rapid ecoregional assessment n.
Skip this Video
Loading SlideShow in 5 Seconds..
Rapid Ecoregional Assessment PowerPoint Presentation
Download Presentation
Rapid Ecoregional Assessment

Rapid Ecoregional Assessment

292 Vues Download Presentation
Télécharger la présentation

Rapid Ecoregional Assessment

- - - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript

  1. Climate and related factors: preliminary results Rapid Ecoregional Assessment Yukon Lowlands-Kuskokwim Mountains-Lime Hills Rapid Ecoregional Assessment Project, Alaska

  2. Schematic of MQs pertaining to climate trends • What are the projected monthly, seasonal, and annual temperature, precipitation, and length of warm and cold seasons for the REA, and how do these projections vary across time, across the region, and across varying global greenhouse gas emissions scenarios? • Where will climate change impact CEs, including subsistence species?

  3. The Scenarios Network for Alaska and Arctic Planning is a collaborative network of the University of Alaska, state, federal, and local agencies, NGOs, and industry partners. Its mission is to provide timely access to scenarios of future conditions in Alaska and the Arctic for more effective planning by decision-makers, communities, and industry. What is SNAP?

  4. Global Circulation Models (GCMs) Complex coupled models created by national and international labs Interactions of oceans, atmosphere, and radiation balance Calculated which 5 of 15 models were most accurate in the far north A1B, B1 and A2 emissions scenarios Temperature and precipitation projections by month to 2100 Measuring and modeling change

  5. Baseline values = PRISM mean monthly precipitation and temperature, 771m, 1971-2000 Adjusted and interpolated GCM outputs to historical baseline Effectively removed model biases while scaling down the GCM projections Downscaling GCM output (ECHAM5) 2.5 x 2.5 degrees Frankenberg et al., Science, Sept. 11, 2009

  6. Climate Model Selection • A composite (average) of all five was used, to minimize model bias • The A2 emission scenario was selected (considered fairly probable) with some cross-comparison to A1B (more conservative). • Monthly decadal averages were used (2020s, 2050s, and 2060s), in order to reduce error due to the stochastic nature of GCM outputs • A historical baseline period of 1971-2000 was selected, to offer congruency across all SNAP-linked models. • The finest-scale (771 m) outputs were used, based on AR4 GCMs, again to provide consistency.

  7. Conceptual model of downscaled climate products Monthly projected data, temp and precip, to 2100, for 3 emission scenarios Global Circulation Models (AR4) 5 highest –performing models Downscaling with PRISM 1971-2000 baseline, 771m resolution GCM selection Data selection Data processing 5-model composite, A2 and A1B scenarios, baseline plus 2020s, 2050s, 2060s (decadal averages) Selection of key variables pertinent to CEs Inputs to ALFRESCO, Cliomes model, and GIPL permafrost model; creation of freeze, thaw, and season length interpolations Data selection Climate Model See model schematics for full inputs Fire Model Permafrost Model Biome shift model

  8. Baseline climate across ecoregional landscape • Between 1949 and 1998, mean temperature increased throughout Alaska • Trends in precipitation are less clear, due to higher variability • Both temperature and precipitation varied considerably from year to year across the historical reference period • This natural variability must be taken into account when considering ongoing and future climate trends

  9. Baseline mean temperatures Typically, the YKL ecoregion is warmest in the south in autumn, winter, and spring. However, in the summer, this pattern is reversed, with the hottest temperatures occurring to the north. This is a result of the moderating effects of the ocean and the relatively more extreme climate in interior regions.

  10. Watershed boundaries used for spatial analysis Third-level HUCs proved to the b

  11. Baseline temperature by ecoregion

  12. Baseline precipitation Typically, the YKL ecoregion is driest in the north in all seasons. However, precipitation varies quite widely across the ecoregion, from less than 40 mm per month to more than 170 mm. Summer rainfall is particularly variable.

  13. Baseline precipitation by ecoregion

  14. Projected Climate January temperature for current and three future decades, A2 scenaro (right) and A1B (below).

  15. January temperature by ecoregion

  16. Projected Climate July temperature for current and three future decades, A2 scenaro (right) and A1B (below).

  17. July temperature by ecoregion

  18. Projected Climate Summer (left) and winter precipitation for current and three future decades

  19. Projected Climate Mean annual precipitation for current and three future decades, A2 scenaro (right) and A1B (below).

  20. Mean annual precipitation by ecoregion

  21. Projected Climate Data at which the running mean temperature crosses the freezing point in the autumn. (Statewide context provides a range of reference).

  22. Date of freeze by ecoregion

  23. Projected Climate Data at which the running mean temperature crosses the freezing point in the srping. (Statewide context provides a range of reference).

  24. Date of thaw by ecoregion

  25. Permafrost: driver of change • Permafrost thaw is both a result of climate change, and a change agent in its own right • In permafrost areas, the formation and drainage of thermokarst lakes plays a key role in the hydrologic dynamics of the ecosystem • Permafrost thaw leads to multiple effects, including frost heaves, pits, gullies, differential tussock growth, localized drying, and changes in shrub and moss species abundance, productivity, and mortality • Permafrost degradation can occur in many different ways, depending on slope, soil texture, hydrology, and ice content, and each of these modes has different effects on ecosystems, human activities, infrastructure, and energy fluxes Torre Jorgenson

  26. Permafrost Modeling • Permafrost modeling was done using SNAP climate projections as described under climate modeling, and the Geophysical Institute Permafrost Lab (GIPL) permafrost model for Alaska • Model outputs include mean annual ground temperature (MAGT) and active layer thickness (ALT) • Algorithms are dependent on the insulating properties of varying ground cover and soil types, as well as on climate variables • Resolution is 1-2km • Although very fine-scale changes in micro-conditions cannot be accurately predicted by the GIPL model, outputs provide a general picture of areas likely to undergo some degree of thaw and associated hydrologic changes

  27. Conceptual model of GIPL permafrost modeling techniques

  28. Schematic of GIPL model

  29. Schematic of MQs related to permafrost • What are the current soil thermal regime dynamics? • Based on the predictions of the best available climate models and soil temperature models, how will soil thermal regimes change in the future? • Where are predicted changes in soil thermal regimes associated with communities and transportation routes? • How and where will changes in permafrost impact vegetation? • How might changes in temperature, precipitation, evapotranspiration, and soil thermal dynamics affect general hydrology and hydrology-dependent CEs such as waterfowl in the region?

  30. Permafrost: MAGT Mean annual ground temperature at one meter depth serves as a reasonable proxy for the presence/absence of ecologically significant permafrost. Blue areas are frozen; white to orange areas are thawed.

  31. Permafrost: MAGT

  32. Permafrost: ALT These maps depict two different variables. In areas with permafrost (temperatures below freezing at one meter depth), the brown shades show seasonal thaw. Blue shades show depth of winter freeze in non-permafrost areas.

  33. Predicted changes in soil thermal regime • Permafrost is expected to undergo significant thaw across much of the REA as mean annual ground temperature at one meter depth rises from below 0°C to above 0°C • Note that thaw at one meter does not equate with total permafrost loss, since deeper permafrost is likely to persist much longer, with a talik layer above it • In addition, areas that are already without permafrost are likely to experience shallower winter freezing, and areas that retain permafrost throughout the study period are likely to experience deeper summer thaw (thicker active layer)

  34. Fire Assessment and Modeling • Fire is being modeled using SNAP climate data and the ALFRESCO model in the larger context of a projected future fire regime and its effects on major vegetation classes • Climate projections, past fire history, and current vegetation patterns will be used to model patterns of fire frequency across the landscape. • Fire behavior involves stochastic elements such as the exact location of lightning strikes and the variability of weather patterns at finer time-scales than are available • Therefore, fire distribution per se will not be modeled; rather its projected average frequency across the landscape will be used to model changes in vegetation patterns and distribution

  35. MQs related to fire • What is the fire history of the ecoregion? • What climatic conditions are likely to result in significant changes to fire activity? • What is the current frequency (return interval) and the likely future frequency for fire in the ecoregion and broad sub-regions?

  36. Conceptual model of ALFRESCO fire simulation methodology

  37. ALFRESCO 1.0

  38. ALFRESCO 2.0

  39. ALFRESCO 2.0

  40. Cumulative Area Burned Historical (1950-2011) ALFRESCO replicates

  41. Cumulative Area Burned Historical (1950-2011) ALFRESCO replicates

  42. ALFRESCO 2.0 • Alaska Frame-Based Ecosystem Code • Spatially explicit state & transition model • Model is driven by disturbance & climate • Historical climate data are derived from CRU • Projected climate data are derived individually for the 5 best models and the A2 emission scenario (ALFRESCO cannot use composite model because variability is too low for calibrations) • Simulates fire & vegetation succession dynamics at a 1-km spatial resolution on a 1-year time step • Transitions constrained based on observed changes • Results averaged across 100 model runs per climate model = 500 model runs.

  43. Fire history and current fire regime • Fire frequency is dependent not only on the flammability of the landscape, but also on fire ignitions from lightning • Although lightning strikes are tracked by the Alaska Fire Service accuracy of measurement has been inconsistent over time, meaning that no consistent trends can be found in historical data • In some cases, climate change appears to be positively correlated with increased cloud-to-ground lightning activity

  44. Sample lightning ignitions

  45. Fire history from 1940 to the present(

  46. Fire regime: ongoing effects • Lichens are slow to regrow after fire • Recent decades have seen marked change in tundra ecosystems due to the interplay of climate change, wildfire, and disturbance by caribou and reindeer • Observed significant reduction of terricolous lichen ground cover and biomass • Fire can also lead to vegetation shift; in one study, it was found that shrub cover was higher on burned plots than unburned plots, and that cover of cottongrass (Eriophorumvaginatum) initially increased following the fire, and remained so for more than 14 years

  47. Tundra to Forest Transition Adapted from Epstein et al. (2004) Journal of Biogeography

  48. Tundra to Forest Transition

  49. Projections for Spatial Transitions Reclassification of NALCMS Land Cover Map (2005)

  50. Tundra to Forest Transition CCCMA ALFRESCO Replicate Basal area of White Spruce low mid high