.|  Baltimore Ecosystem Study
The Baltimore School of Urban Ecology:
 
Space, Scale, and Time for the Study of Cities
 
J. Morgan Grove, Mary Cadenasso, Steward Pickett, William R. Burch, Jr., and Gary Machlis
 
Grove, J.M., Cadenasso, M., Pickett, S.T.A., Burch, W.R., Jr., Machlis, 2015. The Baltimore School of Urban Ecology: Space, Scale, and Time for the Study of Cities. Yale University Press. New Haven and London.
 

 

This book is about an interdisciplinary approach for meeting new needs for the study of cities: patch dynamics. In this book we advance patch dynamics as an approach to integrate science disciplines and practices in order to address the spatial, organizational, and temporal complexity of urban areas. We do not propose patch dynamics as a theory. Rather, we present patch dynamics as an approach towards building theories, developing methods, and advancing practice. While patch dynamics is most familiar from ecology, we believe this approach will resonate with many other disciplines and professions.
 
We view patch dynamics to have broad relevance. This book is intended for ecological and social scientists, for students who are interested in studying cities or working across disciplines and professions, and for practitioners — policy makers, planners, designers, managers, and community activists — who confront complex and interdependent urban issues.
 
This book is relevant to improving the connections between science and decision making. Increasingly, people recognize that complex urban issues require comprehensive approaches and understandings that involve multiple disciplines and account for space, scale, and time. Our goal is that readers of this book will gain new perspectives for how they can study, build, or manage cities.
 
We feel it is important to acknowledge what this book is not. This book is not a textbook about urban ecology. It does not try to account for and synthesize every idea, approach, or tradition. Rather, it is about an approach — patch dynamics — and its development that has served as the basis for the Baltimore Ecosystem Study since its beginning in 1997. Thus, this book reflects our experiences, growth, and history of developing an urban ecology research program and, frankly, our passion for all the nobility, quirks, venality, resilience, and beauty that is the history and future of the City of Baltimore, its region, and the Chesapeake Bay.
 
Visual representations are important for communicating concepts and data associated with patch dynamics. Below, we provide hi-resolution versions of the graphics from our book. In cases where we have created color versions, they are provided here. These graphics are made available for your use. In return, we ask that you include the following citation.
 
Grove, J.M., Cadenasso, M., Pickett, S.T.A., Burch, W.R., Jr., Machlis, 2015. The Baltimore School of Urban Ecology: Space, Scale, and Time for the Study of Cities. Yale University Press. New Haven and London.
 
These figures would not have been possible without the artistry and hard work of Amy Grove, Jarlath O'Neil-Dunne, Dexter Locke, Chris Boone, Michele Romolini, and Neil Bettez for all the work to produce these figures.
 
List of Figures:
 
Figure 1.1. The human ecosystem. The human ecosystem concept, bounded by the bold line, showing its expansion from the bioecological concept of the ecosystem as proposed originally by Tansley in the dashed line. The expansion incorporates a social complex and a built complex, which includes land modifications, buildings, infrastructure, and other artifacts. Both the biotic and the physical environmental complexes of urban systems are expected to differ from those in nonurban ecosystems. Download hi-resolution or low-resolution image.
 
TEST or low-resolution image.
 

Example 1
Download this image


 
Figure 1.2. Framework for complexity of social-ecological systems. The three dimensions of complexity are spatial heterogeneity, organizational connectivity, and temporal contingency. Components of the framework are arrayed along each axis, increasing in complexity from top to bottom. For example, a more complex understanding of spatial heterogeneity is achieved as quantification moves from patch richness, frequency, and configuration to patch change and the shift in the patch mosaic. Complexity in organizational connectivity increases from within-unit processes to the interaction of units and the regulation of that interaction to functional patch dynamics. Finally, historical contingency increases in complexity from contemporary direct effects through lags and legacies to slowly emerging indirect effects. While not shown in the figure, organizational connectivity can be assessed within and between levels of organization. Download hi-resolution or low-resolution image.
 

 
Figure 1.3. The human ecosystem framework: critical resources, social system, and flows. Download hi-resolution or low-resolution image.
 

 
Figure 1.4. Dynamic links between science and decision making: an abstracted cycle of interactions among scientists and decision makers. Download hi-resolution or low-resolution image.
 

 
Figure 1.5. Stokes’s Pasteur’s quadrant. In Pasteur’s quadrant, Stokes categorizes four different types of research. Most research associated with our work in Baltimore would be located in Pasteur’s quadrant: use-inspired basic research. Download hi-resolution or low-resolution image.
 

 
Figure 2.1. An illustrative diagram of the thematic history of biological ecology. The most recent development is a focus on social-ecological systems (SES). Download hi-resolution or low-resolution image.
 

 
Figure 2.2 E. W. Burgess’s zonal model: (left) the idealized pattern; (right) its application to Chicago (Credit, University of Chicago Press). Download hi-resolution or low-resolution image.
 

 
Figure 3.1. Examples of distinct HERCULES patches: (a) and (b) are differentiated by building density, although woody and herbaceous vegetation densities are the same; (c) and (d) are differentiated by density of woody and herbaceous vegetation, but building density is the same. Download hi-resolution or low-resolution image.
 

 
Figure 4.1. An interdisciplinary model of water quality and Bangkok’s klongs. Disciplines are shown in parenthesis. Several of the variables reflect patch dynamics characteristics and are identified with an asterisk. Download hi-resolution or low-resolution image.
 

 
Figure 4.2. The Human Ecosystem Framework: critical resources, social system, and flows. Download hi-resolution or low-resolution image.
 

 
Figure 5.1. Comparison of high- and midresolution land cover classifications derived from high-resolution National Agricultural Imagery Program (NAIP) imagery and midresolution Landsat TM imagery of the same area. Panel A shows high-resolution imagery for an area in New York City. Panel B classifies those portions of the area with tree canopy cover, using the NAIP imagery. Panel C classifies the same area using the Landsat TM data (National Landcover Database). Comparison of panels B and C illustrates that Landsat TM data are insufficient for detecting fine-scale vegetation in the more heterogeneous parts of the study area. Download hi-resolution or low-resolution image.
 

 
Figure 5.2. BES field-sampling types and locations in the Gwynns Falls watershed. BES plots for long-term vegetation sampling, stream-sampling sites, and meteorological stations (MET) are distributed along a gradient from urban to suburban to rural. Field-sampling sites are also located in Baisman’s Run, a native reference site northeast of the Gwynns Falls watershed. Download hi-resolution or low-resolution image.
 

 
Figure 5.3. Central Arizona-Phoenix (CAP) LTER field sampling plan. Field sampling plots are located based upon a grid matrix to distribute plots randomly and spatially-independently. Download hi-resolution or low-resolution image.
 

 
Figure 5.4. Distribution of Baltimore urban forest effects model (UFORE) vegetation field plots. UFORE plots are stratified and randomly distributed based upon twelve land use categories (n=400: 200 Baltimore City, 200 Baltimore County). Download hi-resolution or low-resolution image.
 

 
Figure 5.5. Distribution of BES household telephone surveys. The BES household telephone survey is stratified and randomly distributed based upon a sixty-six lifestyle-group classification (n=3,000). Download hi-resolution or low-resolution image.
 

 
Figure 5.6. Example of planned change: stormwater structures built by Baltimore County Department of Environmental Protection and Sustainability in the Gwynns Falls watershed. Download hi-resolution or low-resolution image.
 

 
Figure 5.7. Linking pixels, plots, and parcels. Pixels (land cover) and plots (telephone, ethnographic, and vegetation surveys, and soil plots) can be colocated at the parcel scale. Download hi-resolution or low-resolution image.
 

 
Figure 5.8. Changes in parcel boundaries over time in Baltimore: “The more things change, the more they stay the same.” The first column shows conditions in 2005, combining a remotely sensed image and parcel boundaries. The second column shows parcel and stream boundaries in 1915. The third column shows the overlay of the two time periods. Row 1 shows significant change during this time period, with the construction of a highway and displacement of a stream, while row 2 shows little change in a downtown neighborhood of Baltimore over this ninety-year period. Download hi-resolution or low-resolution image.
 

 
Figure 5.9. Examples of socioecological data types organized by scale and intensity of analysis. Data types in italics are data that are typically available from other sources. Data types in bold are typically collected by urban ecological research programs. Download hi-resolution or low-resolution image.
 

 
Figure 5.10. Example of Noncensus data organized by scale and time for the city of Baltimore from 1800 to 2000. Download hi-resolution or low-resolution image.
 

 
Figure 5.11. Correlation between race and distance to disamenities in the city of Baltimore from 1940 to 2000. Download hi-resolution or low-resolution image.
 

 
Figure 5.12. 1937 Home Owners’ Loan Corporation map for Baltimore city. HOLC description of neighborhood noted with a star. The inset description reads: “Security Grading (Declining). Location: The portion of the Ward lying south of Druid Hill Park bounded on the East by Mt. Royal Terrace; South, North Avenue; West, Reistertown Road. Description: An old residential section seriously threatened with negro encroachment. A small section along Reistertown Road consists of fairly modern two story brick rows. Mixed–some negroes, some owners of long standing still occupying old residences–converted apartments containing white collar class-skilled mechanics, etc., Population 1930 (whole ward) 38,596, 10.5% negro, 8.7% foreign born. Population increase since 1920 (whole ward) 14.7%. Favorable Features: Druid Hill Park and good transportation. Detrimental Features: Obsolescence and negro encroachment.” Download hi-resolution or low-resolution image.
 

 
Figure 5.13. Long-term social-ecological research (LTSER) platforms are similar to a table with four legs, each of which is essential to the integrity of the whole. The four legs are long-term monitoring, comparative analyses, experimentation, and modeling. Download hi-resolution or low-resolution image.
 

 
Figure 5.14. The LTSER data temple, with specific BES research themes included. Download hi-resolution or low-resolution image.
 

 
Figure 6.1. Snow’s map of the Golden Square neighborhood, the Broad Street pump and nearby pumps, and the location of victims of the epidemic. The height of the stacked lines indicates the number of persons who died at the address. Snow did not make this map, which was part of his report, until after the pump handle had been removed and the epidemic had subsided. Download hi-resolution or low-resolution image.
 

 
Figure 6.2. 40 Broad Street and the home of the Lewis family. A leak from the deteriorated cesspool in the front of the house connected to the Broad Street pump was the source of the cholera contamination. Download hi-resolution or low-resolution image.
 

 
Figure 6.3. Dynamic links between science and decision making: an abstracted cycle of interactions among scientists and decision makers. Download hi-resolution or low-resolution image.
 

 
Figure 6.4. Urban tree canopy cover and PRIZM lifestyle market categories. Neighborhood patches in Baltimore can be classified using PRIZM lifestyle market categories. The size of the circle indicates the amount of residential canopy cover in each neighborhood. Neighborhoods may have similar levels of population density, and households may have similar levels of income and education but are at different life stages. Differences in the amount of urban tree canopy cover per neighborhood may be significantly associated with household life stage. Download hi-resolution or low-resolution image.
 

 
Figure 6.5. Legacies of redlining: vacant lots and urban tree canopy cover. Comparison of panel A with panels B and C illustrates that neighborhoods classified as “hazardous” or “declining” in 1937 have the lowest rates of canopy cover in 2007 and the highest concentration of vacant lots and buildings in 2012. Download hi-resolution or low-resolution image.
 

 
Figure 6.6. Urban tree canopy (UTC) prioritization and similarity of organizations based upon prioritization goals.
 
The first figure is a summary map of UTC prioritization for Baltimore at the neighborhood scale based upon a stakeholder process (n=25 organizations). Download hi-resolution or low-resolution image.
 
The second is a tree diagram showing similarities among stakeholder groups in terms of what they considered to be their most important factors for prioritizing where to plant trees. Groups closer to each other are more similar. Download hi-resolution or low-resolution image.
 

 


 
Figure 6.7. The environmental stewardship network for the city of Baltimore and neighborhoods where organizations work.
 
(a) A network diagram of environmental stewardship organizations in Baltimore in 2011 (n=390 organizations). Larger circles mean that the organization has more connections with other organizations. Download hi-resolution or low-resolution image.
 
(b) Number of environmental stewardships groups per neighborhood. Groups that work at the citywide level are not included in the density mapping. Download hi-resolution or low-resolution image.
 

 

 
Figure 6.8. Mapping tree plantings by UTC priorities and the organizations that planted the trees.
 
(a) Tree plantings in the fall of 2013 by UTC prioritization category at the neighborhood level. Download hi-resolution or low-resolution image.
 
(b) Number of trees planted by organization for each UTC prioritization category. Download hi-resolution or low-resolution image.
 

 

 
Figure 6.9. Odds ratio for urban greening programs in Baltimore using ESRI’s (Environmental Systems Research Institute) tapestry market classification. Expected rate of adoption is equal to a value of 1, based upon household populations for each market type in the city. Two neighborhood types—upscale avenues and high society—have both the highest levels of existing canopy cover and the highest rates of adoption of “free giveaway” programs. Download hi-resolution or low-resolution image.
 

 
Figure 6.10. Extensive-intensive data framework for Baltimore’s UTC goal (cycles 1–7). Download hi-resolution or low-resolution image.
 

 

 

 
This research was supported by funding from the NSF Long-term Ecological Research (LTER) Program. This material is based upon work supported by the National Science Foundation under Grant No. DEB-1027188. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.