Society for Integrative and Comparative Biology, 2010 Annual Meeting. Meeting Abstract 67.11 Wednesday, Jan. 6 Resilience in urban socioecological systems: residential water management as a driver of biodiversity KATTI, M*; SCHLEDER, B; California State Univ, Fresno; California State Univ, Fresno mkatti@csufresno.edu Cities are unique ecosystems where human social-economic-cultural activities prominently shape the landscape, influencing the distribution and abundance of other species, and consequent patterns of biodiversity. The long-term sustainability of cities is of increasing concern as they continue to grow, straining infrastructure and pushing against natural resources constraints. A key resource is water, esp. in the more rapidly urbanizing arid regions. Understanding water management is thus critical for a deeper theoretical understanding of urban ecosystems and for effective urban policy. Landscaping and irrigation at any urban residence is a product of local geophysical/ecological conditions, homeowners’ cultural preferences, socioeconomic status, neighborhood dynamics, zoning laws, and city/state/federal regulations. Since landscape structure and water availability are key determinants of habitat for other species, urban biodiversity is strongly driven by the outcome of interactions between these variables. Yet the relative importance of ecological variables vs human socioeconomic variables in driving urban biodiversity remains poorly understood. Here we analyze data from the Fresno Bird Count, a citizen science project in California’s Central Valley, to show that spatial variation in bird diversity is best explained by a multivariate model including significant negative correlations with % building and grass cover, and positive correlations with interactions between irrigation intensity, median family income, and grass height. We discuss implications of our findings for urban water management policies in general, and for Fresno’s planned switch to metering water use in 2013. Ecological theory, conservation, and urban policy all benefit if we recognize cities as coupled socioecological systems.

1. Resilience in urban socioecological
systems: residential water
management as a driver of
biodiversity
Madhusudan Katti
Bradley Schleder
Department of Biology
California State University, Fresno
http://www.fresnobirds.org/

2. Natural
Environment

3. Natural
Environment
Individual responses
Behavior, Physiology, Life History

4. Regional/Global
Ecosystem
Natural Community
Environment
Population
Individual responses
Behavior, Physiology, Life History

5. Regional/Global
Ecosystem
Natural Community Human
Environment Impact
Population
Individual responses
Behavior, Physiology, Life History

6. A coupled socioecological system (SES)
Actions
interventions
Human System Ecosystem
Ecosystem
services

7. Unpacking the black box of
“human impact”

8. Unpacking the black box of
“human impact”
• Human actions alter the ecosystem (directly and
indirectly), e.g. by altering land-use/land-cover
(LULC) with diverse effects on other species

9. Unpacking the black box of
“human impact”
• Human actions alter the ecosystem (directly and
indirectly), e.g. by altering land-use/land-cover
(LULC) with diverse effects on other species
• Human actions are complex products of
interactions among many variables:

10. Unpacking the black box of
“human impact”
• Human actions alter the ecosystem (directly and
indirectly), e.g. by altering land-use/land-cover
(LULC) with diverse effects on other species
• Human actions are complex products of
interactions among many variables:
• economic characteristics

11. Unpacking the black box of
“human impact”
• Human actions alter the ecosystem (directly and
indirectly), e.g. by altering land-use/land-cover
(LULC) with diverse effects on other species
• Human actions are complex products of
interactions among many variables:
• economic characteristics
• sociological arrangements

12. Unpacking the black box of
“human impact”
• Human actions alter the ecosystem (directly and
indirectly), e.g. by altering land-use/land-cover
(LULC) with diverse effects on other species
• Human actions are complex products of
interactions among many variables:
• economic characteristics
• sociological arrangements
• cultural characteristics and background

13. Unpacking the black box of
“human impact”
• Human actions alter the ecosystem (directly and
indirectly), e.g. by altering land-use/land-cover
(LULC) with diverse effects on other species
• Human actions are complex products of
interactions among many variables:
• economic characteristics
• sociological arrangements
• cultural characteristics and background
• institutional arrangements

14. Unpacking the black box of
“human impact”
• Human actions alter the ecosystem (directly and
indirectly), e.g. by altering land-use/land-cover
(LULC) with diverse effects on other species
• Human actions are complex products of
interactions among many variables:
• economic characteristics
• sociological arrangements
• cultural characteristics and background
• institutional arrangements
• politics

15. Residential water use in Fresno

16. Residential water use in Fresno
• Currently, 51% of water is used residentially

17. Residential water use in Fresno
• Currently, 51% of water is used residentially
• 70% of residential water used for irrigation

18. Residential water use in Fresno
• Currently, 51% of water is used residentially
• 70% of residential water used for irrigation
• Starting in 2013, flat rate water will be
replaced by metered water

19. Residential water use in Fresno
• Currently, 51% of water is used residentially
• 70% of residential water used for irrigation
• Starting in 2013, flat rate water will be
replaced by metered water
• As water becomes more expensive,
residents are expected to reduce
consumption

20. Residential water use in Fresno
• Currently, 51% of water is used residentially
• 70% of residential water used for irrigation
• Starting in 2013, flat rate water will be
replaced by metered water
• As water becomes more expensive,
residents are expected to reduce
consumption
• Residential irrigation is therefore predicted
to decrease

21. Residential water use in Fresno
• Currently, 51% of water is used residentially
• 70% of residential water used for irrigation
• Starting in 2013, flat rate water will be
replaced by metered water
• As water becomes more expensive,
residents are expected to reduce
consumption
• Residential irrigation is therefore predicted
to decrease
• Decreased irrigation is in turn predicted to
alter bird diversity

22. Residential water use in Fresno
• Currently, 51% of water is used residentially
• 70% of residential water used for irrigation
• Starting in 2013, flat rate water will be
replaced by metered water
• As water becomes more expensive,
residents are expected to reduce
consumption
• Residential irrigation is therefore predicted
to decrease
• Decreased irrigation is in turn predicted to
alter bird diversity
• The effects may be more extreme in
impoverished urban areas

23. Challenges and Data Needs for Urban Ecology

24. Challenges and Data Needs for Urban Ecology
• Urban habitats are highly variable
• many different land uses
• rapid changes over short distances
• need many sites for data precision

25. Challenges and Data Needs for Urban Ecology
• Urban habitats are highly variable
• many different land uses
• rapid changes over short distances
• need many sites for data precision
• Many species of birds
• Measurements must be repeatable at
same sites over years

26. Fresno-Clovis Metro Area (FCMA)
Fresno Bird Count Study
Madera
Californi Clovi
Fresno
Fresno
Fresno
FBC Censused in Habitat surveyed
site 2008 (N=200) (N=38)

27. Fresno Bird Count - April 15-May 15:
Survey Design & Census Methodology

28. Fresno Bird Count - April 15-May 15:
Survey Design & Census Methodology
• 460 sites on a 1Km X 1Km grid
• each site located randomly within a single grid cell

29. Fresno Bird Count - April 15-May 15:
Survey Design & Census Methodology
• 460 sites on a 1Km X 1Km grid
• each site located randomly within a single grid cell
• Grouped into 58 clusters (routes) of 7-8 sites each
• possible to cover all sites in a single morning

30. Fresno Bird Count - April 15-May 15:
Survey Design & Census Methodology
• 460 sites on a 1Km X 1Km grid
• each site located randomly within a single grid cell
• Grouped into 58 clusters (routes) of 7-8 sites each
• possible to cover all sites in a single morning
• Counts conducted within 4 hours after sunrise

31. Fresno Bird Count - April 15-May 15:
Survey Design & Census Methodology
• 460 sites on a 1Km X 1Km grid
• each site located randomly within a single grid cell
• Grouped into 58 clusters (routes) of 7-8 sites each
• possible to cover all sites in a single morning
• Counts conducted within 4 hours after sunrise
• Point Count
• 5-minute duration
• count all birds seen & heard
• within a 40m-radius circle

32. Fresno Bird Count - April 15-May 15:
Survey Design & Census Methodology
• 460 sites on a 1Km X 1Km grid
• each site located randomly within a single grid cell
• Grouped into 58 clusters (routes) of 7-8 sites each
• possible to cover all sites in a single morning
• Counts conducted within 4 hours after sunrise
• Point Count
• 5-minute duration
• count all birds seen & heard
• within a 40m-radius circle
• ~200 points surveyed during 2008 and 2009 each

33. Fresno Bird Count - April 15-May 15:
Habitat and Socioeconomic Assessment

34. Fresno Bird Count - April 15-May 15:
Habitat and Socioeconomic Assessment
• A subset of 38 points, located in residential areas only

35. Fresno Bird Count - April 15-May 15:
Habitat and Socioeconomic Assessment
• A subset of 38 points, located in residential areas only
• Rapid survey of habitat variables at 20m-radius plots
• % Canopy Cover
• Ground cover: % grass, impervious, dirt/mulch, buildings, gravel
• Number of Trees
• Average Tree Height
• Number of Shrubs
• Average Shrub Height
• Grass Height
• Irrigation intensity score (4 point scale: 0-3) for each residence
overlapping plot; Mode of score is used in analyses presented here
• % population in surrounding census block group living below
poverty level (<$18,310/yr for a family of three)*
*US Census, Department of Health and Human Services 2008 - 2009

36. Examples of Irrigation Regimes
0
2
3

37. Distribution of Bird Species Richness in the FCMA
In 2008
• 186 points surveyed
by 35 volunteers
• 86 bird species
recorded
• 3263 total birds
seen
• Average species
richness per site:
5.13 ± 0.16 SE
1-3 spp 3-6 spp 6-9 spp 9-12 spp

38. Irrigation is correlated with...
Poverty Bird Species Richness
1.00
10
9
0.75 8
3
Mode Irrigation
# Bird Species
7
0.50 6
5
4
0.25 2
3
2
0.00 1
0 10 20 30 40 50 60 0 2 3
% Poverty Mode Irrigation
R2(U) =0.10, Χ2 p =0.01, F2,35=3.18, p =0.05, R2=0.15
N=38

39. Bird Species Richness is also
correlated with...
% Building Open Canopy
10 10
9 9
8 8
# Bird Species
# Bird Species
7 7
6 6
5 5
4 4
3 3
2 2
1 1
-5 0 5 10 15 20 25 30 35 30 40 50 60 70 80 90 100
% Building % Open Canopy
R2 =0.12, p =0.04, N=38 R2 =0.11, p =0.04, N=38

40. Multivariate Drivers of Bird Diversity in the FCMA
• Stepwise regression: 9 variables, 8 interaction terms, mixed procedure
• Best fit model as shown in table below
• Whole model R2=0.64 (adj. R2=0.53), F(8,24)=5.43, P=0.0006
Source +/- F-ratio Prob >F
Mean Grass Height * Mode Irrigation Score + 10.71 0.003*
% Building - 7.93 0.0096*
Mean Grass Height (cm) - 4.81 0.038*
% Poverty * Mode Irrigation Score + 4.33 0.048*
% Poverty * % Building + 2.55 0.12
Mean Shrub Height (m) - 2.16 0.16
Mode Irrigation Score + 0.86 0.36
% Poverty - 0.048 0.83
• Model comparison based inference supported the above results; the lowest AICc =
119.665 for comparison of 56 8-paramater models

41. Conclusions from the FCMA

42. Conclusions from the FCMA
• Urban bird diversity increased significantly with
residential irrigation intensity.

43. Conclusions from the FCMA
• Urban bird diversity increased significantly with
residential irrigation intensity.
• Residential irrigation intensity decreased
significantly with increasing poverty levels.

44. Conclusions from the FCMA
• Urban bird diversity increased significantly with
residential irrigation intensity.
• Residential irrigation intensity decreased
significantly with increasing poverty levels.
• Neighborhood poverty level (i.e., economic status)
has strong effects on irrigation intensity even
without water metering.

45. Conclusions from the FCMA
• Urban bird diversity increased significantly with
residential irrigation intensity.
• Residential irrigation intensity decreased
significantly with increasing poverty levels.
• Neighborhood poverty level (i.e., economic status)
has strong effects on irrigation intensity even
without water metering.
• Poverty also has strong indirect effects on bird
diversity through intermediate variables including
irrigation intensity, grass height, % building cover,
and shrub height.

46. Conclusions from the FCMA

47. Conclusions from the FCMA
• Socioeconomic status and related irrigation /
landscape management practices by residents
appear to be strong drivers of habitat structure
for birds, and overall bird diversity in the FCMA.

48. Conclusions from the FCMA
• Socioeconomic status and related irrigation /
landscape management practices by residents
appear to be strong drivers of habitat structure
for birds, and overall bird diversity in the FCMA.
• In an arid region like the San Joaquin Valley
irrigation dramatically alters the landscape and
may provide resources (food and cover)
otherwise unavailable to bird species.

49. Differences in landscape structure
Phoenix:
Stream channels larger, more perennial
More irrigation, pseudo-riparian
vegetation, golf courses
mesic yards
Tucson:
Mostly seasonal washes
xeric yards
Indian Bend Wash, Scottsdale Rillito Wash, Tucson

50. Abert’s Towhee
Endemic to lower Sonoran desert
Prefers riparian habitat, especially
Pipilo aberti for breeding

51. Abert’s Towhee in two cities
Phoenix Tucson
2.0 2.0
Ln (Towhee Abundance)
SPRING
Ln (Towhee Abundance)
SPRING
Spearman’s ρ = -0.336 Spearman’s ρ = -0.183
P = 0.016 P < 0.0001
1.5 Autumn 1.5
Winter
Spring
1.0 1.0
0.5 0.5
0.0 0.0
0 3000 6000 9000 12000 15000
0 3000 6000 9000 12000 15000
Distance to Nearest River (m) Distance to Nearest Major Wash (m)
M. Katti, W. Turner, E. Shochat (2001)

52. Ongoing and proposed research
in the FCMA

53. Ongoing and proposed research
in the FCMA
• Results provide baseline (Before) data as part of a
Before-After-Control-Impact experiment (a
“found” experiment)

54. Ongoing and proposed research
in the FCMA
• Results provide baseline (Before) data as part of a
Before-After-Control-Impact experiment (a
“found” experiment)
• Mapping of tree cover and species diversity

55. Ongoing and proposed research
in the FCMA
• Results provide baseline (Before) data as part of a
Before-After-Control-Impact experiment (a
“found” experiment)
• Mapping of tree cover and species diversity
• Detailed socioeconomic and cultural surveys of
individual homeowners (to begin in 2010)

56. Ongoing and proposed research
in the FCMA
• Results provide baseline (Before) data as part of a
Before-After-Control-Impact experiment (a
“found” experiment)
• Mapping of tree cover and species diversity
• Detailed socioeconomic and cultural surveys of
individual homeowners (to begin in 2010)
• Survey of institutional agents

57. Ongoing and proposed research
in the FCMA
• Results provide baseline (Before) data as part of a
Before-After-Control-Impact experiment (a
“found” experiment)
• Mapping of tree cover and species diversity
• Detailed socioeconomic and cultural surveys of
individual homeowners (to begin in 2010)
• Survey of institutional agents
• Analysis of water policies

58. Ongoing and proposed research
in the FCMA
• Results provide baseline (Before) data as part of a
Before-After-Control-Impact experiment (a
“found” experiment)
• Mapping of tree cover and species diversity
• Detailed socioeconomic and cultural surveys of
individual homeowners (to begin in 2010)
• Survey of institutional agents
• Analysis of water policies
• Detailed Land-Use/Land-Cover (LULC) modeling

59. Ongoing and proposed research
in the FCMA
• Results provide baseline (Before) data as part of a
Before-After-Control-Impact experiment (a
“found” experiment)
• Mapping of tree cover and species diversity
• Detailed socioeconomic and cultural surveys of
individual homeowners (to begin in 2010)
• Survey of institutional agents
• Analysis of water policies
• Detailed Land-Use/Land-Cover (LULC) modeling
• Develop and test a more comprehensive model of
the urban socioecological system (SES)

60. A conceptual model of an
urban SES focusing on water

61. Thanks are due to:
• Inspiration:
• Will Turner - Tucson Bird Count;
CAP-LTER, BES
• Mary Cadenasso, Derya Ozgoc-
Caglar, Andrew Jones, Lara Triona,
Henry Delcore, John Bushoven,
Sarah Wallace
• Perspiration:
• All the volunteers of the FBC!
• Web Database (work in progress):
• NiJeL.org (Lela Prashad, Nancy
Jones, JD Godchaux)
• Funding:
• Provost, CSU-Fresno
• College of Science & Mathematics,
CSU-Fresno