Anthony J. Parolari

Ecohydrological modeling in agroecosystems: Examples and challenges

Human societies are increasingly altering the water and biogeochemical cycles to both improve ecosystem productivity and reduce risks associated with the unpredictable variability of climatic drivers. These alterations, however, often cause large negative environmental consequences, raising the question as to how societies can ensure a sustainable use of natural resources for the future. Here we discuss how ecohydrological modeling may address these broad questions with special attention to agroecosystems. The challenges related to modeling the two-way interaction between society and environment are illustrated by means of a dynamical model in which soil and water quality supports the growth of human society but is also degraded by excessive pressure, leading to critical transitions and sustained societal growth-collapse cycles. We then focus on the coupled dynamics of soil water and solutes (nutrients or contaminants), emphasizing the modeling challenges, presented by the strong nonlinearities in the soil and plant system and the unpredictable hydroclimatic forcing, that need to be overcome to quantitatively analyze problems of soil water sustainability in both natural and agricultural ecosystems. We discuss applications of this framework to problems of irrigation, soil salinization, and fertilization and emphasize how optimal solutions for large-scale, long-term planning of soil and water resources in agroecosystems under uncertainty could be provided by methods from stochastic control, informed by physically and mathematically sound descriptions of ecohydrological and biogeochemical interactions.

An ecohydrological perspective on drought‐induced forest mortality

Regional-scale drought-induced forest mortality events are projected to become more frequent under future climates due to changes in rainfall patterns. The occurrence of these mortality events is driven by exogenous factors such as frequency and severity of drought and endogenous factors such as tree water and carbon use strategies. To explore the link between these exogenous and endogenous factors underlying forest mortality, a stochastic ecohydrological framework that accounts for random arrival and length of droughts as well as responses of tree water and carbon balance to soil water deficit is proposed. The main dynamics of this system are characterized with respect to the spectrum of anisohydric-isohydric stomatal control strategies. Using results from a controlled drought experiment, a maximum tolerable drought length at the point where carbon starvation and hydraulic failure occur simultaneously is predicted, supporting the notion of coordinated hydraulic function and metabolism. We find qualitative agreement between the model predictions and observed regional-scale canopy dieback across a precipitation gradient during the 2002–2003 southwestern United States drought. Both the model and data suggest a rapid increase of mortality frequency below a precipitation threshold. The model also provides estimates of mortality frequency for given plant drought strategies and climate regimes. The proposed ecohydrological approach can be expanded to estimate the effect of anticipated climate change on drought-induced forest mortality and associated consequences for the water and carbon balances.

Beyond the SCS‐CN method: A theoretical framework for spatially lumped rainfall‐runoff response

Acknowledgments This work was supported through the USDA Agricultural Research Service cooperative agreement 58-6408-3-027; and National Science Foundation (NSF) grants CBET-1033467, EAR-1331846, FESD-1338694, EAR-1316258, and the Duke WISeNet grant DGE-1068871. The data used for Figure 9 are reproduced from Tedela et al. [2011, 2008]. Processed data and code are available by e-mail from the corresponding author. We thank the reviewers for their useful and constructive comments that helped improve the paper.

Stochastic rainfall-runoff model with explicit soil moisture dynamics

Stream runoff is perhaps the most poorly represented process in ecohydrological stochastic soil moisture models. Here we present a rainfall-runoff model with a new stochastic description of runoff linked to soil moisture dynamics. We describe the rainfall-runoff system as the joint probability density function (PDF) of rainfall, soil moisture and runoff forced by random, instantaneous jumps of rainfall. We develop a master equation for the soil moisture PDF that accounts explicitly for a general state-dependent rainfall-runoff transformation. This framework is then used to derive the joint rainfall-runoff and soil moisture-runoff PDFs. Runoff is initiated by a soil moisture threshold and a linear progressive partitioning of rainfall based on the soil moisture status. We explore the dependence of the PDFs on the rainfall occurrence PDF (homogeneous or state-dependent Poisson process) and the rainfall magnitude PDF (exponential or mixed-exponential distribution). We calibrate the model to 63 years of rainfall and runoff data from the Upper Little Tennessee watershed (USA) and show how the new model can reproduce the measured runoff PDF.

Tapping Environmental History to Recreate America’s Colonial Hydrology

To properly remediate, improve, or predict how hydrological systems behave, it is vital to establish their histories. However, modern-style records, assembled from instrumental data and remote sensing platforms, hardly exist back more than a few decades. As centuries of data is preferable given multidecadal fluxes of both meteorology/climatology and demographics, building such a history requires resources traditionally considered only useful in the social sciences and humanities. In this Feature, Pastore et al. discuss how they have undertaken the synthesis of historical records and modern techniques to understand the hydrology of the Northeastern U.S. from Colonial times to modern day. Such approaches could aid studies in other regions that may require heavier reliance on qualitative narratives. Further, a better insight as to how historical changes unfolded could provide a “past is prologue” methodology to increase the accuracy of predictive environmental models.

Framework for event‐based semidistributed modeling that unifies the SCS‐CN method, VIC, PDM, and TOPMODEL

Hydrologists and engineers may choose from a range of semidistributed rainfall-runoff models such as VIC, PDM, and TOPMODEL, all of which predict runoff from a distribution of watershed properties. However, these models are not easily compared to event-based data and are missing ready-to-use analytical expressions that are analogous to the SCS-CN method. The SCS-CN method is an event-based model that describes the runoff response with a rainfall-runoff curve that is a function of the cumulative storm rainfall and antecedent wetness condition. Here, we develop an event-based probabilistic storage framework and distill semidistributed models into analytical, event-based expressions for describing the rainfall-runoff response. The event-based versions called VICx, PDMx, and TOPMODELx also are extended with a spatial description of the runoff concept of “pre-threshold” and “threshold-excess” runoff, which occur, respectively, before and after infiltration exceeds a storage capacity threshold. For total storm rainfall and antecedent wetness conditions, the resulting ready-to-use analytical expressions define the source areas (fraction of the watershed) that produce runoff by each mechanism. They also define the probability density function (PDF) representing the spatial variability of runoff depths that are cumulative values for the storm duration, and the average unit area runoff, which describes the so-called runoff curve. These new event-based semidistributed models and the traditional SCS-CN method are unified by the same general expression for the runoff curve. Since the general runoff curve may incorporate different model distributions, it may ease the way for relating such distributions to land use, climate, topography, ecology, geology, and other characteristics. This article is protected by copyright. All rights reserved.

Fertilization effects on the ecohydrology of a southern California annual grassland

Nitrogen limits leaf gas exchange, canopy development, and evapotranspiration in many ecosystems. In dryland ecosystems, it is unclear whether increased anthropogenic nitrogen inputs alter the widely recognized dominance of water and energy constraints on ecohydrology. We use observations from a factorial irrigation and fertilization experiment in a nitrogen-limited southern California annual grassland to explore this hypothesis. Our analysis shows growing season soil moisture and canopy-scale water vapor conductance are equivalent in control and fertilized plots. This consistency arises as fertilization-induced increases in leaf area index (LAI) are offset by reduced leaf area-based stomatal conductance,g s . We interpret this as evidence of a hydraulic feedback between LAI, plant water status, and g s , not commonly implemented in evapotranspiration models. These results support the notion that canopy physiology and structure are coordinated in water-limited ecosystems to maintain a transpiration flux tightly controlled by hydraulic constraints in the soil-vegetation-atmosphere pathway.

Reply to Comment by Fred L. Ogden et al. on "Beyond the SCS-CN Method: A Theoretical Framework for Spatially Lumped Rainfall-Runoff Response"

Though Ogden et al. list several shortcomings of the original SCS-CN method, fit for purpose is a key consideration in hydrological modelling, as shown by the adoption of SCS-CN method in many design standards. The theoretical framework of Bartlett et al. [2016a] reveals a family of semidistributed models, of which the SCS-CN method is just one member. Other members include event-based versions of the Variable Infiltration Capacity (VIC) model and TOPMODEL. This general model allows us to move beyond the limitations of the original SCS-CN method under different rainfall-runoff mechanisms and distributions for soil and rainfall variability. Future research should link this general model approach to different hydrogeographic settings, in line with the call for action proposed by Ogden et al.

Accounting for landscape heterogeneity improves spatial predictions of tree vulnerability to drought

As climate change continues, forest vulnerability to droughts and heatwaves is increasing, but vulnerability varies regionally and locally through landscape position. Also, most models used in forecasting forest responses to heat and drought do not incorporate relevant spatial processes. In order to improve spatial predictions of tree vulnerability, we employed a nonlinear stochastic model of soil moisture dynamics accounting for landscape differences in aspect, topography and soils. Across a watershed in central Texas we modeled dynamic water stress for a dominant tree species, Juniperus ashei, and projected future dynamic water stress through the 21st century. Modeled dynamic water stress tracked spatial patterns of remotely sensed drought-induced canopy loss. Accuracy in predicting drought-impacted stands increased from 60%, accounting for spatially variable soil conditions, to 72% when also including lateral redistribution of water and radiation/temperature effects attributable to aspect. Our analysis also suggests that dynamic water stress will increase through the 21st century, with trees persisting at only selected microsites. Favorable microsites/refugia may exist across a landscape where trees can persist; however, if future droughts are too severe, the buffering capacity of an heterogeneous landscape could be overwhelmed. Incorporating spatial data will improve projections of future tree water stress and identification of potential resilient refugia.