal catchments: implications for environmental assessment

To evaluate the scale-dependent spatial variability of water chemistry within two Swedish boreal catchments (subcatchment areas 0.01-78 km2), samples were taken at every junction in the stream network during June 2000 and August 2002. The values of most chemical constituents spanned more than an order of magnitude, and the range was similar to that found in all of Northern Sweden by the national stream survey in 2000. According to the official assessment tools used in Sweden, the entire range of environmental status (for pH, absorbance, alkalinity, dissolved organic carbon (DOC)) and human acidification influence existed within these two study catchments. The water chemistry parameters were relatively stable at catchment areas greater than 15 km2. Sampling at that scale may be adequate if generalized values for the landscape are desired. However the chemistry of headwaters, where much of the stream length and aquatic ecosystem is found would not be characterized. Map parameters correlated to the variability in a key chemical parameter, DOC, but the best predictive map parameters differed markedly between catchments. This study highlights the importance of accounting for headwater spatial variability in environmental assessments of running waters, even in relatively pristine areas. The nature of drainage networks with many headwaters and progressively fewer downstream watercourses makes this a considerable challenge.     

Chemical recovery of surface waters across the northeastern united states from reduced inputs of acidic deposition: 1984-2001

Changes in lake water chemistry between 1984 and 2001 at 130 stratified random sites across the northeastern United States were studied to evaluate the population-level effects of decreases in acidic deposition. Surface-water S04(2-) concentrations decreased across the region at a median rate of -1.53 microequiv L(-1) year(-1). Calcium concentrations also decreased, with a median rate of -1.73 microequiv L(-1) year(-1). This decrease in Ca2+ retarded the recovery of surface water acid neutralizing capacity (Gran ANC), which increased at a median rate of 0.66 microequiv L(-1) year(-1). There were small increases in pH in all subregions except central New England and Maine, where the changes were not statistically significant. Median NO3- trends were not significant except in the Adirondacks, where NO3- concentrations increased at a rate of 0.53 microequiv L(-1) year(-1). A regionwide decrease in the concentration of total Al, especially in ponds with low ANC values (ANC < 25 microequiv L(-1)), was observed in the Adirondack subregion. These changes in Al were consistent with the general pattern of increasing pH and ANC. Despite the general pattern of chemical recovery, many ponds remain chronically acidic or are susceptible to episodic acidification. The continued chemical and biological recovery at sites in the northeastern United States will depend on further controls on S and N emissions.     

Ion budgets and sediment-water interactions during the experimental acidification and recovery of Little Rock Lake, Wisconsin

Ion budgets for the two basins of experimentally acidified Little Rock Lake (Vilas County, WI, U.S.A.) indicate that Ca2+, Mg2+, and K+ were released from the bottom sediments to the water column during 1984-1994, and NH4+, NO3-, and SO4(2-) were removed for a net internal alkalinity generation (IAG). Sulfate removal contributed approximately 50% of the IAG in the reference basin, and cation production generated approximately 40%. In-lake processes in the reference basin removed approximately 38% of the sulfate input; 58% was lost to outflow, and 4% remained in the water column. As a result of acid additions that stimulated sulfate reduction and lower pH that enhanced ion exchange, sulfate removal and Ca2+ production were more important for IAG in the treatment basin. During 1984-1994, sulfate removal contributed about 61% of the IAG, and Ca2+ production contributed about half of the IAG from cation production. In the treatment basin, in-lake processes removed about 46% of the total input of sulfate (including acid additions); 36% was lost to outflow and 18% remained in the water column (representing approximately 25% of the added acid). In both basins of LRL, NH4+ consumption roughly balanced NO3- consumption, and net N transformations provided only 3-12% of the IAG. Overall, Na+ and Cl- were conservative in both basins during 1984-1994. Most ion budget components, including calculated internal reaction terms, showed fairly large interannual variations; e.g., ion inputs (dominated by atmospheric deposition) varied by a factor of about two. Over the 10-year period, ANC terms calculated from the budgets as the difference between base cation and acid anion terms agreed well with measured ANC terms for the budget components, indicating that the budgets accounted for all important IAG constituents.     

Regional assessment of the response of the acid-base status of lake watersheds in the Adirondack region of New York to changes in atmospheric deposition using PnET-BGC

Understanding the response of soil and surface waters to changes in atmospheric deposition is critical for guiding future legislation on air pollution. The Adirondack region of New York experiences among the most severe ecological impacts from acidic deposition. The region is characterized by considerable variability in atmospheric deposition, surficial and bedrock geology, hydrologic flow paths, and vegetation resulting in variability in effects of acidic deposition. In this study, an integrated biogeochemical model (PnET-BGC) was applied to 37 forest lake watersheds to assess the response of soil and surface waters of the Adirondacks to changes in atmospheric deposition at a regional scale. Model-simulated surface water chemistry was validated against data from two synoptic surveys conducted in 1984 and 2001. Results indicate that the model is able to capture the observed changes in surface water chemistry during this period. The model was further used to forecast the response of soil and surface waters to three future emission control scenarios. Results indicate that under the Clean Air Act, surface water SO4(2-) concentrations will continue to decrease at a median rate of -0.38 microeq/L-yr, and surface water ANC is predicted to increase at a median rate of 0.11 microeq/L-yr. More aggressive emission reductions will accelerate the rate of recovery. Under an aggressive control scenario, which represents an additional 75% reduction in SO2 emissions beyond the implementation of the Clean Air Act, surface water SO4(2-) concentrations are predicted to decrease at a median rate of -0.88 microeq/L-yr, and surface water ANC is predicted to increase at a median rate of 0.43 microeq/L-yr. Model predictions of several biologically relevant chemical indicators are also reported.     

Effect of declining lake base cation concentration on freshwater critical load calculations

Steady-state critical load models have been extensively used as the scientific underpinning for air pollution control policies in Europe and are currently being applied to other parts of the world. An important assumption of steady-state models is that critical load estimates do not change through time (or time scale of interest). The most commonly used model for estimating freshwater critical loads is the steady-state water chemistry (SSWC) model. In this study we examined changes in SSWC critical load estimates for 29 lakes in south-central Ontario using data collected 13 years apart (1985-1998), during which time bulk sulfate (SO4(2-)) deposition decreased by 35%. In lakes with the lowest base cation concentrations (<120 microequiv L(-1) Ca2+), the decrease in SO4(2-) concentration was accompanied by an approximately equivalent decrease in base cation concentration, resulting in only a minimal increase in Acid Neutralizing Capacity (ANC) during the 13-year period (median increase 1.6 microequiv L(-1)), and the median critical load for acidity (CL(A)) estimated by the SSWC model decreased by 14.6%. These changes may have been brought about by declining base cation concentrations owing to continued soil acidification in the region. In contrast, in lakes with higher base cation concentrations (>150 microequiv L(-1) Ca2+), the relative decline in base cation concentration was not as great, resulting in a larger increase in ANC (median increase 13.1 microequiv L(-1)) and an increase in the estimated CL(A) (median 5.1%). Lakes with moderate base cation concentrations (120-150 microequiv L(-1) Ca2+) exhibited an intermediate response; the median ANC increased by 8.8 microequiv L(-1) and the estimated CL-(A) decreased by 2.2%. In central Ontario, SSWC critical load estimates based on data taken only 13 years apart change quite dramatically due to changing lake base cation concentrations, and the response appears to depend on the base status of the lakes. The changing values obtained from the SSWC model have important consequences for policy decisions regarding acceptable levels of acid deposition. The application of dynamic models that take into account changes in lake/soil chemistry appears more appropriate for estimating acceptable levels of acid deposition in the region.     

Evaluation of indicator-based pH measurements for freshwater over a wide range of buffer intensities

Two different sulfonephthalein indicators, cresol red (CR) with a pKa of approximately 8.3 and bromothymol blue (BTB) with pKa of approximately 7.4, were tested for an analysis of freshwater over a broad range of pH and total alkalinity values. Measurements from an autonomous sensor system using a 1 cm optical path length were compared to those using a 10 cm path length on a benchtop spectrophotometer. The indicator pH perturbation was quantified with a thermodynamic model and nonlinear least-squares analysis. The laboratory study found that the perturbation-corrected pH differed between the 1 cm (large indicator perturbation) and 10 cm (small indicator perturbation) optical path length measurements from -0.017 to +0.15 with a median of +0.0041 pH units for CR and from -0.015 to +0.026 with a median of -0.0008 pH units for BTB. Precision was +0.0005-0.013 and +0.0001-0.0027 pH units for the 1 and 10-cm-path-length measurements, respectively. The autonomous sensor was deployed for 14 days in a local creek. Simultaneous glass pH electrode measurements had a large negative and drifting offset (-0.15 to -0.40 pH units) compared to the indicator-based measurements. This study is the first in situ comparison between potentiometric and spectrophotometric pH methods in a freshwater system.     

CALIBRATION OF THE PEA TENDEROMETER AND A COMPARISON OF TENDEROMETER READINGS FROM THREE COUNTRIES

The validity of the Texture Test System (TTS) as a reference calibration instrument for the Pea Tenderometers has been evaluated. The factors influencing the results are discussed. A linear correlation between TTS values and tenderometer units has been obtained with a correlation coefficient of at least 0.98. Calibration of the Pea Tenderometers by means of the TTS is now the standard procedure in Sweden. A comparison has been made between Swedish, English and American tenderometer readings. The results show that Swedish and English tenderometer readings are quite similar, whereas the U.S. values differ by up to 9 TU depending on the sample tenderness.     

Chemical equilibrium modeling of organic acids, pH, aluminum, and iron in Swedish surface waters

A consistent chemical equilibrium model that calculates pH from charge balance constraints and aluminum and iron speciation in the presence of natural organic matter is presented. The model requires input data for total aluminum, iron, organic carbon, fluoride, sulfate, and charge balance ANC. The model is calibrated to pH measurements (n = 322) by adjusting the fraction of active organic matter only, which results in an error of pH prediction on average below 0.2 pH units. The small systematic discrepancy between the analytical results for the monomeric aluminum fractionation and the model results is corrected for separately for two different fractionation techniques (n = 499) and validated on a large number (n = 3419) of geographically widely spread samples all over Sweden. The resulting average error for inorganic monomeric aluminum is around 1 μM. In its present form the model is the first internally consistent modeling approach for Sweden and may now be used as a tool for environmental quality management. Soil gibbsite with a log *Ks of 8.29 at 25°C together with a pH dependent loading function that uses molar Al/C ratios describes the amount of aluminum in solution in the presence of organic matter if the pH is roughly above 6.0.     

Anthropogenically driven changes in chloride complicate interpretation of base cation trends in lakes recovering from acidic deposition

Declines in Ca and Mg in low ANC lakes recovering from acidic deposition are widespread across the northern hemisphere. We report overall increases between 1984 and 2004 in the concentrations of Ca + Mg and Cl in lakes representing the statistical population of nearly 4000 low ANC lakes in the northeast U.S. Increases in Cl occurred in nearly all lakes in urbanized southern New England, but only 18% of lakes in more remote Maine had Cl increases. This spatial pattern implicates road salt application as the major source of the increased Cl salts. Among the 48% of the lake population classified as salt-affected, the median changes in Cl (+133 microeq/L) and Ca + Mg (+47 microeq/ L) were large and positive in direction over the 20 years. However, in the unaffected lakes, Cl remained stable and Ca + Mg decreased (-3 microeq/L), consistent with reported long-term trends in base cations of acid-sensitive lakes. This discrepancy between the Cl groups suggests that changes in ion exchange processes in salt-affected watersheds have altered the geochemical cycling of Ca and Mg. One policy-relevant implication is that waters influenced by Cl salts complicate regional assessments of surface water recovery from "acid rain" related to the passage of the Clean Air Act.     

Scale-dependent temporal variations in stream water geochemistry

A year-long study of four western Montana streams (two impacted by mining and two "pristine") evaluated surface water geochemical dynamics on various time scales (monthly, daily, and bi-hourly). Monthly changes were dominated by snowmelt and precipitation dynamics. On the daily scale, post-rain surges in some solute and particulate concentrations were similar to those of early spring runoff flushing characteristics on the monthly scale. On the bi-hourly scale, we observed diel (diurnal-nocturnal) cycling for pH, dissolved oxygen, water temperature, dissolved inorganic carbon, total suspended sediment, and some total recoverable metals at some or all sites. A comparison of the cumulative geochemical variability within each of the temporal groups reveals that for many water quality parameters there were large overlaps of concentration ranges among groups. We found that short-term (daily and bi-hourly) variations of some geochemical parameters covered large proportions of the variations found on a much longer term (monthly) time scale. These results show the importance of nesting short-term studies within long-term geochemical study designs to separate signals of environmental change from natural variability.     

Hydroxy-PCBs, PBDEs, and HBCDDs in serum from an elderly population of Swedish fishermen's wives and associations with bone density

Lack of human exposure data is frequently reported as a critical gap in risk assessments of environmental pollutants, especially regarding "new" pollutants. The objectives of this study were to assess serum levels of the persistent 2,2',4,4',5,5'-hexachlorobiphenyl (CB-153), hydroxylated polychlorinated biphenyl metabolites (OH-PCBs), polybrominated diphenyl ethers (PBDEs), and hexabromocyclodode-canes (HBCDDs) in a group of Swedish middle-aged and elderly women expected to be relatively highly exposed, and to evaluate the impact of potential determinants (e.g., fish intake, age) for the inter-individual variation, as well as to investigate the association between these pollutants and bone density. No associations were found between bone mineral density or biochemical markers of bone metabolism and the analyzed environmental pollutants. Relatively high levels of CB-153 (median 260 ng/g fat) and sigma 3-OH-PCBs (median 1.7 ng/mL serum), and low concentrations of sigma6PBDEs (median 3.6 ng/g fat) were determined. Total level of HBCDDs in serum was quantified by gas chromatography with mass spectrometric detection (median 0.5 ng/g fat). HBCDD diastereomeric and enantiomeric patterns were determined by liquid chromatography with mass spectrometric detection. The dominating stereoisomer was (-)alpha-HBCDD, but 1-3% of gamma-HBCDD was also detected in the serum samples.     

Use of Monte Carlo analysis to characterize nitrogen fluxes in agroecosystems

Intensive agricultural systems are largely responsible for the increase in global reactive nitrogen compounds, which are associated with significant environmental impacts. The nitrogen cycle in agricultural systems is complex and highly variable, which complicates characterization in environmental assessments. Appropriately representing nitrogen inputs into an ecosystem is essential to better understand and predict environmental impacts, such as the extent of seasonally occurring hypoxic zones. Many impacts associated with reactive nitrogen are directly related to annual nitrogen loads, and are not adequately represented by average values that de-emphasize extreme years. To capture the inherent variability in agricultural systems, this paper employs Monte Carlo analysis (MCA) to model major nitrogen exports during crop production, focusing on corn-soybean rotations within the U.S. Corn Belt. This approach yields distributions of possible emission values and is the first step in incorporating variable nutrient fluxes into life cycle assessments (LCA) and environmental impact assessments. Monte Carlo simulations generate distributions of nitrate emissions showing that 80% of values range between 15 and 90 kg NO39-) N/ha (mean 38.5 kg NO3(-) N/ha; median 35.7 kg NO3(-) N/ha) for corn fields and 5-60 kg NO3(-) N/ha (mean 20.8 kg NO3(-) N/ha; median 16.4 kg NO3(-) N/ha) for soybean fields. Data were also generated for grain and residue nitrogen, N2O, NO(x), and NH3. Results indicate model distributions are in agreement with available measured emissions.     

Model prognoses for future acidification recovery of surface waters in norway using long-term monitoring data

During the past 20 years, acid deposition in Europe has decreased by more than 60%, yet still a large number of lakes and streams in southern Norway have not recovered to a water quality sufficient to support sustainable populations of trout or salmon. Long-term (30 years) monitoring data were used hereto constrain the calibration of the acidification model MAGIC to three Norwegian calibrated catchments. The model accounted for 60-80% of the variance in the year-to-year variations in concentrations of most of the major ions in streamwater. The results support the use of the lumped parameter acid neutralizing capacity (ANC) to link chemical parameters to biological response, as the calibration efficiency for ANC is considerably higher than for other biologically important parameters such as inorganic aluminum (Al(n+)) and pH. Three different scenarios for future deposition of sulfur were run: current legislation, maximum feasible reductions, and an illustrative scenario removing all anthropogenic deposition. These analyses show that much of the potential improvement in water quality has already occurred and that only limited further improvement can be expected from the current legislation. The current legislation is unlikely to produce ANC values sufficiently high to allow self-reproducing populations of trout at two of the three sites. Most of the response in water chemistry to reduced acid deposition has been rapid; the water chemical responses largely occur the same year or a few years after reduction in the input. The soil pool of exchangeable base cations depleted during 150 years of acid deposition, however, requires several centuries for replenishment. The uncertainties in future predictions come from several factors, such as future nitrogen dynamics and impacts from changes in seasalt and precipitation events. The differences in future water chemistry predicted from changed seasalt deposition or nitrogen dynamics are larger that the differences between different deposition scenarios. Hence, these factors must be included in future assessments of recovery from acidification.     

Assessing long-term pH change in an Australian river catchment using monitoring and palaeolimnological data

Reviews of stream monitoring data suggest that there has been significant acidification (>1.0 pH unit at some sites) of Victorian streamwaters over the past 3 decades. To assess whether these declines are within the range of natural variability, we developed a diatom model for inferring past pH and applied it to a ca. 3500-yr diatom record from a flood plain lake, Callemondah 1 Billabong, on the Goulburn River, which has among the most substantial observed pH declines. The model has a jackkniffed r2 between diatom inferred and measured pH of 0.77 and a root mean square error of prediction of 0.35 pH units. In the pre-European period, pH was stable (range 6.5-6.7) for approximately 3000 yr. Since European settlement around 160 yr ago, diatom-inferred billabong pH has increased significantly by >0.5 units. We hypothesize that this increase in pH is related to processes associated with land clearance (e.g., increased base cation load and decreased organic acid load). There is no evidence of the recent monitored declines in the Callemondah record, which may indicate that that flood plain lakes and the main stream are experiencing divergent pH trends or that the temporal resolution in the billabong sediment record is insufficient to register recent declines.     

Inhibitory activity of phosphates on molds isolated from foods and food processing plants

Six commercial phosphates were evaluated for inhibition of the growth of 17 molds isolated from food sources. The assays were performed at neutral and natural (without pH adjustment) pH values, and the molds were streaked on plate count agar with added phosphates. Phosphate concentrations of 0.1, 0.3, 0.5, 1.0, and 1.5% (wt/vol) were used, and the MIC was determined. The resistance of molds to phosphates depended on the species. At a neutral pH, Aspergillus ochraceus and Fusarium proliferatum were resistant to all phosphates at all concentrations assayed, and Byssochlamys nivea, Aureobasidium pullulans, and Penicillium glabrum were most sensitive. The most inhibitory phosphates were those with chain lengths greater than 15 phosphate units and the highest sequestering power. At natural pH values (resulting from dissolving the phosphate in the medium), inhibitory activity changed dramatically for phosphates that produced alkaline or acidic pH in the medium. Phosphates with alkaline pH values (sodium tripolyphosphate of high solubility, sodium tripolyphosphate, and sodium neutral pyrophosphate) were much more inhibitory than phosphates at a neutral pH, but sodium acid pyrophosphate (acidic pH) had decreased inhibitory activity. The results indicate that some phosphates could be used in the food industry to inhibit molds linked to food spoilage.     

Retrospective analyses and future predictions of snowmelt-induced acidification: example from a heavily impacted stream in the Czech Republic

We have combined a long-term hydrochemistry model (MAGIC) with a model that predicts short-term transient changes in hydrochemistry (pBDM) during hydrological events in order to improve the temporal resolution of retrospective analyses and future predictions of streamwater acidification. The model has been applied to a heavily impacted catchment in the Czech Republic. Spring flood acid-neutralizing capacity (ANC), pH, and inorganic monomeric aluminum (Ali(n+)) were simulated for the years of 1860, 1900, 1930, 1950, 1965, and 1985, measured in 1999, and predicted for 2030 using two different emission control scenarios. If the emission reduction according to the current legislation scenario is implemented, the model predicts that the spring flood pH, ANC, and Ali(n+) will recover close to the level of the 1950s by 2030. This will occur despite the annual average chemistry being farfrom having recovered to that level. The results suggest that the recovery of spring flood events is faster then the recovery of annual average chemistry and that much of what is won by further emission reduction will not be fully realized on an annual time scale.     

On the construction, comparison, and variability of airsheds for interpreting semivolatile organic compounds in passively sampled air

Air mass origin as determined by back trajectories often aids in explaining some of the short-term variability in the atmospheric concentrations of semivolatile organic contaminants. Airsheds, constructed by amalgamating large numbers of back trajectories, capture average air mass origins over longer time periods and thus have found use in interpreting air concentrations obtained by passive air samplers. To explore some of their key characteristics, airsheds for 54 locations on Earth were constructed and compared for roundness, seasonality, and interannual variability. To avoid the so-called "pole problem" and to simplify the calculation of roundness, a "geodesic grid" was used to bin the back-trajectory end points. Departures from roundness were seen to occur at all latitudes and to correlate significantly with local slope but no strong relationship between latitude and roundness was revealed. Seasonality and interannual variability vary widely enough to imply that static models of transport are not sufficient to describe the proximity of an area to potential sources of contaminants. For interpreting an air measurement an airshed should be generated specifically for the deployment time of the sampler, especially when investigating long-term trends. Samples taken in a single season may not represent the average annual atmosphere, and samples taken in linear, as opposed to round, airsheds may not represent the average atmosphere in the area. Simple methods are proposed to ascertain the significance of an airshed or individual cell. It is recommended that when establishing potential contaminant source regions only end points with departure heights of less than ～700 m be considered.     

Reliabilities of genomic prediction using combined reference data of the Nordic Red dairy cattle populations

This study investigated the possibility of increasing the reliability of direct genomic values (DGV) by combining reference populations. The data were from 3,735 bulls from Danish, Swedish, and Finnish Red dairy cattle populations. Single nucleotide polymorphism markers were fitted as random variables in a Bayesian model, using published estimated breeding values as response variables. In total, 17 index traits were analyzed. Reliabilities were estimated using a 5-fold cross validation, and calculated as the within-year squared correlation between estimated breeding values and DGV. Marker effects were estimated using reference populations from individual countries, as well as using a combined reference population from all 3 countries. Single-country reference populations gave mean reliabilities across 17 traits of 0.19 to 0.23, whereas the combined reference gave mean reliabilities of 0.26 for all populations. Using marker effects from 1 population to predict the other 2 gave a loss in mean reliability of 0.14 to 0.21 when predicting Swedish or Finnish animals with Danish marker effects, or vice versa. Using Swedish or Finnish marker effects to predict each other only showed a loss in mean reliability of 0.03 to 0.05. A combined Swedish-Finnish reference population led to an average reliability as high as that from the 3-country reference population, but somewhat different for individual traits. The results from this study show that it is possible to increase the reliability of DGV by combining reference populations from related populations.     

Investigations on the trypsin inhibitor,urease and cooking behaviour of soybean Glycine max Merr.

Eleven varieties of soybeans were analysed for their trypsin inhibitor activity (TIA), urease activity and cooking behaviour. The TIA ranged between 8·1 and 38·5 trypsin units inhibited per mg meal. In general, black soybeans contain higher trypsin inhibitor activities than yellow varieties. There was no appreciable variation in urease activity whose values ranged between 1·93 and 2·12 pH units. Considerable variability was evident in cooking behaviour of soybeans. The cooking period ranged between 140 min and 318 min. Thus the present study suggests the possibility of lowering trypsin inhibitor activity and cooking periods by further selection and breeding programmes of soybean cultivars.