Various epidemiological studies have linked exposure to Ultrafine Particles (UFP; diameter< 100 nm) to adverse health impacts. Roadway traffic is one of the major sources of UFPs and heavily influences UFP concentrations in the nearby vicinity of major roadways. Modeling efforts to predict UFPs have been limited due to the scarcity of reliable information on emissions, lack of monitoring data and limited understanding of complex processes affecting UFP concentrations near sources. In this study continuous measurement of ultrafine particle number concentrations (PNC) and mass concentrations of nitric oxide (NO), nitrogen dioxide (NO2) and PM2.5 was conducted near an arterial road and freeway at different seasons and meteorological conditions and integrated with traffic count data. PNC showed high correlation with NO (r=0.64 for arterial; 0.61 for freeway), NO2 (r=0.57 for arterial; 0.53 for freeway) and NOx (NOx=NO+NO2; r=0.63 for arterial; 0.59 for freeway) and moderate to low correlation with traffic volume (r=0.33 for arterial; 0.32 for freeway) and PM2.5 (r=0.28 for arterial; 0.23 for freeway); respectively; for both sites at 15 minute averages. The PNC-NOx relationship prevailed on a shorter term (15 min), hourly, and throughout the day basis. Both PNC and NOx showed comparatively higher correlation with traffic during the morning period but became lower during evening which can be attributed to the higher boundary layer and wind speeds. The variable meteorology in the evening affects both PNC and NOx concentrations in the same way and the correlation between NOx and PNC is maintained high both during morning (r=0.74 for arterial; 0.69 for freeway), and evening (r=0.62 for arterial; 0.59 for freeway) periods. Thus nitrogen oxides can be used as a proxy for traffic-related UFP number concentration reflecting the effect of both traffic intensity and meteorological dilution. The PNC-NOx relation was explored for various meteorological parameters i.e. wind speed and temperature. It is found that NOx emission is temperature independent and can be used to reflect the effect of traffic intensity and meteorological dilution. Once the effect of traffic intensity and dilution is removed, the effect of temperature on PNC-NOx ratio becomes important which can be attributed to the variation in PNC emission factors with temperature. The high morning PNC-NOx ratio found at the arterial road is a result of new particle formation due to lower temperature and low concentration of exhaust gases in the morning air favoring nucleation over condensation. This finding has important implication when calculating emission factors for UFP number concentrations. Thus it can be concluded that roadside concentration of ultrafine particles not only depends on traffic intensity but also on meteorological parameters affecting dilution or new particle formation. High concentrations of ultrafine particle number concentration close to a roadway is expected due to higher traffic intensity , as well as during low wind speed causing low dilution and low temperature conditions favoring new particle formation. Finally a simplified approach of calculating particle number emission factor was developed using existing and easily available emission inventory for traffic related tracer gases. Using NOx emission factors from MOVES emission model, the emission ratio of PNC to NOx was converted to develop particle number emission factors. NOx was selected as the traffic related tracer gas since the number concentration of particles is closely correlated to NOx, NOx and particles are diluted in the same way and NOx emission factors are available for a variety of traffic situations. To ensure contribution of fresh traffic exhaust, the average of the difference of pollutant concentrations at high traffic condition and background condition was used to calculate PNC-NOX ratio. Using nitrogen oxides to define background and high-traffic conditions and MOVES emission factor for NOX to convert corresponding PNC-NOX ratio, an average emission factor of (1.82 ± 0.17) E+ 14 particle/ vehicle-km was obtained, suitable for summertime. When compared to existing particle number emission factors derived from dynamometer tests, it was found that there exits reasonable agreement between the calculated real world particle number emission factors and emission factors from dynamometer tests. The calculated emission factor and R-Line dispersion model was tested in predicting near-road particle number concentrations. Although only 23% of the variability in PNC was explained by the dispersion model, 84.33% of the measurements fell within the factor of two envelope. This suggests that there is potential to effectively use these models and thus warrants more in-depth analysis. Finally, a simple map of PNC gradients from major roads of Portland was developed. The results of this study helped identify proxy-indicators to provide reference values for estimating UFP concentrations and emissions that can be used for simple evaluation of particle concentration near major roadways for environmental and urban planning purposes and to assess expected impact of UFP pollution on population living near roadways exposed to elevated concentrations.

The effects of roads on landscapes and wildlife and their ecological processes are substantial and represent a major anthropogenic disruption to the natural environment. Current understanding of the impacts of roads for their strategic management is hampered by a lack of information on 1) the influence of the ecosystems on road effects, 2) the effects of roads on higher-order ecosystem responses (populations and communities), and 3) the overall impacts of roads on ecosystems and their wildlife (on both abiotic and biotic ecosystem components). This study used the Silver City Highway in arid New South Wales, Australia, as a model for a typical road in an arid ecosystem to investigate the ecological effects of arid-zone roads and their management. The study examined the spatial variation of soil, vegetation, kangaroo and small mammal variables in relation to the road, explored the factors contributing to these respective spatial variations, investigated the effects of arid-zone road management on vegetation, assessed the fragmentation effect of the arid-zone road on small mammals, and determined the patterns, causes and effects of kangaroo-vehicle collisions. The study revealed that the arid-zone road influenced most of the variables measured, skewed the population demographics of two kangaroo species, and altered the community composition of small mammals, with two small mammal species listed as threatened in NSW and of national significance (Sminthopsis macroura and Leggadina forresti) negatively impacted by the road. Increases in the amount of water along the road edge drove many of the subsequent effects of the arid-zone road as arid-zone flora and fauna are adapted to exploiting limited and patchily distributed resources. However, current arid-zone road management also influenced vegetation quality, microclimates around the road influenced kangaroo densities and small mammal communities, and kangaroo flight behaviour and temporal variations in traffic volume affected roadkill frequency. Together, these results suggest that roads have a high overall ecological impact in arid ecosystems. Conservation managers need to first rank the impacts of arid-zone roads in order of their conservation importance and need for mitigation, and from there, devise relevant informed management frameworks to target these impacts.

Unpaved forestry roads can significantly affect surface runoff and sediment production, with consequential impacts for stream water quality. The potential impact of road runoff on stream water quality is mitigated by the redistribution of runoff into the forest compartments through road drains. The objective of this study was to assess runoff and the associated nutrient loads from unpaved forest access roads, and to evaluate the effectiveness of road runoff redistribution onto the forest compartments. Unpaved road segments in Mondi Forest Plantation in Seele Estate, New Hanover, South Africa were instrumented for runoff measurement in response to natural rainfall. Two road segment classes were investigated for water quality from unbounded runoff plots: steep sloped road segments of road gradients of 9.5° and 7.5°, and gentle sloped road segments of road gradients of 1.6° and 2.0°. Water quality was also assessed by monitoring road runoff, and stream water quality was analysed for water quality parameters including; pH, Nitrates, Nitrites, Phosphates, Total dissolved Oxygen, Oxygen consumption, Ammonium and temperature upstream and downstream of the Estate. The effectiveness of road runoff redistribution into the forest compartments was evaluated through relating water distribution to tree breast height diameter. Two sets of road drains corresponding to the plots of different road gradients were selected as for runoff, and sampled, and corresponding plots or allotments were established to determine tree breast height diameter measurements. The results of the study revealed that, as might have been expected, runoff production increases with the increasing road gradient. The quality of road runoff water was lower than the stream water. There were no significant differences observed in nutrient levels upstream and downstream of the road stream crossings. The nutrient concentrations however, were higher upstream of the estate than downstream. Significant differences in tree breast height diameter were noted between plots of different road gradients. This suggested that the gradient determines the infiltration of redistributed runoff and hence the availability of the water that can be used by the trees within a compartment. The results of the study suggest that unpaved roads are important in the generation of nutrient loads. Much of the nutrient value is redistributed within the compartment itself rather than being transferred to the stream. This suggests that, provided that road runoff can be contained within the compartments, the potentially negative impact of road runoff can be mitigated and may enhance tree growth. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2011.

M.Sc. / The purpose of this study was to determine whether unpaved roads in commercial forests have a detrimental impact on aquatic macro-invertebrates in the receiving rivers associated with these roads. The upper section of the headwaters of the Motlhaka-Semeetse River was chosen as the Study Area because a portion is situated in the Wolkberg Wilderness area (natural area with no roads) and a portion in a commercial forestry plantation (high road network density). Stream conditions based on a rapid bio-assessment of aquatic macro-invertebrates using SASS5 were then compared between the two catchments amid 2002 and 2010. In order to support the SASS5 results, daily rainfall data from December 1959 to October 2010 was collated and compared to daily stream flow data from January 1960 to October 2010. In addition, geomorphological and instream habitat changes in the river between the two catchment areas were recorded, as well as the condition of stream crossings and their linkage to stream networks in the afforested catchment. The SASS5 results revealed that stream conditions at the upper unimpaired site were significantly better, visibly and statistically (p < 0.05) than conditions at the lower site within the forestry plantation. In addition, there were noteworthy changes in the catchment hydrology, the geomorphology and instream habitat between the natural and afforested catchments. Road network densities in commercial forests are mostly considerably higher than suggested in literature and exceed stream network densities. Roads increase the surface area for interception of rainfall and the runoff from this high density of roads results in modification of the catchment hydrology, geomorphology and instream habitat of receiving streams. This physical change to the receiving streams is one of the main reasons for the deterioration in SASS5 results, disproving the hypothesis that road networks in forestry areas have no impact on receiving aquatic ecosystems. Commercial forestry in South Africa needs to improve their road planning, layout, management and maintenance to reduce these environmental impacts. In doing so, the road network density will be reduced with considerable environmental and economic benefits.

Scenic corridor protection can be viewed as a microcosm of all land use and protection concerns. The issue has been a topic of discussion among a select few land use professionals, including landscape architects, highway engineers and land use lawyers, for over fifty years. However, the acceleration of land development into previously undisturbed scenic areas, coupled with recent federal initiatives aimed at the inclusion of scenic byways programs into the 1991 Transportation Bill, has increased discussions on scenic byway issues, including questions on the success or failures of scenic corridor protection. To date, there has been little research on the issues and concerns of scenic corridor management and protection, especially at the state level. This thesis is an attempt to gauge the status of scenic corridor protection among those states with several years experience in the field and to offer other states some insights as to tools, issues and concerns involved with corridor management and protection. The process was achieved through a combination of telephone interviews, a questionnaire and case studies of each participant state. Indications are that the field is currently fragmented in terms of tools and methods of management and many managers feel a sense of frustration at the lack of a holistic approach to scenic corridor protection. / Master of Landscape Architecture

As replacement and removal of undersized culverts gains momentum as an effective technique for restoring natural stream flows and removing fish passage barriers, it is important to evaluate the benefits of these efforts on the in-stream and adjacent riparian habitat for other species of potential concern. This study compares stream-associated amphibian (SAA) occurrence in streams adjacent to different road crossing structures on unpaved forest roads in the Wilson River watershed located within the Tillamook State Forest, Oregon. Surveys were conducted at road crossing structures for three taxa of SAA; Pacific giant salamander (Dicamptodon tenebrosus), coastal tailed frog (Ascaphus truei), and Columbia torrent salamander (Rhyacotriton kezeri). Statistical models were created to analyze the effect of habitat variables on SAA occupancy, and determine whether those variables changed relative to road crossing structures. Results showed that coastal tailed frog occupancy was positively associated with dissolved oxygen and crayfish presence, negatively associated with longitude, and had a slight quadratic relationship to channel confinement. Because all sites were highly saturated with oxygen, percent dissolved oxygen was likely a surrogate for one or more other covariates, such as in-stream habitat or substrate size. Detection rates of Pacific giant salamander were too high and detection rates of Columbia torrent salamander were too low to provide reliable models, but they did provide some insight into the factors affecting occupancy in the study area, including information about their interactions with road crossing structures. Occupancy rates of Pacific giant salamanders in the study area appear to be unaffected by road crossings, fish passable or otherwise. Conversely, torrent salamanders seemed to be mostly absent from the study sites altogether, but based on the one stream where they were detected, they are not excluded from occupying fish barrier culvert sites. Top weighted habitat covariates, including dissolved oxygen, channel confinement, crayfish and fish presence, and flow constriction are all indicators of disturbance that can be linked to road crossings. Although the size and structure of road crossings did not prove to significantly influence SAA occupancy, indicators of stream disturbance that can be related to road crossings were included in the best models for predicting occupancy, demonstrating the importance of reducing disturbance related to road crossings on streams.