Accuracy of tropical peat and non-peat fire forecasts enhanced by simulating hydrology.

Soil moisture deficits and water table dynamics are major biophysical controls on peat and non-peat fires in Indonesia. Development of modern fire forecasting models in Indonesia is hampered by the lack of scalable hydrologic datasets or scalable hydrology models that can inform the fire forecasting models on soil hydrologic behaviour. Existing fire forecasting models in Indonesia use weather data-derived fire probability indices, which often do not adequately proxy the sub-surface hydrologic dynamics. Here we demonstrate that soil moisture and water table dynamics can be simulated successfully across tropical peatlands and non-peatland areas by using a process-based eco-hydrology model (ecosys) and publicly available data for weather, soil, and management. Inclusion of these modelled water table depth and soil moisture contents significantly improves the accuracy of a neural network model in predicting active fires at two-weekly time scale. This constitutes an important step towards devising an operational fire early warning system for Indonesia.

Water level changes in Lake Erie drive 21st century CO2 and CH4 fluxes from a coastal temperate wetland.

Wetland water depth influences microbial and plant communities, which can alter the above- and below-ground carbon cycling of a wetland. Wetland water depths are likely to change due to shifting precipitation patterns, which will affect projections of greenhouse gas emissions; however, these effects are rarely incorporated into wetland greenhouse gas models. Seeking to address this gap, we used a mechanistic model, ecosys, to simulate a range of water depth scenarios in a temperate wetland, and analyzed simulated predictions of carbon dioxide (CO2) and methane (CH4) fluxes over the 21st century. We tested our model using eddy covariance measurements of CO2 and CH4 fluxes collected at the Old Woman Creek National Estuarine Research Reserve (OWC) during 2015 and 2016. OWC is a lacustrine, estuarine, freshwater, mineral-soil marsh. An empirical model found that the wetland water depth is highly dependent on the water depth of the nearby Lake Erie. Future wetland surface water depths were modeled based on projection of Lake Erie's water depth using four separate NOAA projections, resulting in four wetland water-depth scenarios. Two of the four 21st century projections for Lake Erie water depths used in this study indicated that the water depth of the wetland would remain nearly steady; however, the other two indicated decreases in the wetland water depth. In our scenario where the wetland dries out, we project the wetland's climatological warming effect will decrease due to smaller CH4 fluxes to the atmosphere and larger CO2 uptake by the wetland. We also found that increased water level can lower emissions by shifting the site towards more open water areas, which have lower CH4 emissions. We found that decreased water depths would cause more widespread colonization of the wetland by macrophyte vegetation. Using an empirical relationship, we also found that further drying could result in other, non-wetland vegetation to emerge, dramatically altering soil carbon cycling. In three of our four projections, we found that in general the magnitude of CO2 and CH4 fluxes steadily increase over the next 100 years in response to higher temperatures. However, in our driest simulations, we projected a different response due to increased oxidation of soil carbon, with CH4 emissions decreasing substantially from an annual cumulative peak of 224.6 to a minimum of 104.7 gC m-2 year-1. In that same simulation, net cumulative flux of CO2 changed from being a sink of 56.5 gC m-2 year-1 to a source of 369.6 gC m-2 year-1 over the same period, despite a temperature increase from 13.7 °C to 14.2 °C. This temperature shift in our other three cases with greater water depths increased the source strength of CH4 and the sink strength of CO2. We conclude that the magnitude of wetland greenhouse-gas fluxes depended on the water depth primarily as it affected the areal percentage of the wetland available for plant colonization, but dramatic decreases in water depths could cause significant reductions in the wetland CH4 fluxes, while simultaneously altering the wetland vegetation.

Sustainable irrigation based on co-regulation of soil water supply and atmospheric evaporative demand.

Irrigation is an important adaptation to reduce crop yield loss due to water stress from both soil water deficit (low soil moisture) and atmospheric aridity (high vapor pressure deficit, VPD). Traditionally, irrigation has primarily focused on soil water deficit. Observational evidence demonstrates that stomatal conductance is co-regulated by soil moisture and VPD from water supply and demand aspects. Here we use a validated hydraulically-driven ecosystem model to reproduce the co-regulation pattern. Specifically, we propose a plant-centric irrigation scheme considering water supply-demand dynamics (SDD), and compare it with soil-moisture-based irrigation scheme (management allowable depletion, MAD) for continuous maize cropping systems in Nebraska, United States. We find that, under current climate conditions, the plant-centric SDD irrigation scheme combining soil moisture and VPD, could significantly reduce irrigation water use (-24.0%) while maintaining crop yields, and increase economic profits (+11.2%) and irrigation water productivity (+25.2%) compared with MAD, thus SDD could significantly improve water sustainability.

Alaskan carbon-climate feedbacks will be weaker than inferred from short-term experiments.

Climate warming is occurring fastest at high latitudes. Based on short-term field experiments, this warming is projected to stimulate soil organic matter decomposition, and promote a positive feedback to climate change. We show here that the tightly coupled, nonlinear nature of high-latitude ecosystems implies that short-term (<10 year) warming experiments produce emergent ecosystem carbon stock temperature sensitivities inconsistent with emergent multi-decadal responses. We first demonstrate that a well-tested mechanistic ecosystem model accurately represents observed carbon cycle and active layer depth responses to short-term summer warming in four diverse Alaskan sites. We then show that short-term warming manipulations do not capture the non-linear, long-term dynamics of vegetation, and thereby soil organic matter, that occur in response to thermal, hydrological, and nutrient transformations belowground. Our results demonstrate significant spatial heterogeneity in multi-decadal Arctic carbon cycle trajectories and argue for more mechanistic models to improve predictive capabilities.

Towards a multiscale crop modelling framework for climate change adaptation assessment.

Predicting the consequences of manipulating genotype (G) and agronomic management (M) on agricultural ecosystem performances under future environmental (E) conditions remains a challenge. Crop modelling has the potential to enable society to assess the efficacy of G × M technologies to mitigate and adapt crop production systems to climate change. Despite recent achievements, dedicated research to develop and improve modelling capabilities from gene to global scales is needed to provide guidance on designing G × M adaptation strategies with full consideration of their impacts on both crop productivity and ecosystem sustainability under varying climatic conditions. Opportunities to advance the multiscale crop modelling framework include representing crop genetic traits, interfacing crop models with large-scale models, improving the representation of physiological responses to climate change and management practices, closing data gaps and harnessing multisource data to improve model predictability and enable identification of emergent relationships. A fundamental challenge in multiscale prediction is the balance between process details required to assess the intervention and predictability of the system at the scales feasible to measure the impact. An advanced multiscale crop modelling framework will enable a gene-to-farm design of resilient and sustainable crop production systems under a changing climate at regional-to-global scales.

Expansion of high-latitude deciduous forests driven by interactions between climate warming and fire.

High-latitude regions have experienced rapid warming in recent decades, and this trend is projected to continue over the twenty-first century1. Fire is also projected to increase with warming2,3. We show here, consistent with changes during the Holocene4, that changes in twenty-first century climate and fire are likely to alter the composition of Alaskan boreal forests. We hypothesize that competition for nutrients after fire in early succession and for light in late succession in a warmer climate will cause shifts in plant functional type. Consistent with observations, our ecosystem model predicts evergreen conifers to be the current dominant tree type in Alaska. However, under future climate and fire, our analysis suggests the relative dominance of deciduous broadleaf trees nearly doubles, accounting for 58% of the Alaska ecosystem's net primary productivity by 2100, with commensurate declines in contributions from evergreen conifer trees and herbaceous plants. Post-fire deciduous broadleaf tree growth under a future climate is sustained from enhanced microbial nitrogen mineralization caused by warmer soils and deeper active layers, resulting in taller trees that compete more effectively for light. The expansion of deciduous broadleaf forests will affect the carbon cycle, surface energy fluxes and ecosystem function, thereby modifying important feedbacks with the climate system.

Multimodel ensembles of wheat growth: many models are better than one.

Crop models of crop growth are increasingly used to quantify the impact of global changes due to climate or crop management. Therefore, accuracy of simulation results is a major concern. Studies with ensembles of crop models can give valuable information about model accuracy and uncertainty, but such studies are difficult to organize and have only recently begun. We report on the largest ensemble study to date, of 27 wheat models tested in four contrasting locations for their accuracy in simulating multiple crop growth and yield variables. The relative error averaged over models was 24-38% for the different end-of-season variables including grain yield (GY) and grain protein concentration (GPC). There was little relation between error of a model for GY or GPC and error for in-season variables. Thus, most models did not arrive at accurate simulations of GY and GPC by accurately simulating preceding growth dynamics. Ensemble simulations, taking either the mean (e-mean) or median (e-median) of simulated values, gave better estimates than any individual model when all variables were considered. Compared to individual models, e-median ranked first in simulating measured GY and third in GPC. The error of e-mean and e-median declined with an increasing number of ensemble members, with little decrease beyond 10 models. We conclude that multimodel ensembles can be used to create new estimators with improved accuracy and consistency in simulating growth dynamics. We argue that these results are applicable to other crop species, and hypothesize that they apply more generally to ecological system models.

Modeling topographic effects on net ecosystem productivity of boreal black spruce forests.

Current regional estimates of net primary productivity (NPP) of boreal black spruce overlook the large variation in NPP caused by small-scale topographic effects on soil water, temperature and nutrient availability. Topographic effects on black spruce NPP could likely be modeled by simulating the lateral and vertical movement of water, and its effects on soil nutrient transformation and uptake, through three-dimensional watersheds defined by aspects and slopes of their topographic positions. To examine this likelihood, the ecosystem model 'ecosys' was run for 120 years on a transect that included upper- and lower-slope positions and a basin in which a basal water table was set 0.5 m below the soil surface. For the run, we used soil properties and weather conditions recorded at the 115-year-old BOREAS Southern Old Black Spruce site. Short-term model performance was tested by comparing diurnal and annual carbon (C) transfers simulated under 1994 weather conditions during the 115th year of the model run with those measured at this site during 1994 by eddy covariance, surface chambers and allometry. After 115 years, annual spruce NPP simulated at the upper-slope positions was twice that at the basin (350 versus 170 g C m-2), whereas accumulated wood C was almost three times as large (6.8 versus 2.4 kg C m-2). In the model, increases in NPP and wood growth in upper-slope positions were caused by lower soil water contents, higher soil temperatures, and more rapid O2 uptake that accelerated heterotrophic respiration and hence nutrient mineralization and uptake. Modeled differences in wood growth with topographic position were quantitatively consistent with measurements of boreal black spruce at several research sites differing in water table depth. Modeled differences also agreed with differences in wood growth rates derived from allometric measurements at boreal black spruce sites differing in productivity indices as a result of differences in subsurface hydrology. The magnitude of these differences clearly indicates the importance of accounting for subsurface hydrology in regional estimates of boreal forest productivity.

Changes in mass and energy transfer between the canopy and the atmosphere: model development and testing with a free-air CO2 enrichment (FACE) experiment.

The rationale for this study is found in the probable higher temperatures and changes in rainfall patterns that are expected in the future as a result of increasing levels of CO2 in the atmosphere. In particular, higher air temperatures may cause an increase in evapotranspiration demand while a reduction in rainfall could increase the severity and duration of drought in arid and semi-arid regions. Representation of the water transfer scheme includes water uptake by roots and the interaction between evapotranspiration and CO2 enrichment. The predicted response of a spring wheat (Triticum aestivum L. cv. Yecora rojo) canopy in terms of energy exchange processes to elevated atmospheric CO2 level was tested against measurements collected at the FACE (Free Air Enrichment Experiment) site in 1994. Simulated and measured canopy conductances were reduced by about 30% under elevated [CO2] under optimum conditions of water supply. Reductions in latent heat fluxes under elevated instead of ambient [CO2] caused reductions in both simulated and measured seasonal water use of 6% under optimum and 2% under suboptimum irrigation. The soil-plant-atmosphere water transfer scheme proposed here offers several advances in the simulation of land surface interactions. First, the stomatal resistance model minimizes assumptions in existing land surface schemes about the effects of interactions among environmental conditions (radiation, temperature, CO2) upon stomatal behavior. These interactions are resolved in the calculation of CO2 in which processes are already well understood.