Sacramento Growth 1992-2001

Title Measuring the sprawl of the Sacramento Metropolitan Area, 1992-2001
Natasha J. Vidic American River College, Geography 350: Data Acquisition in GIS; Spring 2009
Abstract The Sacramento Metropolitan Area is a sprawling urban area with total population close to 2,000,000 people. The objective of this paper was to quantify the growth of the contiguous Sacramento urban area between 1992 and 2001. To fulfill this objective, I acquired USGS National Land Cover Database layers for 1992 and 2001, and the USGS National Land Cover Database 1992/2001 Change Product. Because the land use categories changed between 1992 and 2001 layers, the individual categories of urban use (e.g., low intensity residential, commercial/industrial, etc.) were not directly comparable. I summed all urban categories to calculate the total urban area for each layer. The total urban area in 1992 encompassed 626 km2 compared to 899 km2 in 2001, an increase of 144%, which can justifiably be called urban sprawl. However, the analysis of the 1992/2001 Change Product resulted in the change of only 83 km2. The latter value implied that the 1992-2001 rate of sprawl decreased significantly compared to 1970-1990 when the Sacramento urban area increased by 233 km2. Therefore, I obtained another estimate using the population change divided by average 2001 population density and arrived to an estimate of area increase of 160 km2. This value implies an increase in the rate of sprawl of about 33% and is likely closer to the real increase than the value derived from the 1992/2001 Change Product. Despite the differences in land use categories between the 1992 and 2001 Land Cover layers, they may be more suitable for change analysis than the 1992/2001 Change Product which seems to underestimate the degree of land use change.
Introduction Urban sprawl is a term referring to the spreading of urban area over rural land at the fringe of an urban area (Sprawl City, 2000). Sprawling neighborhoods of low population density single-family homes force their residents to commute by automobile to work (Squires, 2002). Urban planners emphasize the qualitative aspects of sprawl such as the lack of transportation options and pedestrian friendly neighborhoods. Walkability and bikeability of such neighborhoods is poor; it has been linked with higher rates of obesity (Plantinga and Bernell, 2007). Prime agricultural land is often consumed by the sprawl and causes a decline in land and water quantity and quality. Soil sealing (paved surfaces) increase the flood risk of the area (Haase, 2009). In California, population growth has been shown to be the major driving force of urban sprawl at the end of 20th century (Kolankiewicz and Beck, 2000). The population of the four counties which converge in the Sacramento Metropolitan Area (Sacramento, Placer, ElDorado and Yolo) has increased from about 1,520,000 in 1990 to about 2,000,000 at the present (U.S. Census facts). Several neighboring towns have merged into a single contiguous metropolitan area (Figure 1), which has been ranked to be 57 of 100 largest U.S. urban areas based on the amount of land it consumed between 1970 and 1990 (ca. 233 km2 or 90 mi2; Sprawl City, 2000). I set out to measure the growth of the contiguous Sacramento urban area shown in Figure 1. I acquired the National Land Cover layers for 1992 and 2001 for the Sacramento region. I used raster algebra to calculate the differences in the extent of urban areas in 1992 and 2001 and found a large increase in the extent of urban areas. I compared the results to the significantly more conservative estimates derived from National Land Cover Database 1992/2001 Change Product and discuss their viability. I use the population increase and 2001 population density to estimate the increase of the urban area. I use the estimates to calculate rates of sprawl and compare them to the 1970-1990 estimates.
Methods The National Land Cover Database layers for 1992 and 2001 were dowloaded from the USGS National Map Seamless Server. National Land Cover Database 1992/2001 Change Product was dowloaded from the USGS Multi-resolution Land Use Consortium, National Land Cover Data Access Portal. The data in all three products were available in raster form (30x30 m) and dowloaded for the area of interest. The downloaded 1991 and 1992 layers were rectangular in shape, but were extracted using a polygon shapefile to trim the surrounding urban areas from the data range (Figs. 2,3). Because of the differening land use categories for 1992 and 2001 land cover layers, I merged all urban categories to one. Using new data, I used raster algebra to identify the urban areas on the 2001 land cover that did not exist in 1992. I analyzed the data and compared them to data derived by attribute selection from the National Land Cover Database 1992/2001 Change Product (referred to as 1992/2001 Change Product from now on).
Results Urban areas in the National Landcover Database layers from 1992 and 2001 A visual inspection of land cover maps for the Sacramento area suggests a significant increase in area and intensity of urban/developed areas from 1992 to 2001 (Figures 2, 3). However, because land use categories have changed from 1992 to 2001 layers, a quantitative analysis of individual urban categories was not feasible. I summed the areas of individual urban categories to obtain the total urban areas for each map. They summed up to 626 and 899 km2 for 1992 and 2001, respectively, an increase of 144% relative to the total 1992 area. I used ArcMap (ESRI) to merge the residential/commercial/open space categories into a single urban category. The total 1992 urban area added up to 633 km2 (244 mi2; Fig. 4), a bit more than obtained by summing the areas of individual urban use categories. I obtained a similar result for the 2001 layer: the total urban area of 941 km2 (363 mi2), and a 150% increase relative to the 1992. I employed map algebra to subtract 1992 urban areas from the ones in the 2001 layer (Fig. 6). The new urban areas shown in Fig. 6 sum to 300 km2 of (116 mi2) increase of urban use, a bit more than the 273 km2 difference calculated from the layers. Urban areas in the National Land Cover Database 1992/2001 Change Product The total urban area extent derived from the 1992/2001 Change Product (Fig. 7) is about 895 km2, similar to the total urban area derived from the 2001 Land Cover layer. However, the degree of change derived from the 1992/2001 Change Product (83 km2, 32 mi2) is significantly lower than those calculated or derived from the individual land cover layers (273-300 km2 or 105-116 mi2). These calculations would bring the extent of 1992 urban areas to 812 km2 (313 mi2), about 130% of the 1992 Land Cover layer estimate and reduce the increase in urban areas to about 110% relative to 1992. Figure 1: ESRI World Imagery 1:250,000 (published July 2008) showing the sprawling Sacramento urban area and the names of several towns that have been agglomerated into a single contiguous urban area. Figure 2: The downloaded and extracted 1992 National Landcover Data Layer for the Sacramento Metropolitan area overlying the ESRI World Imagery basemap. Scale 1:250,000. Urban areas in hues of pink and red. The total urban area encompasses 626 km2 (242 mi2). Figure 3: The downloaded and cropped 1992 National Landcover Data Layer for the Sacramento Metropolitan area overlying the ESRI World Imagery base map. Scale 1:250,000. Urban areas in hues of pink and red. The total urban area encompasses 899 km2 (347 mi2). Figure 4: The total extent of urban areas (pink) derived from the 1992 Land Cover Database Layer is 633 km2 (244 mi2). ESRI World Imagery base map underneath. Scale 1:250,000. Figure 5: The total extent of urban area of 941 km2 (363 mi2) derived from the 1992 (pink) and 2001 (cyan) Land Cover Database Layer overlying the ESRI World Imagery base map. Scale 1:250,000. Figure 6: The result of raster algebra subtracting 1992 urban areas from 2001 (cyan) resulting in about 300 km2 (115 mi2) of new urban use areas. Scale 1:250,000. Figure 7: The Sacramento Metropolitan urban areas from the The 1992/2001 Change Product. Red – urban (812 km2), cyan – change to urban (83 km2; 32 mi2)). Scale 1:250,000.
Discussion The results of my research suggest a significant growth of the extent or the Sacramento Metropolitan Area between 1992 and 2001. The analyses of the 1992 and 2001 National Land Cover layers show an increase of urban areas from 626 to 899 km2 (from 242 to 347 mi2), or about 144% of 1992 area. The analysis of the 1992/2001 Change Product resulted in an increase of urban areas of only about 10%. I have not attempted to analyze the sophisticated methodology of the 1992/2001 Change Product, however, I think it underestimates the degree of change, especially for the urban areas. To test this hypothesis, I made several assumptions to arrive to another estimate of the extent of newly urbanized land. I assumed that the growth was driven by population change. I used the average population density calculated for 2001 (ca. 2000 people/ km2) and the 1990-2000 increase in population (315,755; U.S. Census Data). The result was about 160 km2 or 60 mi2, twice the 1992/2001 Change Product estimate, and about 60% of the areas derived from the two land cover layers. I used the sprawl data reported by Sprawl City, 2000, to compare my sprawl estimates for 1992-2001. I calculated the sprawl rate for the 1970-1900 interval to be 11.7 km2/yr (4.5 mi2/yr). The 1992/2001 Change Product area estimate yields a rate of 8.3 km2/yr (3.2 mi2/yr), suggestint a decrease in sprawl rate, which is not very likely. The area estimates derived from the Land Cover layers yield high rates of sprawl, 27.3 km2/yr (10.5 mi2/yr). The estimates derived from the population growth and density data show an increase in sprawl rate of about 33%, which is probably the closest to the reality.
Future Research I would like to obtain other data, such as county parcel data with the year of construction, to try to provide a better model of the growth of the Sacramento Metropolitan Area over a longer time period. I am especially interested in real estate boom period between 2002-2007.As a soil scientist, I am also interested in asessing the quality of agricultural soils lost to urbanization. I have acquired the data for the counties which converge in the Sacramento Metropolitan Area.
Conclusions The Sacramento Metropolitan Area has undergone significant growth between 1992 and 2001. According to the analyses of Land Cover Layers, the urban area has increased from 626 to 899 km2, about 144%. However, although the 1992/2001 Change Product yielded similar estimate for the total urban area for 2001, the estimate of change for considerably lower (83 km2). My third approach of dividing the population increase with 2001 average population density resulted in an increase of urban area of 160 km2. The higher values are more likely to be close to the real increase than that derived from the 1992/2001 Change Product. The comparisons of sprawl rates supports my hypothesis as it is not likely that the rate of sprawl decreased relative to the 1970-1990 interval. Despite the differences in land use categories between the 1992 and 2001 Land Cover layers, they may be more suitable for change analysis than the 1992/2001 Change Product which seems to underestimate the degree of change.
References Haase, D., Effects of urbanisation on the water balance - A long-term trajectory, Environmental Impact Assessment Review, In Press, Corrected Proof, Available online 15 February 2009, ISSN 0195-9255, DOI: 10.1016/j.eiar.2009.01.002. Huang, S.L., Szu-Hua Wang, William W. Budd, Sprawl in Taipei's peri-urban zone: Responses to spatial planning and implications for adapting global environmental change, Landscape and Urban Planning, Volume 90, Issues 1-2, 15 March 2009, Pages 20-32, ISSN 0169-2046, DOI: 10.1016/j.landurbplan.2008.10.010. Kolankiewicz L. and Beck R. , Sprawl in California, 32 p. CAPS 2000 Conference at the University of Southern California on August 13, 2000 http://sprawlcity.org/studyCA/CAsprawl.pdf Multi-resolution Land Use Consortium, National Land Cover Data Access Portal: http://gisdata.usgs.gov/website/mrlc/ Plantinga, A., Bernell, S. (2007). The Association Between Urban Sprawl and Obesity: Is it a Two Way Street, Journal of Regional Science, Volume 47, Number 5: 857-879. Squires G.D., 2002, Editor, Urban sprawl: causes, consequences and policy responses, The Urban Institute Press, Washington D.C. Wassmer, Robert W.,Urban Sprawl in a U.S. Metropolitan Area: Ways to Measure and a Comparison of the Sacramento Area to Similar Metropolitan Areas in California and the U.S.(September 8, 2000). CSUS Public Policy and Administration Working Paper No. 2000-03. Available at SSRN: http://ssrn.com/abstract=241975 or DOI: 10.2139/ssrn.241975 The National Map Seamless Server: http://seamless.usgs.gov/index.php U. S. Census Data: http://www.census.gov/
Acknowledgements I would like to thank Paul Veisze for providing the incentive for this research, and Hugh Howard for invaluable assistance in raster analyses.