State-and-transition Models: Ecosystem Science & Management

Transcription

1 State-and-transition Models: Current Status and Future Direction David D. Briske Ecosystem Science & Management Texas A&M University

2 Professional Reinvention

3 State-and-transition Model Framework State 1 State 2 Threshold Community pathway Threshold Community phases within stable states Stringham et al State 3

4 Presentation Objectives Explore linkage between STMs and resilience Assess the current effectiveness of STMs Investigate role of empirical data in STMs Comment on the future direction of STMs

5 Thresholds vs Resilience Thresholds Resilience Resilience degree of modification that an ecosystem can absorb prior to transform to an alternative state. Threshold resilience limit of an ecosystem.

6 Resilience-based Management State I Ecological Resilience State II Ecological Resilience Trigger Threshold Community Phase At-risk Community Feedback switch Restoration pathway Community Phase Negative Feedbacks Negative Feedbacks Positive Feedbacks Positive Feedbacks Modified from Briske et al. 2008

7 Positive and Negative Feedbacks Grassland State Threshold Progression Woodland State ack Sw witch Feedb Positive Feedbacks woody plant cover coarse fuel loads propagule limitations Negative Feedbacks grassland productivity fine, continuous fuel loads propogule limitations

8 Mountain Clay Ecological Site, Oregon Idaho Fescue Bluebunch wheatgrass Reference State Indicators: High perennial grass cover, dispersed sagebrush cover, minimal juniper and bare soil. Idaho Fescue Bluebunch wheatgrass Feedbacks: Herbaceous cover retains water on site and provides fuel to support a fire return Idaho Fescue Mountain big sagebrush interval of less than 50 years. Bluebunch wheatgrass Juniper At-risk Community Phase: Herbaceous cover reduced, Mountain big sagebrush decadence, juniper visible and bare sagebrush soil patches increasing, potential fire frequency reduced. Juniper Sagebrush Idaho Fescue Bluebunch wheatgrass Juniper Idaho Fescue Sandberg bluegrass Trigger: Drought and intensive grazing promote juniper establishment through reduced fire frequency. Threshold: Juniper attains a height and density that reduces fine fuel load and fire-induced tree mortality. Large, inter-connected bare soil patches occur with Restoration Pathway: Bunchgrass redistribution of nutrients/soil beneath juniper canopies. (BG) density > 1 m 2 requires mechanical juniper removal only; BG density < 1 m 2 requires juniper removal and grass reseeding, if soil is intact. Alternative State Indicators: Mature juniper dominant, Idaho fescue only beneath juniper canopies, large interconnected bare soil patches, sagebrush decadence.. Feedbacks: Juniper dominates resource use, water and wind redistribute soil and nutrients beneath juniper, minimal grass and sagebrush establishment.

9 Effectiveness of STMs Say, aren t here supposed to be holes in this mask!

10 How Effective are STMs? Survey 47 rangeland professionals 26 Agency Managers 21 Research Scientists Purposes of STMs Model Strengths th Model Weaknesses Construction and Review

11 STM Purposes Guide management (87%) Managers 92%; Researchers 81% Describe ecological l dynamics (70%) Managers 65%; Researchers 76% Identify testable hypotheses h (40%) Managers 12%; Researcher 76% Communications tool (38%) Managers 35%; Researchers 43%

12 STM Strengths Improve decision making (87%) Managers 92%; Researchers 81% Describe system dynamics (70%) Managers 65%; Researchers 76% Improve communication i (38%) Managers 35%; Researchers 43% Identify relevant questions (34%) Managers 19%; Researchers 52%

13 STM Weaknesses Insufficient information (43%) Managers 30%; Researchers 57% Models overly complex (26%) Managers 38%; Researchers 10% Lack of time and resources (21%) Managers 27%; Researchers 14% Potential misrepresentation (17%) Managers 8%; Researchers 29%

14 Construction & Review Expert knowledge critical (43%) Managers 47%; Researchers 37% Minimal i empirical i knowledge (43%) Managers 34%; Researchers 61% Model inconsistency i (26%) Managers 34%; Researchers 13% Mechanisms for validation (87%) Managers 87%; Researchers 88%

15 Areas of STM Refinement Management vs ecological drivers Role of expert vs empirical knowledge Criteria to define thresholds Appropriate model complexity Model review and revision

16 Value of Empirical Data Then a miracle occurs!!

17 Attributes Idaho Data Sets Idaho National Lab Plant density 34 m 2 plots Sampled 10 times N = 340 samples Species = 55 MAP = 220 mm Idaho Falls ID US Sheep Station Plant density m 2 plots Sampled 23 times N = 545 samples Species = 54 MAP = 300 mm Dubois ID

18 Data Analysis Dr Sumanta Bagchi Identify communities with cluster analysis Verify community membership against species dissimilarity BIC-parsimony, ANOSIM, SIMPER Record community transitions in time Categorize transition frequency and attributes

19 Species Composition b) INEEL others Sa.ka Pa.sm Al.de Me.al Br.te La.oc Op.po Average abundance (%) He.co El.el El.la Ch.vi Ar.trd a) USSES others El.al Ko.ma Po.se Br.te Ps.sp Te.ca Ph.lo Average abundance (%) Ph.ho ho Cr.ac Ch.vi Ba.sa Ar.trp

20 Transition Dissimilarity US Sheep Station Idaho National Lab 0 0 Freq quency Freque ency Species dissimilarityi il it Species dissimilarity

21 Empirical STMs a) USSES D 5 A B b) INEEL D C A C E 4 F E 3 5 B

22 Community Transitions Precipitation (cm m) a) USSES Precipitation (cm) Cheatgrass b) INEEL Precipitation (cm) Cheatgrass Precip pitation (cm) ive grass relative ndance (%) Cheatg abun ) Ch heatgrass relat abundance (%

23 Temporal Dynamics 1.0 a) Abrupt change 1.0 c) Gradual change DCA axis b) Frequent reversal d) Relative stability DCA ax is DCA axis 1 DCA axis 1

24 Summary Idaho Data Sets Transitions occurred in a 10 yr window Associated with increasing cheatgrass density Transitions decreased at maximum density Alternative stable state formed Cheatgrass is a biotic trigger Interaction with precipitation patterns Feedbacks rapid and unrelated to fire Likely induced by plant-soil processes Similar patterns occurred at both sites

25 Value of Empirical Data? Empirical data can support STMs: Describe community transitions Identify temporal scales Assess feedback mechanisms Refine resilience hypotheses Vegetation records insufficient: Adaptive management best approach Monitor management outcomes Consider autogenic & climatic processes

26 Future of STMs Strong, consistent support among stakeholders Continue resilience-based foundation Adaptive management supported w/ monitoring Science-management partnerships