In recent years, green roofs have become an accepted solution in ecological urban design to mitigate the impacts of impervious surfaces (Berghage, Beattie, Jarrett, Thuring, & Razaei, 2009). An experimental research project was conducted at the Mississippi Agriculture and Forestry Experiment Station (MAFES) Green Infrastructure Research Area at South Farm of Mississippi State University to determine how medium depth and slope gradient on rooftops affect plant cover and survival. Plant cover was monitored monthly by photographing the experimental green roof platforms. Photoshop and AutoCAD software programs were employed to digitize and to calculate plant cover from the images. All recorded data was analyzed with Analysis of Variance (ANOVA) tests. It was determined that the effects of medium depth and slope are statistically significant on plant cover and survival.

Water quality and nonpoint source (NPS) pollution are important issues in many areas of the world, including the Greater Toronto Area where urban development is changing formerly rural watersheds into impervious surfaces. Impervious surfaces (i.e. roads, sidewalks, parking lots, strip malls, building rooftops, etc.) made out of impenetrable materials directly impact hydrological attributes of a watershed. Therefore, understanding the degree and spatial distribution of impervious surfaces in a watershed is an important component of overall watershed management. According to Environment Canada’s estimates, road salts, also considered nonpoint source pollutants, represent the largest chemical loading to Canadian surface waters. The main objective of this study is to verify the often assumed correlation between impervious surfaces and chlorides that result from the application of road salts, focusing on a case study in the selected six major watersheds within the Greater Toronto Area. In this study, Landsat-5 TM images from 1990, 1995, 2000, and 2005 were used in mapping urban impervious surface changes within the study area. Pixel-based unsupervised classification technique was utilized in estimation of percentage impervious surface coverage for each watershed. Chloride concentrations collected at Water Quality Monitoring Stations within the watersheds were then mapped against impervious surface estimates and their spatiotemporal distribution was assessed. In a GIS environment, remotely sensed impervious surface maps and chloride maps were overlaid for the investigation of their potential correlation. The main findings of this research demonstrate an average of 12.9% increase in impervious surface areas as well as a three-fold increase in chloride concentrations between 1990 and 2005. Water quality monitoring stations exhibiting the highest amounts of chloride concentrations correspond with the most impervious parts of the watersheds. The results also show a correlation (coefficient of determination of 0.82) between impervious surfaces and chloride concentrations. The findings demonstrate that the increase in imperviousness do generate higher chloride concentrations. Correspondingly, the higher levels of chloride can potentially degrade quality of surface waters in the region. Through an innovative integrated remote sensing approach, the study was successful in identifying areas most vulnerable to surface water quality degradation by road salts.

Water quality and nonpoint source (NPS) pollution are important issues in many areas of the world, including the Greater Toronto Area where urban development is changing formerly rural watersheds into impervious surfaces. Impervious surfaces (i.e. roads, sidewalks, parking lots, strip malls, building rooftops, etc.) made out of impenetrable materials directly impact hydrological attributes of a watershed. Therefore, understanding the degree and spatial distribution of impervious surfaces in a watershed is an important component of overall watershed management. According to Environment Canada’s estimates, road salts, also considered nonpoint source pollutants, represent the largest chemical loading to Canadian surface waters. The main objective of this study is to verify the often assumed correlation between impervious surfaces and chlorides that result from the application of road salts, focusing on a case study in the selected six major watersheds within the Greater Toronto Area. In this study, Landsat-5 TM images from 1990, 1995, 2000, and 2005 were used in mapping urban impervious surface changes within the study area. Pixel-based unsupervised classification technique was utilized in estimation of percentage impervious surface coverage for each watershed. Chloride concentrations collected at Water Quality Monitoring Stations within the watersheds were then mapped against impervious surface estimates and their spatiotemporal distribution was assessed. In a GIS environment, remotely sensed impervious surface maps and chloride maps were overlaid for the investigation of their potential correlation. The main findings of this research demonstrate an average of 12.9% increase in impervious surface areas as well as a three-fold increase in chloride concentrations between 1990 and 2005. Water quality monitoring stations exhibiting the highest amounts of chloride concentrations correspond with the most impervious parts of the watersheds. The results also show a correlation (coefficient of determination of 0.82) between impervious surfaces and chloride concentrations. The findings demonstrate that the increase in imperviousness do generate higher chloride concentrations. Correspondingly, the higher levels of chloride can potentially degrade quality of surface waters in the region. Through an innovative integrated remote sensing approach, the study was successful in identifying areas most vulnerable to surface water quality degradation by road salts.

This research conducted a GIS analysis of twenty-eight sub-watersheds in Massachusetts and considered five factors which the literature suggested influenced base flow. The literature suggested a positive relation between precipitation, forest cover and base flow and a negative relationship between impervious surfaces, soil drainage class, slope and base flow. A multiple regression analysis of the sub-watershed information confirmed the literature for two factors (soil drainage class and slope) and questioned it on three factors (impervious surfaces, precipitation, and forest cover). The resulting predictive equation indicated that imperviousness and precipitation were the most significant factors affecting base flow. The first derivative of the predictive equation indicated that the interaction effects between these factors had a substantial impact on the base flow values. A number of land use impacts were also considered in this research. Natural land use features such as open land, wetland, forest cover, agricultural, and recreation uses were found to promote the infiltration and increase base flow. These uses promote base flow because their surfaces are more pervious than unnatural surfaces such as high density/multi-family residential, industrial and commercial areas which are mostly covered by impervious surfaces reducing base flow. The research also used a case study of two Massachusetts sub-watersheds to consider the effectiveness of structural stormwater Best Management Practices for promoting base flow. Two sub-watersheds were considered: the Neponset where stormwater BMPs had been implemented and Quinsigamond that had not implemented them. The case study results suggested that structural stormwater BMPs were effective in increasing base flow.

The rapid expansion of urban systems in both area and population represents the most significant landuse/landcover change occurring in the world today. Urbanization is often accompanied by increasing environmental degradation. This degradation is related to stormwater runoff, air temperatures greater than surrounding rural areas, increased air and water pollution, losses of vegetated lands, and lack of access to sufficient and healthy foods in lower-income areas. Urban agriculture (UA), a practice long established in previous eras but neglected for many decades, can mediate such concerns by providing greenspaces to improve ecosystem services. Successful practice of UA requires recognition of interactions between social and environmental patterns. Neglect of these interactions leads to failure in spatially integrating social and environmental dimensions of the urban landscape, limiting the success of UA. This study investigates siting of UA within Roanoke, Virginia, a compact urban region characterized by social and environmental conditions that can be addressed by effective siting and practice of UA. This research takes a broader perspective than prior studies on UA and urban greenspaces. It proposes innovative applications of geospatial technologies for urban assessment. Studies on UA have typically focused on food insecurity, while studies on greenspaces focus on parks and tree canopy cover, without investigating interactions that promote synergies between these two efforts. Research over the past few years is now recognizing potential contributions for urban agriculture to alleviate environmental issues such as stormwater runoff, soil infertility, and the urban heat island effect. Little of this research has been devoted to the actual siting of urban agriculture to specifically alleviate both socio-economic and environmental issues. This research applies geospatial technologies to evaluate spatial patterns characterizing both environmental and socio-economic disparities within the City of Roanoke, Virginia. This approach has identified specific locations that are open and available for urban agriculture, and has appraised varying levels of socio-economic and environmental parameters. This research identified, at the census block group level, areas with varying levels of degradation. Thus, those locations in which a new urban agriculture greenspace can contribute to both socio-economic and environmental reparation. This research has identified spatial dimensions in which UA will assist in restoring ecosystem services to guide various food production activities. These results can be generalized to other urban locations and contribute to efficient use of urban land and space, improving the three pillars of worldwide sustainability – economic, environment, and social. / Ph. D.

Severe flooding events in recent years have underlined the importance of accurate hydrological modelling in urban areas. There are many important parameters relating to both the rainfall distribution and properties of the land on which the rain falls that controls the impacts of the rain event. While the importance of input parameters such as initial water content, topography and extent of hardened surfaces is widely known, their impact on hydrological response in urban areas is not thoroughly understood. In 2017, scientists from the Swedish Meteorological and Hydrological Institute, SMHI, presented five empirical hyetographs based on long-term Swedish municipal rain data (Olsson, et al., 2017). The hyetographs vary in location of peak and distribution of intensity. Olsson (2019) evaluated the hydrological response in terms of water depth of the hyetographs in relation to Chicago Design Storm (CDS), a common design storm used in Swedish hydrological modelling, and found that CDS nearly always overestimates the hydrological response in comparison to the empirical hyetographs, meaning historical rainfall intensity distributions, developed by SMHI.  The aim of this thesis is to analyse the robustness of empirical hyetographs by conducting a sensitivity analysis of a generic urban model with a variation of input parameters. A statistical analysis of data on hardened surfaces, topography and initial water content was conducted to find the median, 5th and 95th percentile respectively of the range of values in Swedish conditions. These values were applied to a synthetic urban model and run together with the five empirical hyetographs and CDS in MIKE 21. The results indicate that the empirical hyetographs are not very robust relating to variations in infiltration capacity (initial water content and amount of hardened areas), while they are more robust when varying the topography. The variation of topography also resulted in large variations in water depth, time to peak and extent of flooded area, while variation of initial water content and amount of hardened surfaces had smaller, although still clear, effects. Furthermore, the results show that hyetographs with a late peak are more sensitive to variations in initial water content and hardened surfaces than hyetographs with an early peak. On average, CDS estimates the response in comparison to the empirical hyetographs accurately, with an average overestimation of 0.5%, but does not capture the range and complexity of the empirical hyetographs. Including the frequency of the different hyetographs, CDS gave an overestimation of 5% compared to the empirical hyetographs. The highlight the uncertainty in using CDS as input for urban cloudburst modelling, but limitations in form of catchment properties in a study area and the importance of the input values limits the generability of the study. / De senaste årens allvarliga översvämningshändelser har understrukit vikten av noggrann hydrologisk modellering i stadsområden. Även om betydelsen av parametrar som initial vattenmättnad, topografi och andelen hårdgjorda ytor är allmänt känd, är deras inverkan på urban hydrodynamisk modellering inte helt klarlagd. Under 2017 presenterade forskare från Sveriges Meteorologiska och Hydrologiska Institut, SMHI, fem empiriska hyetografer baserade på långvariga kommunala regndata (Olsson et al., 2017). Hyetograferna varierar i fördelning av intensitet och tidsmässig placering av maxintensitet. Olsson (2019) utvärderade hyetografernas hydrologiska respons i relation till Chicago Design Storm (CDS), en vanlig designstorm som används i svensk hydrologisk modellering, och fann att CDS nästan alltid överskattar responsen i jämförelse med de empiriska hyetograferna.  Syftet med detta examensarbete är att analysera robustheten hos de empiriska hyetograferna genom att utföra en känslighetsanalys av en generisk stadsmodell med en variation av ingångsparametrar. En statistisk analys av data om hårdgjorda ytor, topografi och initial vattenmättnad genomfördes för att hitta median, 5:e respektive 95:e percentilen av värdena i svenska förhållanden. Dessa värden applicerades på en generisk stadsmodell och kördes tillsammans med de fem empiriska hyetograferna och CDS i MIKE 21. Resultaten indikerar att de empiriska hyetograferna inte är helt robusta vad gäller variationer i infiltrationskapacitet (initial vattenmättnad och andel hårdgjorda ytor), medan de är mer robusta vid variation av topografin. Variationen av topografi resulterade också i stora variationer i vattendjup, tid till maxflöde och utbredning av översvämmat område, medan variation av initial vattenmättnad och mängden hårdgjorda ytor hade mindre, men fortfarande tydliga, effekter. Vidare visar resultaten att hyetografer med sen topp är känsligare för variationer i initial vattenhalt och hårdgjorda ytor än hyetografer med tidig topp. I genomsnitt uppskattar CDS responsen från de empiriska hyetograferna korrekt, med en genomsnittlig överskattning på 0.5%, men fångar inte omfånget och komplexiteten hos de empiriska hyetograferna. Inräknat frekvensen av de olika hyetograferna gav CDS en överskattning på 5 % jämfört med de empiriska hyetograferna. Resultaten belyser osäkerheterna i att använda CDS regn som drivdata i skyfallsmodelleringar för stadsplanering, men begränsningar i form av markegenskaper i ett enskilt studieområde och vikten av modellens initialvärden begränsar dock generaliteten för studien.

The studies in this dissertation address the issue of variability in runoff generation and pollutant concentration in urban areas, and specifically in the catchments of stormwater control measures. There is an imperfect correlation between runoff volumes and the capture area and land uses of urban catchments. Variable capture areas and uncertainty in urban runoff sources complicate stormwater control measure design and urban stream assessment. Four stormwater control measures in upstream suburban Philadelphia, ranging in capture area from 0.11 ha to 32 ha, were monitored, sampled, and modeled. Sampling was conducted in the watersheds of Wissahickon Creek, Tookany Creek, and Pennypack Creek. The approaches discussed below have the goal of better understanding runoff and the movement of associated contaminants into stormwater retention basins and streams. Rather than viewing runoff generation and contaminant transport as a static process, this work proposes that the amount of runoff contributed from different areas of a catchment changes during and between storm events, and that the origin and concentration of contaminants change as a result. Linking source areas to runoff volumes through natural and modeled tracers could improve predictions of water quality and quantity in stormwater control measures in urban streams. Nitrate (NO3–) isotope ratios were used as tracer of flow from different urban land uses. Time series samples of stormwater runoff entering two stormwater control measures (a constructed wetland and a small bioretention basin) were collected and analyzed to distinguish sources of NO3– by samples’ δ15N and δ18O ratios. A Bayesian mixing model was used to determine that NO3– sources were a mix of soil nitrogen (N) and atmospheric deposition across six storm events. Furthermore, atmospheric versus soil N sources varied throughout the storms. The large catchment of the constructed wetland had more NO3– source variability between samples compared to the small catchment of the bioretention basin. Thus, the NO3– isotopes suggest more distinct flow paths in the large catchment and more mixing of flow across land uses in the small catchment. Quantifying flow path variability from storm to storm and between different catchments can improve design and placement of urban stormwater control measures. A distributed hydrologic model, GSSHA, was used to simulate overland runoff from impervious and semi-pervious land covers in the catchment of a stormwater control measure. The positions of low vegetation and impervious land uses over the catchment were rearranged to create hypothetical catchments during four storm events. Fluctuating source proportions over time suggested that grab samples might not be adequate for capturing average overland runoff chemistry. It was also found that the portion of total runoff volume from impervious areas varied from 50 to 75% while the relative proportion of impervious cover remained constant at 54%. Land use percentages averaged over capture areas are frequently used to estimate runoff amounts and pollutant concentrations, but this model disrupts the assumption that urban hydrologic responses can be predicted from imperviousness alone. Overland runoff was measured and modeled before and after the installation of two stormwater control measures, a berm and a bioswale. Discharge in the stream was modeled for 9 storms ranging in size from 14 to 54 mm. We found that during 4 of the modeled storms there was no decrease in stream discharge and decreases in discharge were generally only observed for low intensity storms. Furthermore, only 5% of the stream catchment was captured by SCMs. Modeled tracers, used to track runoff contributions from areas upslope of the SCMs found that the size of upslope contributing areas did not predict the proportion of runoff generated in each area. Field data to support the models included water level loggers and samples of overland runoff collected in subsurface stormwater casing. After the SCMs were installed, less water was captured in downslope sampling bottles, but new flow paths developed. Furthermore, significant variation was observed in upslope concentrations of dissolved nutrients and total suspended solids, casting doubt on whether point samples of urban overland runoff geochemistry can be representative given variable runoff generation and heterogeneous land uses. This study points out the challenges in evaluating stormwater control measures and reveals that source areas’ contribution to stream flow varies independently of their size. Therefore, modeling before stormwater control measure installation is recommended to determine the factors that influence a capture area’s contribution to urban streamflow. / Geoscience

Understanding the extent to which human activities impact surface water resources has become increasingly important as both human population growth and related landscape changes impact water quality and quantity across varying geographical scales. Skypark, Santa’s Village is a 233.76-acre tourism-based outdoor recreation area located in Skyforest, California residing within the San Bernardino National Forest. The park is situated at Hooks Creek, the headwaters of the Mojave River Watershed, and is characterized by a diverse landscape that includes forest cover and human development, including impervious surfaces, a restored meadow, and recreational trails. In 2016, Hencks Meadow was considered degraded by human activity and restored by the Natural Resources Conservation Services (NRCS) using best management practices (BMPs) to manage stormwater runoff and mitigate pollutants entering recreational downstream surface water. Three BMP detention basins were constructed to store and improve water quality from stormwater runoff. The purpose of this study is to observe the extent to which the engineered BMP detention basins design were effective in mitigating stormwater pollution from entering Hooks Creek. Over a six to eight month period (January to August), ponds were tested in situ bi-weekly for temperature (ºC), dissolved oxygen (mg/L), pH, turbidity (NTU), conductivity (µS/cm), nitrate (mg/L), and ammonium (mg/L), with additional laboratory tests for total suspended solids (mg/L), total dissolved solids (mg/L), chemical oxygen demand (mg/L), total coliform (MPN/100mL), Escherichia coli (MPN/100mL), and trace metals (µg/L). The results of this study support that the BMP design is improving surface stormwater runoff from impervious surfaces before it enters Hooks Creek. Findings could also promote the design and implementation of stormwater BMP detention basins at other site locations where water degradation is evident. Furthermore, this research can be used to promote the necessary improvement of water quality and quantity on a widespread geographical scale.

Since the early 1800's, many studies have recognized increased heat in urban areas, known as the urban heat island (UHI) effect, as one of the results of human modification to the natural landscape. UHI is related to differences in land surface temperature (LST) between rural areas and urban areas where factors of the built environment such as the thermodynamic capacities of materials, structural geometry, and heat generating activities cause increased storage and re-radiation of heat to the atmosphere. This thesis examines the correlation between factors of urbanization and differences in land surface temperature (LST) in the subtropical climate of Pinellas County, Florida using remote sensing techniques. It describes the spatial pattern of LST, analyzes its relationship to factors of urbanization relative to NDVI, percentage of impervious surface, and land use land cover in the study area. It also assesses the effectiveness of remote sensing as an efficient method of identifying LST patterns at the local and neighborhood level for mitigation strategies. Landsat TM thermal band imagery for three dates; April 1986, 2001 and 2009 was processed using Qin's mono-window algorithm (MWA) technique to derive LST levels. This data was compared to in-situ readings, then normalized and statistically analyzed for correlation with vegetation ratio (NDVI) and imperviousness percentages derived using linear spectral mixing/unmixing, and also with land use/land cover classification. The resulting LST spatial pattern is a gradient across the peninsular landscape, from cooler water and wetland areas to a generally warmer interior, interspersed with micro-urban heat islands (MUHIs), corresponding to urban structures and "cool-islands" of parkland and lakes. Correspondence between LST pattern and urban structures and land use demonstrates the suitability of medium resolution remote sensing data and techniques for identifying micro-urban heat islands (MUHIs) for possible mitigation. Mitigation could include relatively low-cost measures like replacement of inefficient asphalt roofs with more reflective and emissive "cool roofs," placement of "street trees" to enhance shade, and replacement of impervious pavements by permeable surfaces. The thesis concludes that Landsat TM imagery processed with the MWA provides an efficient, relatively low-cost method for locating MUHIs. Satellite remote sensing, combined with aerial photography can facilitate neighborhood level analysis for the implementation of low-cost mitigation techniques. Previous studies have demonstrated that these are successful ways to mitigate the UHI effect at the micro-scale level; lowering urban heat and saving energy, and also facilitating the reintegration of natural elements into the urban environment.

Skyfall har blivit mer och mer frekventa i Sverige till följd av den globala uppvärmningen och de klimatförändringar som sker, och kraftiga skyfall kan leda till omfattande skador som ibland är svåra att förutse. Studien undersöker potentiella samband mellan skador på byggnader som uppstod vid ett skyfallsevent i Jönköping 2013 och byggnadernas omgivande topografi, hydrologi samt hårdgjorda ytor. Tidigare studier inom området har undersökt samband mellan flera variabler och skadade områden, men genom att studera topografi/hydrologi samt hårdgjorda ytor tillsammans med försäkringsdata, undersöks dessa förutsättningars påverkan mer specifikt i detta arbete. Olika stora buffertzoner runt byggnader användes för att extrahera flödesackumulering baserade på höjddata för analys av hydrologin samt klassificerad data för analys av andel hårdgjord yta. Data från hela studieområdet samt ett mindre område koncentrerat till två villakvarter studerades och på extraherad data utfördes visualiseringar, statistiska beräkningar och T-tester för att undersöka potentiella samband. Resultaten visade att den större buffertzonen på 50 m med metoden identifierar ett större antal skadade byggnader än de mindre buffertzonerna. En klassificerad hårdgjord yta kan vara användbar för att identifiera mer sårbara områden i ett större perspektiv medan flödesackumulerad data på grund av sin stora variation verkar lämpa sig bättre för analyser av mindre områden. / Due to global warming and climate change, heavy rainfalls have become more frequent in Sweden and can lead to extensive damage that is sometimes difficult to predict. This study examines potential correlations between damage to buildings that occurred during a cloudburst event in Jönköping in 2013 and the surrounding hydrology, topography and degree of surface sealing. Previous studies in the field have examined correlations between several variables and damaged areas, but by studying hydrology and impervious surfaces individually and their correlation with insurance data, their impact is examined more specifically in this study. Different sized buffer zones around buildings were used to extract flow-accumulation data and percentage of impervious surfaces. Data from the entire study area and a smaller area concentrated on two residential blocks were studied and visualizations, statistical calculations and T-tests were performed on extracted data to investigate potential correlations. The results showed that the larger 50 m buffer zone with the method identifies a larger number of damaged buildings than the smaller buffer zones. Classified impervious surface can be useful for identifying more vulnerable areas in a larger perspective while flow-accumulation data due to its large variation seems better suited for analysis of smaller areas.