Wetting-induced soil CO2 emission pulses are driven by interactions among soil temperature, carbon, and nitrogen limitation in the Colorado Desert.

Warming-induced changes in precipitation regimes, coupled with anthropogenically enhanced nitrogen (N) deposition, are likely to increase the prevalence, duration, and magnitude of soil respiration pulses following wetting via interactions among temperature and carbon (C) and N availability. Quantifying the importance of these interactive controls on soil respiration is a key challenge as pulses can be large terrestrial sources of atmospheric carbon dioxide (CO2 ) over comparatively short timescales. Using an automated sensor system, we measured soil CO2 flux dynamics in the Colorado Desert-a system characterized by pronounced transitions from dry-to-wet soil conditions-through a multi-year series of experimental wetting campaigns. Experimental manipulations included combinations of C and N additions across a range of ambient temperatures and across five sites varying in atmospheric N deposition. We found soil CO2 pulses following wetting were highly predictable from peak instantaneous CO2 flux measurements. CO2 pulses consistently increased with temperature, and temperature at time of wetting positively correlated to CO2 pulse magnitude. Experimentally adding N along the N deposition gradient generated contrasting pulse responses: adding N increased CO2 pulses in low N deposition sites, whereas adding N decreased CO2 pulses in high N deposition sites. At a low N deposition site, simultaneous additions of C and N during wetting led to the highest observed soil CO2 fluxes reported globally at 299.5 µmol CO2  m-2  s-1 . Our results suggest that soils have the capacity to emit high amounts of CO2 within small timeframes following infrequent wetting, and pulse sizes reflect a non-linear combination of soil resource and temperature interactions. Importantly, the largest soil CO2 emissions occurred when multiple resources were amended simultaneously in historically resource-limited desert soils, pointing to regions experiencing simultaneous effects of desertification and urbanization as key locations in future global C balance.

Irrigated urban trees exhibit greater functional trait plasticity compared to natural stands.

Urbanization creates novel ecosystems comprised of species assemblages and environments with no natural analogue. Moreover, irrigation can alter plant function compared to non-irrigated systems. However, the capacity of irrigation to alter functional trait patterns across multiple species is unknown but may be important for the dynamics of urban ecosystems. We evaluated the hypothesis that urban irrigation influences plasticity in functional traits by measuring carbon-gain and water-use traits of 30 tree species planted in Southern California, USA spanning a coastal-to-desert gradient. Tree species respond to irrigation through increasing the carbon-gain trait relationship of leaf nitrogen per specific leaf area compared to their native habitat. Moreover, most species shift to a water-use strategy of greater water loss through stomata when planted in irrigated desert-like environments compared to coastal environments, implying that irrigated species capitalize on increased water availability to cool their leaves in extreme heat and high evaporative demand conditions. Therefore, irrigated urban environments increase the plasticity of trait responses compared to native ecosystems, allowing for novel response to climatic variation. Our results indicate that trees grown in water-resource-rich urban ecosystems can alter their functional traits plasticity beyond those measured in native ecosystems, which can lead to plant trait dynamics with no natural analogue.

Impacts of Soil NOx Emission on O3 Air Quality in Rural California.

Nitrogen oxides (NOx) are a key precursor in O3 formation. Although stringent anthropogenic NOx emission controls have been implemented since the early 2000s in the United States, several rural regions of California still suffer from O3 pollution. Previous findings suggest that soils are a dominant source of NOx emissions in California; however, a statewide assessment of the impacts of soil NOx emission (SNOx) on air quality is still lacking. Here we quantified the contribution of SNOx to the NOx budget and the effects of SNOx on surface O3 in California during summer by using WRF-Chem with an updated SNOx scheme, the Berkeley Dalhousie Iowa Soil NO Parameterization (BDISNP). The model with BDISNP shows a better agreement with TROPOMI NO2 columns, giving confidence in the SNOx estimates. We estimate that 40.1% of the state's total NOx emissions in July 2018 are from soils, and SNOx could exceed anthropogenic sources over croplands, which accounts for 50.7% of NOx emissions. Such considerable amounts of SNOx enhance the monthly mean NO2 columns by 34.7% (53.3%) and surface NO2 concentrations by 176.5% (114.0%), leading to an additional 23.0% (23.2%) of surface O3 concentration in California (cropland). Our results highlight the cobenefits of limiting SNOx to help improve air quality and human health in rural California.

The influence of vegetation, mesoclimate and meteorology on urban atmospheric microclimates across a coastal to desert climate gradient.

Many cities are increasing vegetation in part due to the potential for microclimate cooling. However, the magnitude of vegetation cooling and sensitivity to mesoclimate and meteorology are uncertain. To improve understanding of the variation in vegetation's influence on urban microclimates we asked: how do meso- and regional-scale drivers influence the magnitude and timing of vegetation-based moderation on summertime air temperature (Ta), relative humidity (RH) and heat index (HI) across dryland cities? To answer this question we deployed a network of 180 temperature sensors in summer 2015 over 30 high- and 30 low-vegetated plots in three cities across a coastal to inland to desert climate gradient in southern California, USA. In a followup study, we deployed a network of temperature and humidity sensors in the inland city. We found negative Ta and HI and positive RH correlations with vegetation intensity. Furthermore, vegetation effects were highest in evening hours, increasing across the climate gradient, with reductions in Ta and increases in RH in low-vegetated plots. Vegetation increased temporal variability of Ta, which corresponds with increased nighttime cooling. Increasing mean Ta was associated with higher spatial variation in Ta in coastal cities and lower variation in inland and desert cities, suggesting a climate dependent switch in vegetation sensitivity. These results show that urban vegetation increases spatiotemporal patterns of microclimate with greater cooling in warmer environments and during nighttime hours. Understanding urban microclimate variation will help city planners identify potential risk reductions associated with vegetation and develop effective strategies ameliorating urban microclimate.

Multivariate regulation of soil CO2 and N2 O pulse emissions from agricultural soils.

Climate and land-use models project increasing occurrence of high temperature and water deficit in both agricultural production systems and terrestrial ecosystems. Episodic soil wetting and subsequent drying may increase the occurrence and magnitude of pulsed biogeochemical activity, affecting carbon (C) and nitrogen (N) cycles and influencing greenhouse gas (GHG) emissions. In this study, we provide the first data to explore the responses of carbon dioxide (CO2 ) and nitrous oxide (N2 O) fluxes to (i) temperature, (ii) soil water content as percent water holding capacity (%WHC), (iii) substrate availability throughout, and (iv) multiple soil drying and rewetting (DW) events. Each of these factors and their interactions exerted effects on GHG emissions over a range of four (CO2 ) and six (N2 O) orders of magnitude. Maximal CO2 and N2 O fluxes were observed in environments combining intermediate %WHC, elevated temperature, and sufficient substrate availability. Amendments of C and N and their interactions significantly affected CO2 and N2 O fluxes and altered their temperature sensitivities (Q10 ) over successive DW cycles. C amendments significantly enhanced CO2 flux, reduced N2 O flux, and decreased the Q10 of both. N amendments had no effect on CO2 flux and increased N2 O flux, while significantly depressing the Q10 for CO2 , and having no effect on the Q10 for N2 O. The dynamics across DW cycles could be attributed to changes in soil microbial communities as the different responses to wetting events in specific group of microorganisms, to the altered substrate availabilities, or to both. The complex interactions among parameters influencing trace gas fluxes should be incorporated into next generation earth system models to improve estimation of GHG emissions.

Quantifying the timescales over which exogenous and endogenous conditions affect soil respiration.

Understanding how exogenous and endogenous factors and above-ground-below-ground linkages modulate carbon dynamics is difficult because of the influences of antecedent conditions. For example, there are variable lags between above-ground assimilation and below-ground efflux, and the duration of antecedent periods are often arbitrarily assigned. Nonetheless, developing models linking above- and below-ground processes is crucial for estimating current and future carbon dynamics. We collected data on leaf-level photosynthesis (Asat ) and soil respiration (Rsoil ) in different microhabitats (under shrubs vs under bunchgrasses) in the Sonoran Desert. We evaluated timescales over which endogenous and exogenous factors control Rsoil by analyzing data in the context of a semimechanistic temperature-response model of Rsoil that incorporated effects of antecedent exogenous (soil water) and endogenous (Asat ) conditions. For both microhabitats, antecedent soil water and Asat significantly affected Rsoil , but Rsoil under shrubs was more sensitive to Asat than that under bunchgrasses. Photosynthetic rates 1 and 3 d before the Rsoil measurement were most important in determining current-day Rsoil under bunchgrasses and shrubs, respectively, indicating a significant lag effect. Endogenous and exogenous controls are critical drivers of Rsoil , but the relative importance and the timescale over which each factor affects Rsoil depends on above-ground vegetation and ecosystem structure characteristics.

Research needs on the biodiversity-ecosystem functioning relationship in drylands.

Research carried out in drylands over the last decade has provided major insights on the biodiversity-ecosystem functioning relationship (BEFr) and about how biodiversity interacts with other important factors, such as climate and soil properties, to determine ecosystem functioning and services. Despite this, there are important gaps in our understanding of the BEFr in drylands that should be addressed by future research. In this perspective we highlight some of these gaps, which include: 1) the need to study the BEFr in bare soils devoid of perennial vascular vegetation and biocrusts, a major feature of dryland ecosystems, 2) evaluating how intra-specific trait variability, a key but understudied facet of functional diversity, modulate the BEFr, 3) addressing the influence of biotic interactions on the BEFr, including plant-animal interactions and those between microorganisms associated to biocrusts, 4) studying how differences in species-area relationships and beta diversity are associated with ecosystem functioning, and 5) considering the role of temporal variability and human activities, both present and past, particularly those linked to land use (e.g., grazing) and urbanization. Tackling these gaps will not only advance our comprehension of the BEFr but will also bolster the effectiveness of management and ecological restoration strategies, crucial for safeguarding dryland ecosystems and the livelihoods of their inhabitants.

Precipitation legacies amplify ecosystem nitrogen losses from nitric oxide emissions in a Pinyon-Juniper dryland.

Climate change is increasing the variability of precipitation, altering the frequency of soil drying-wetting events and the distribution of seasonal precipitation. These changes in precipitation can alter nitrogen (N) cycling and stimulate nitric oxide (NO) emissions (an air pollutant at high concentrations), which may vary according to legacies of past precipitation and represent a pathway for ecosystem N loss. To identify whether precipitation legacies affect NO emissions, we excluded or added precipitation during the winter growing season in a Pinyon-Juniper dryland and measured in situ NO emissions following experimental wetting. We found that the legacy of both excluding and adding winter precipitation increased NO emissions early in the following summer; cumulative NO emissions from the winter precipitation exclusion plots (2750 ± 972 µg N-NO m-2 ) and winter water addition plots (2449 ± 408 µg N-NO m-2 ) were higher than control plots (1506 ± 397 µg N-NO m-2 ). The increase in NO emissions with previous precipitation exclusion was associated with inorganic N accumulation, while the increase in NO emissions with previous water addition was associated with an upward trend in microbial biomass. Precipitation legacies can accelerate soil NO emissions and may amplify ecosystem N loss in response to more variable precipitation.

Bacterial denitrification drives elevated N2O emissions in arid southern California drylands.

Soils are the largest source of atmospheric nitrous oxide (N2O), a powerful greenhouse gas. Dry soils rarely harbor anoxic conditions to favor denitrification, the predominant N2O-producing process, yet, among the largest N2O emissions have been measured after wetting summer-dry desert soils, raising the question: Can denitrifiers endure extreme drought and produce N2O immediately after rainfall? Using isotopic and molecular approaches in a California desert, we found that denitrifiers produced N2O within 15 minutes of wetting dry soils (site preference = 12.8 ± 3.92 per mil, δ15Nbulk = 18.6 ± 11.1 per mil). Consistent with this finding, we detected nitrate-reducing transcripts in dry soils and found that inhibiting microbial activity decreased N2O emissions by 59%. Our results suggest that despite extreme environmental conditions-months without precipitation, soil temperatures of ≥40°C, and gravimetric soil water content of <1%-bacterial denitrifiers can account for most of the N2O emitted when dry soils are wetted.

Differential daytime and night-time stomatal behavior in plants from North American deserts.

Night-time stomatal conductance (g(night)) occurs in many ecosystems, but the g(night) response to environmental drivers is relatively unknown, especially in deserts. Here, we conducted a Bayesian analysis of stomatal conductance (g) (N=5013) from 16 species in the Sonoran, Chihuahuan, Mojave and Great Basin Deserts (North America). We partitioned daytime g (g(day)) and g(night) responses by describing g as a mixture of two extreme (dark vs high light) behaviors. Significant g(night) was observed across 15 species, and the g(night) and g(day) behavior differed according to species, functional type and desert. The transition between extreme behaviors was determined by light environment, with the transition behavior differing between functional types and deserts. Sonoran and Chihuahuan C(4) grasses were more sensitive to vapor pressure difference (D) at night and soil water potential (Ψ(soil)) during the day, Great Basin C(3) shrubs were highly sensitive to D and Ψ(soil) during the day, and Mojave C(3) shrubs were equally sensitive to D and Ψ(soil) during the day and night. Species were split between the exhibition of isohydric or anisohydric behavior during the day. Three species switched from anisohydric to isohydric behavior at night. Such behavior, combined with differential D, Ψ(soil) and light responses, suggests that different mechanisms underlie g(day) and g(night) regulation.

Soil metabolic pulses: water, substrate, and biological regulation.

Pulses of metabolic activity are a common ecological response to intermittently available resources, and in soils these pulses often occur in response to wetting. To better understand variation in soil pulses, we conducted a distributed field experiment at seven sites along a 2200-m elevation transect in southern California, USA. Treatments included both water and water + substrate additions and two measurements of soil respiration within one hour. These experiments were repeated 11 times throughout 2009-2010. Additions of substrate led to consistently higher pulse fluxes, exceeding 10 micromol CO2 x m(-2( x s(-1), than additions of water alone. These results support a sequential limitation by two resources where an initial limiting resource acts as a switch and, after activation, processes are regulated by a second resource. In contrast to general expectations of increasing pulses with higher soil organic matter (SOM), pulses exhibited strong scale dependencies. Pulses during the summer period and SOM were correlated positively within sites and negatively between sites. This cross-scale divergence implies that, at low elevations, the proportion of SOM available for pulse metabolism was a much larger fraction than at higher elevations. With expected climate changes leading to more frequent drying-wetting cycles, regulation of metabolic pulses will increasingly influence long-term biogeochemical dynamics.

Urban landcover differentially drives day and nighttime air temperature across a semi-arid city.

Semi-arid urban environments are undergoing an increase in both average air temperatures and in the frequency and intensity of extreme heat events. Within cities, different composition and densities of urban landcovers (ULC) influence local air temperatures, either mitigating or increasing heat. Currently, understanding how combinations of ULC influence air temperature at the block to neighborhood scale is necessary for heat mitigation plans, and yet limited due to the complexities integrating high-resolution ULC with spatial and temporally high-resolution microclimate data. We quantify how ULC influences air temperature at 60 m resolution for day and nighttime climate normals and extreme heat conditions by integrating microclimate sensor data sensor and high-resolution (1 m2) ULC for Denver, Colorado's urban core. We derive ULC drivers of air temperature using a structural equation model, then use a random forest algorithm to predict air temperatures for 30-year climate normals and an extreme heat condition. We find that, in conjunction with other ULC, urban tree canopy reduces daytime air temperatures (-0.026 °C per % cover), and the combination of impervious surfaces and buildings increases daytime air temperature (0.021 °C per % cover). Compared to daytime hours, nighttime irrigated turf temperature cooling effects are increased from being non-significant to -0.022 °C per % cover, while tree canopy effects are reduced from -0.026 °C during the day to -0.016 °C at night. Overall, ULC drives ~17% and 25% of local air temperature during the day and night, respectively. ULC influence on daytime air temperatures is altered in extreme heat events, both depending on the ULC type and time of day. Our findings inform urban planners seeking to identify potential hot and cool spots within a semi-arid city and mitigate high urban air temperatures through using ULC within larger urban climate mitigation strategies.

Resolving and Predicting Neighborhood Vulnerability to Urban Heat and Air Pollution: Insights From a Pilot Project of Community Science.

Urban heat and air pollution, two environmental threats to urban residents, are studied via a community science project in Los Angeles, CA, USA. The data collected, for the first time, by community members, reveal the significance of both the large spatiotemporal variations of and the covariations between 2 m air temperature (2mT) and ozone (O3) concentration within the (4 km) neighborhood scale. This neighborhood variation was not exhibited in either daily satellite observations or operational model predictions, which makes the assessment of community health risks a challenge. Overall, the 2mT is much better predicted than O3 by the weather and research forecast model with atmospheric chemistry (WRF-Chem). For O3, diurnal variation is better predicted by WRF-Chem than spatial variation (i.e., underestimated by 50%). However, both WRF-chem and the surface observation show the overall consistency in describing statistically significant covariations between O3 and 2mT. In contrast, satellite-based land surface temperature at 1 km resolution is insufficient to capture air temperature variations at the neighborhood scale. Community engagement is augmented with interactive maps and apps that show the predictions in near real time and reveals the potential of green canopy to reduce air temperature and ozone; but different tree types and sizes may lead to different impacts on air temperature, which is not resolved by the WRF-Chem. These findings highlight the need for community science engagement to reveal otherwise impossible insights for models, observations, and real-time dissemination to understand, predict, and ultimately mitigate, urban neighborhood vulnerability to heat and air pollution.

Ecosystem services and urban heat riskscape moderation: water, green spaces, and social inequality in Phoenix, USA.

Urban ecosystems are subjected to high temperatures--extreme heat events, chronically hot weather, or both-through interactions between local and global climate processes. Urban vegetation may provide a cooling ecosystem service, although many knowledge gaps exist in the biophysical and social dynamics of using this service to reduce climate extremes. To better understand patterns of urban vegetated cooling, the potential water requirements to supply these services, and differential access to these services between residential neighborhoods, we evaluated three decades (1970-2000) of land surface characteristics and residential segregation by income in the Phoenix, Arizona, USA metropolitan region. We developed an ecosystem service trade-offs approach to assess the urban heat riskscape, defined as the spatial variation in risk exposure and potential human vulnerability to extreme heat. In this region, vegetation provided nearly a 25 degrees C surface cooling compared to bare soil on low-humidity summer days; the magnitude of this service was strongly coupled to air temperature and vapor pressure deficits. To estimate the water loss associated with land-surface cooling, we applied a surface energy balance model. Our initial estimates suggest 2.7 mm/d of water may be used in supplying cooling ecosystem services in the Phoenix region on a summer day. The availability and corresponding resource use requirements of these ecosystem services had a strongly positive relationship with neighborhood income in the year 2000. However, economic stratification in access to services is a recent development: no vegetation-income relationship was observed in 1970, and a clear trend of increasing correlation was evident through 2000. To alleviate neighborhood inequality in risks from extreme heat through increased vegetation and evaporative cooling, large increases in regional water use would be required. Together, these results suggest the need for a systems evaluation of the benefits, costs, spatial structure, and temporal trajectory for the use of ecosystem services to moderate climate extremes. Increasing vegetation is one strategy for moderating regional climate changes in urban areas and simultaneously providing multiple ecosystem services. However, vegetation has economic, water, and social equity implications that vary dramatically across neighborhoods and need to be managed through informed environmental policies.

Beyond 'trees are good': Disservices, management costs, and tradeoffs in urban forestry.

The provision of ecosystem services is a prominent rationale for urban greening, and there is a prevailing mantra that 'trees are good'. However, understanding how urban trees contribute to sustainability must also consider disservices. In this perspective article, we discuss recent research on ecosystem disservices of urban trees, including infrastructure conflicts, health and safety impacts, aesthetic issues, and environmentally detrimental consequences, as well as management costs related to ecological disturbances and risk management. We also discuss tradeoffs regarding species selection and local conservation concerns, as well as the central role of human perception in the interpretation of ecosystem services and disservices, particularly the uncritical assertion that 'everybody loves trees'. Urban forestry decision-making that fails to account for disservices can have unintended negative consequences for communities. Further research is needed regarding life cycle assessments, stakeholder decision-making, return-on-investment, and framings of services and disservices in urban forestry.

Michaelis-Menten kinetics of soil respiration feedbacks to nitrogen deposition and climate change in subtropical forests.

China experiences some of the highest rates of anthropogenic nitrogen deposition globally, with further increases projected. Understanding of soil feedbacks to the combined anthropogenic influences of climate change and nitrogen deposition in these systems is critical to improve predictive abilities for future climate scenarios. Here we used a Michaelis-Menten substrate-based kinetics framework to explore how soil CO2 production (Rsoil) responds to changes in temperature and available soil nitrogen (N) by combining field experiments with laboratory manipulations from sites experiencing elevated rates of anthropogenic N deposition but varying in soil N availabiltiy. The temperature sensitivity of Rsoil was strongly influenced by labile C additions. Furthermore, estimation of the temperature response of the Michaelis-Menten parameters supports the use of substrate-based kinetics in modeling efforts. Results from both field and laboratory experiments demonstrated a general decrease in Rsoil with increasing soil available N that was variably dependent on carbon (C) availability. Both the field and the laboratory measurements demonstrated a consistent decrease in the Michaelis-Menten parameter kM with increasing soil available N, indicating an increase in the efficiency of soil C decomposition with increasing N. Furthermore, these results provide evidence of interactions between N deposition and temperature sensitivity, which could influence C storage under combined anthropogenic global change drivers.

Variation in the urban vegetation, surface temperature, air temperature nexus.

Our study examines the urban vegetation - air temperature (Ta) - land surface temperature (LST) nexus at micro- and regional-scales to better understand urban climate dynamics and the uncertainty in using satellite-based LST for characterizing Ta. While vegetated cooling has been repeatedly linked to reductions in urban LST, the effects of vegetation on Ta, the quantity often used to characterize urban heat islands and global warming, and on the interactions between LST and Ta are less well characterized. To address this need we quantified summer temporal and spatial variation in Ta through a network of 300 air temperature sensors in three sub-regions of greater Los Angeles, CA, which spans a coastal to desert climate gradient. Additional sensors were placed within the inland sub-region at two heights (0.1m and 2m) within three groundcover types: bare soil, irrigated grass, and underneath citrus canopy. For the entire study region, we acquired new imagery data, which allowed calculation of the normalized difference vegetation index (NDVI) and LST. At the microscale, daytime Ta measured along a vertical gradient, ranged from 6 to 3°C cooler at 0.1 and 2m, underneath tall canopy compared to bare ground respectively. At the regional scale NDVI and LST were negatively correlated (p<0.001). Relationships between diel variation in Ta and daytime LST at the regional scale were progressively weaker moving away from the coast and were generally limited to evening and nighttime hours. Relationships between NDVI and Ta were stronger during nighttime hours, yet effectiveness of mid-day vegetated cooling increased substantially at the most arid region. The effectiveness of vegetated Ta cooling increased during heat waves throughout the region. Our findings suggest an important but complex role of vegetation on LST and Ta and that vegetation may provide a negative feedback to urban climate warming.

Urban legacies and soil management affect the concentration and speciation of trace metals in Los Angeles community garden soils.

Heavy metals in urban soils can compromise human health, especially in urban gardens, where gardeners may ingest contaminated dust or crops. To identify patterns of urban garden metal contamination, we measured concentrations and bioavailability of Pb, As, and Cd in soils associated with twelve community gardens in Los Angeles County, CA. This included sequential extractions to partition metals among exchangeable, reducible, organic, or residual fractions. Proximity to road increased all metal concentrations, suggesting vehicle emissions sources. Reducible Pb increased with neighborhood age, suggesting leaded paint as a likely pollutant source. Exchangeable Cd and As both increased with road proximity. Only cultivated soils showed an increase in exchangeable As with road proximity, potentially due to reducing humic acid interactions while Cd bioavailability was mitigated by organic matter. Understanding the geochemical phases and metal bioavailability allows incorporation of contamination patterns into urban planning.

Apparent plasticity in functional traits determining competitive ability and spatial distribution: a case from desert.

Species competitive abilities and their distributions are closely related to functional traits such as biomass allocation patterns. When we consider how nutrient supply affects competitive abilities, quantifying the apparent and true plasticity in functional traits is important because the allometric relationships among traits are universal in plants. We propose to integrate the notion of allometry and the classical reaction norm into a composite theoretical framework that quantifies the apparent and true plasticity. Combining the framework with a meta-analysis, a series of field surveys and a competition experiment, we aimed to determine the causes of the dune/interdune distribution patterns of two Haloxylon species in the Gurbantonggut Desert. We found that (1) the biomass allocation patterns of both Haloxylon species in responses to environmental conditions were apparent rather than true plasticity and (2) the allometric allocation patterns affected the plants' competition for soil nutrient supply. A key implication of our results is that the apparent plasticity in functional traits of plants determines their response to environmental change. Without identifying the apparent and true plasticity, we would substantially overestimate the magnitude, duration and even the direction of plant responses in functional traits to climate change.

Multivariate analysis of the ecoregion delineation for aquatic systems.

The ecoregion concept is a popular method of understanding the spatial distribution of the environment', however, it has yet to be adequately demonstrated that the environment is distributed in accordance with these bounded units. In this paper, we generated a testable hypothesis based on the current usage of ecoregions: the ecoregion classification will allow for discrimination between lakes of different water quality. The ecoregion classification should also be more effective better than a comparably scaled classification based on political boundaries, land-use class, or random grouping. To test this hypothesis we used the Environmental Monitoring and Assessment Program (EMAP) lake water chemistry data from the northeast United States. The water chemistry data were reduced to four components using principal component analysis. For comparison to an optimal grouping of these data we used K-means cluster analysis to define the extent at which these lakes could be segregated into distinct classes. Jackknifed discriminant analysis was used to determine the classification rate of ecoregions, the three alternative spatial classification methods, and the clustering algorithm. The classification based on ecoregions was successful for 35% of the lakes included in this study, in comparison to the clustered groups accuracy of 98%. These results suggest that the large scale spatial distribution of ecosystem types is more complicated than that suggested by the present ecoregion boundaries. Further tests of ecoregion delineations are needed and alternative large-scale management strategies should be investigated.