This paper analyzes day-of-week variations in concentrations of particulate matter (PM) in California. Because volatile organic compounds (VOCs) and oxides of nitrogen (NO x ) are not only precursors of ozone (O 3 ) but also of secondary PM, it is useful to know whether the variations by day of week in these precursors are also evident in PM data. Concentrations of PM Յ10 m (PM 10 ) and Յ2.5 m in aerodynamic diameter (PM 2.5 ) were analyzed.

Although networks of environmental monitors are constantly improving through advances in technology and management, instances of missing data still occur. Many methods of imputing values for missing data are available, but they are often difficult to use or produce unsatisfactory results. I-Bot (short for "Imputation Robot") is a context-intensive approach to the imputation of missing data in data sets from networks of environmental monitors. I-Bot is easy to use and routinely produces imputed values that are highly reliable. I-Bot is described and demonstrated using more than 10 years of California data for daily maximum 8-hr ozone, 24-hr PM 2.5 (particulate matter with an aerodynamic diameter <2.5 μm), mid-day average surface temperature, and midday average wind speed. I-Bot performance is evaluated by imputing values for observed data as if they were missing, and then comparing the imputed values with the observed values. In many cases, I-Bot is able to impute values for long periods with missing data, such as a week, a month, a year, or even longer. Qualitative visual methods and standard quantitative metrics demonstrate the effectiveness of the I-Bot methodology.Implications: Many resources are expended every year to analyze and interpret data sets from networks of environmental monitors. A large fraction of those resources is used to cope with difficulties due to the presence of missing data. The I-Bot method of imputing values for such missing data may help convert incomplete data sets into virtually complete data sets that facilitate the analysis and reliable interpretation of vital environmental data. PAPER HISTORY

California's Phase 2 Reformulated Gasoline (CaRFG), introduced early in 1996, represents an important step toward attainment of ozone standards. Studies of vehicle emissions and ambient air quality data have reported substantial reductions of ozone precursors due to CaRFG. This study uses daily measurements of regional ozone and meteorology to estimate the effect of CaRFG on ozone concentrations in three areas of California. In each area, a regression model was used to partially account for the daily effects of meteorology on area-wide ozone maxima for May-October. The statistical models are based on combinations of air temperature aloft (~5000 ft), surface air temperatures, and surface wind speeds. Estimated ozone benefits were attributed to CaRFG after accounting for meteorology, which improved the precision of the estimates by approximately 37-57% based on a resampling analysis. The ozone benefits were calculated as the difference in ozone times the proportion of the reductions of hydrocarbons and nitrogen oxides attributed to CaRFG by the best available emission inventories. Ozone benefits attributed to CaRFG (with 90% confidence) are 8-13% in the Los Angeles area, -2-6% in the San Francisco Bay area overall with greater benefits in two major subregions, and 3-15% in the Sacramento area. IMPLICATIONSBecause ozone formation in the lower troposphere is a complex process, the future effects of alternative regulations designed to reduce ozone precursors can be somewhat uncertain. Decision-makers are frequently limited to the results of emissions tests and air quality simulation models as the basis for prioritizing alternative emission control measures. The analysis presented here provides an additional, empirical evaluation of the ozone-reducing benefits of California's reformulated gasoline. The results may help in determining whether California's fuel specifications are adopted in other parts of the country.

A herbicide phytotoxicity model was used to investigate differences in herbicide tolerance between two species of Amsinckia (A. intermedia Fischer and Meyer and A. gloriosa Suksdorf) to the herbicide bromoxynil (3,5‐dibromo‐4‐hydroxybenzonitrile). The model included two primary parameters that predicted the plant's maximum response and the response rate to a given herbicide level. The models for the two species were compared at each herbicide rate (0.06, 0.12, 0.25, and 0.50 kg ai/ha), and the modeled reaction of the two species to the herbicide was significantly different at all but the lowest herbicide rate. Fitted curves were compared using a t‐test for each model coefficient and alternatively by using a multivariate statistic that considers the joint behavior of the coefficients in the model. Zusammenfassung Ein Modell für die Pflanzentoxizität von Herbiziden zur Einschätzung von Herbizid‐toleranzen Ein mathematisches Modell für die Pflanzentoxizität von Herbiziden wurde zur Unter‐suchung von Unterschieden in der Herbizid‐toleranz von zwei Arten von Amsinckia (A. intermedia Fischer & Meyer and A. gloriosa Suksdorf) gegenüber dem Herbizid Bromoxy‐nil (3,5‐dibromo‐4‐hydroxybenzonitrile) angewandt. Das Modell basiert auf zwei Hauptparametern, die es ermöglichen, die maximale Reaktion und die Reaktionsrate vor‐herzusagen. Die zwei Arten wurden für ver‐schiedene Herbiziddosen (0,06, 0,12, 0,25, und 0,50 kg/ha) mit Hilfe des Modells ver‐glichen. Bezüglich aller Level, ausgenommen dem niedrigsten, zeigten sich signifikante Unterschiede. Der Vergleich der Kurven wurde sowohl mittels t‐test für jeden Koef‐fizienten als auch mittels multivarianter Statis‐tik durchgeführt, die das Verhalten beider Koeffizienten gleichzeitig berücksichtigte.

A 3-yr field experiment evaluated average daily (ADG) and total seasonal (LG) weight gains of growing beef cattle (Bos taurus) on fertilized California foothill annual range. Soils were a mixture of four series of alfisols. Nitrogen was applied alone at 45 or 90 kg ha-•, P and S were applied together at 34P 37S or 67P 74S kg ha-•, and the two N rates were combined with 34P 37S. Two 13.2-ha replications were stocked with 215-kg steers in late November, and stocking rates were increased twice during a grazing season averaging 190 d. Forage on offer (FL) varied from 400 to 2500 kg ha-•. Annual forage production (FP) varied from 2500 to 5700 kg ha -• and differed (P<0.05) between treatments and years. Year-to-year and within-year variations in rainfall influenced FP. Average daily gains over the season varied linearly from less than 0.2 to over 1.0 kg steer-•. Differences in ADG due to treatments were largely restricted to the first year and to treatments with N, P, and S or the higher level of PS, but 3-yr trends were not significant. Average daily gain was not strongly influenced by either FL or legume content of the forage. Three-year LG was highest for the two NPS and the higher PS treatment. Fertilization increased forage organic matter digestibility and N concentrations over the control. Results indicate that either N, P, and S or a sufficient level of P and S to stimulate production of legume N was required for high forage production.