Rabin, S, Daniel S Ward, Sergey Malyshev, B I Magi, Elena Shevliakova, and Stephen W Pacala, March 2018: A fire model with distinct crop, pasture, and non-agricultural burning: Use of new data and a model-fitting algorithm for FINALv1. Geoscientific Model Development, 11(2), DOI:10.5194/gmd-11-815-2018. Abstract
This study describes and evaluates the Fire Including Natural & Agricultural Lands model (FINAL) which, for the first time, explicitly simulates cropland and pasture management fires separately from non-agricultural fires. The non-agricultural fire module uses empirical relationships to simulate burned area in a quasi-mechanistic framework, similar to past fire modeling efforts, but with a novel optimization method that improves the fidelity of simulated fire patterns to new observational estimates of non-agricultural burning. The agricultural fire components are forced with estimates of cropland and pasture fire seasonality and frequency derived from observational land-cover and satellite fire datasets. FINAL accurately simulates the amount, distribution, and seasonal timing of burned cropland and pasture over 2001–2009 (global totals: 0.434 × 106 and 2.02 × 106 km2 yr−1 modeled, 0.454 × 106 and 2.04 × 106 km2 yr−1 observed), but carbon emissions for cropland and pasture fire are overestimated (global totals: 0.297 PgC yr−1 and 0.712 PgC yr−1 modeled, 0.194 PgC yr−1 and 0.538 PgC yr−1 observed). The non-agricultural fire module underestimates global burned area (1.66 × 106 km2 yr−1 modeled, 2.44 × 106 km2 yr−1 observed) and carbon emissions (1.33 PgC yr−1 modeled, 1.84 PgC yr−1 observed). The spatial pattern of total burned area and carbon emissions is generally well reproduced across much of sub-Saharan Africa, Brazil, central Asia, and Australia, whereas the boreal zone suffers from underestimates. FINAL represents an important step in the development of global fire models, and offers a strategy for fire models to consider human-driven fire regimes on cultivated lands. At the regional scale, simulations would benefit from refinements in the parameterizations and improved optimization datasets.
Globally, fires are a major source of carbon from the terrestrial biosphere to the atmosphere, occurring on a seasonal cycle and with substantial interannual variability. To understand past trends and variability in sources and sinks of terrestrial carbon, we need quantitative estimates of global fire distributions. Here we introduce an updated version of the Fire Including Natural and Agricultural Lands model, version 2 (FINAL.2), modified to include multi-day burning and enhanced fire spread rate in forest crowns. We demonstrate that the improved model reproduces the interannual variability and spatial distribution of fire emissions reported in present day remotely sensed inventories. We use FINAL.2 to simulate historical (post-1700) fires and attribute past fire trends and variability to individual drivers: land use and land cover change, population growth, and lightning variability. Global fire emissions of carbon increase by about 10% between 1700 and 1900, reaching a maximum of 3.4 PgC yr-1 in the 1910s, followed by a decrease to about 5% below year 1700 levels by 2010. The decrease in emissions from the 1910s to the present day is driven mainly by land use change, with a smaller contribution from increased fire suppression due to increased human population, and is largest in Sub-Saharan Africa and South Asia. Interannual variability of global fire emissions is similar in the present day as in the early historical period, but present day wildfires would be more variable in the absence of land use change.
Clark, Spencer K., Daniel S Ward, and Natalie M. Mahowald, March 2017: Parameterization-based uncertainty in future lightning flash density. Geophysical Research Letters, 44(6), DOI:10.1002/2017GL073017. Abstract
In this study we implement eight lightning parameterizations in the Community Atmospheric Model (CAM5), evaluate the performance of the parameterizations in the present climate, and test the sensitivity of future lightning activity to the choice of parameterization. In the present-day, the annual mean lightning flash densities in simulations constrained by reanalysis data show the highest spatial correlation to satellite observations for parameterizations based either on cloud top height (0.83) or cold cloud depth (0.80). Under future scenarios using representative concentration pathways, changes in global mean lightning flash density are highly sensitive to the parameterization chosen, with cloud top height schemes, a cold cloud depth scheme, and a scheme based on convective mass flux projecting large increases (36% to 45%), a mild increase (12.6%), and a decrease (-6.7%) in lightning flash density respectively under the RCP8.5 scenario, which causes a 3.4 K warming between 1996-2005 and 2079-2088.
Kok, Jasper F., D A Ridley, Q Zhou, Ron L. Miller, Chun Zhao, C L Heald, and Daniel S Ward, et al., April 2017: Smaller desert dust cooling effect estimated from analysis of dust size and abundance. Nature Geoscience, 10(4), DOI:10.1038/ngeo2912. Abstract
Desert dust aerosols affect Earth’s global energy balance through direct interactions with radiation, and through indirect interactions with clouds and ecosystems. But the magnitudes of these effects are so uncertain that it remains unclear whether atmospheric dust has a net warming or cooling effect on global climate. Consequently, it is still uncertain whether large changes in atmospheric dust loading over the past century have slowed or accelerated anthropogenic climate change, or what the effects of potential future changes in dust loading will be. Here we present an analysis of the size and abundance of dust aerosols to constrain the direct radiative effect of dust. Using observational data on dust abundance, in situ measurements of dust optical properties and size distribution, and climate and atmospheric chemical transport model simulations of dust lifetime, we find that the dust found in the atmosphere is substantially coarser than represented in current global climate models. As coarse dust warms the climate, the global dust direct radiative effect is likely to be less cooling than the ~−0.4 W m−2 estimated by models in a current global aerosol model ensemble. Instead, we constrain the dust direct radiative effect to a range between −0.48 and +0.20 W m−2, which includes the possibility that dust causes a net warming of the planet.
Mahowald, Natalie M., J Randerson, Keith Lindsay, Hugh Morrison, Scott C Doney, P Lawrence, Sarah Schlunegger, and Daniel S Ward, et al., January 2017: Interactions between land use change and carbon cycle feedbacks. Global Biogeochemical Cycles, 31(1), DOI:10.1002/2016GB005374. Abstract
Using the Community Earth System Model, we explore the role of human land use and land cover change (LULCC) in modifying the terrestrial carbon budget in simulations forced by Representative Concentration Pathway 8.5, extended to year 2300. Overall, conversion of land (e.g., from forest to croplands via deforestation) results in a model-estimated, cumulative carbon loss of 490 Pg C between 1850 and 2300, larger than the 230 Pg C loss of carbon caused by climate change over this same interval. The LULCC carbon loss is a combination of a direct loss at the time of conversion and an indirect loss from the reduction of potential terrestrial carbon sinks. Approximately 40% of the carbon loss associated with LULCC in the simulations arises from direct human modification of the land surface; the remaining 60% is an indirect consequence of the loss of potential natural carbon sinks. Because of the multicentury carbon cycle legacy of current land use decisions, a globally averaged amplification factor of 2.6 must be applied to 2015 land use carbon losses to adjust for indirect effects. This estimate is 30% higher when considering the carbon cycle evolution after 2100. Most of the terrestrial uptake of anthropogenic carbon in the model occurs from the influence of rising atmospheric CO2 on photosynthesis in trees, and thus, model-projected carbon feedbacks are especially sensitive to deforestation.
Mahowald, Natalie M., and Daniel S Ward, et al., September 2017: Are the impacts of land use on warming underestimated in climate policy?Environmental Research Letters, 12(9), DOI:10.1088/1748-9326/aa836d. Abstract
While carbon dioxide emissions from energy use must be the primary target of climate change mitigation efforts, land use and land cover change (LULCC) also represent an important source of climate forcing. In this study we compute time series of global surface temperature change separately for LULCC and non-LULCC sources (primarily fossil fuel burning), and show that because of the extra warming associated with the co-emission of methane and nitrous oxide with LULCC carbon dioxide emissions, and a co-emission of cooling aerosols with non-LULCC emissions of carbon dioxide, the linear relationship between cumulative carbon dioxide emissions and temperature has a two-fold higher slope for LULCC than for non-LULCC activities. Moreover, projections used in the Intergovernmental Panel on Climate Change (IPCC) for the rate of tropical land conversion in the future are relatively low compared to contemporary observations, suggesting that the future projections of land conversion used in the IPCC may underestimate potential impacts of LULCC. By including a 'business as usual' future LULCC scenario for tropical deforestation, we find that even if all non-LULCC emissions are switched off in 2015, it is likely that 1.5 °C of warming relative to the preindustrial era will occur by 2100. Thus, policies to reduce LULCC emissions must remain a high priority if we are to achieve the low to medium temperature change targets proposed as a part of the Paris Agreement. Future studies using integrated assessment models and other climate simulations should include more realistic deforestation rates and the integration of policy that would reduce LULCC emissions.
Rabin, S, Joe R Melton, G Lasslop, D Bachelet, M Forrest, S Hantson, J O Kaplan, F Li, S Mangeon, and Daniel S Ward, et al., March 2017: The Fire Modeling Intercomparison Project (FireMIP), phase 1: experimental and analytical protocols with detailed model descriptions. Geoscientific Model Development, 10(3), DOI:10.5194/gmd-10-1175-2017. Abstract
The important role of fire in regulating vegetation community composition and contributions to emissions of greenhouse gases and aerosols make it a critical component of dynamic global vegetation models and Earth system models. Over 2 decades of development, a wide variety of model structures and mechanisms have been designed and incorporated into global fire models, which have been linked to different vegetation models. However, there has not yet been a systematic examination of how these different strategies contribute to model performance. Here we describe the structure of the first phase of the Fire Model Intercomparison Project (FireMIP), which for the first time seeks to systematically compare a number of models. By combining a standardized set of input data and model experiments with a rigorous comparison of model outputs to each other and to observations, we will improve the understanding of what drives vegetation fire, how it can best be simulated, and what new or improved observational data could allow better constraints on model behavior. In this paper, we introduce the fire models used in the first phase of FireMIP, the simulation protocols applied, and the benchmarking system used to evaluate the models. We have also created supplementary tables that describe, in thorough mathematical detail, the structure of each model.
Nevison, Cynthia D., Peter G Hess, S Riddick, and Daniel S Ward, March 2016: Denitrification, leaching, and river nitrogen export in the Community Earth System Model. Journal of Advances in Modeling Earth Systems, 8(1), DOI:10.1002/2015MS000573. Abstract
River nitrogen export is simulated within the Community Earth System Model (CESM) by coupling nitrogen leaching and runoff fluxes from the Community Land Model (CLM) to the River Transport Model (RTM). The coupled CLM-RTM prognostically simulates the downstream impact of human N cycle perturbation on coastal areas. It also provides a framework for estimating denitrification fluxes of N2 and associated trace gases like N2O in soils and river sediments. An important limitation of the current model is that it only simulates dissolved inorganic nitrogen (DIN) river export, due to the lack of dissolved organic nitrogen (DON) and particulate nitrogen (PN) leaching fluxes in CLM. In addition, the partitioning of soil N loss in CLM between the primary loss pathways of denitrification and N leaching/runoff appears heavily skewed toward denitrification compared to other literature estimates, especially in non-agricultural regions, and also varies considerably among the 4 model configurations presented here. River N export is generally well predicted in the model configurations that include mid-latitude crops, but tends to be underpredicted in rivers that are less perturbed by human agriculture. This is especially true in the tropics, where CLM likely underestimates leaching and runoff of all forms of nitrogen. River export of DIN is overpredicted in some relatively unperturbed Arctic rivers, which may result from excessive N inputs to those regions in CLM. Better representation of N loss in CLM can improve confidence in model results with respect to the core model objective of simulating nitrogen limitation of the carbon cycle. This article is protected by copyright. All rights reserved.
Riddick, S, and Daniel S Ward, et al., June 2016: Estimate of changes in agricultural terrestrial nitrogen pathways and ammonia emissions from 1850 to present in the Community Earth System Model. Biogeosciences, 13(11), DOI:10.5194/bg-13-3397-2016. Abstract
Nitrogen applied to the surface of the land for agricultural purposes represents a significant source of reactive nitrogen (Nr) that can be emitted as a gaseous Nr species, be denitrified to atmospheric nitrogen (N2), run off during rain events or form plant-useable nitrogen in the soil. To investigate the magnitude, temporal variability and spatial heterogeneity of nitrogen pathways on a global scale from sources of animal manure and synthetic fertilizer, we developed a mechanistic parameterization of these pathways within a global terrestrial land model, the Community Land Model (CLM). In this first model version the parameterization emphasizes an explicit climate-dependent approach while using highly simplified representations of agricultural practices, including manure management and fertilizer application. The climate-dependent approach explicitly simulates the relationship between meteorological variables and biogeochemical processes to calculate the volatilization of ammonia (NH3), nitrification and runoff of Nr following manure or synthetic fertilizer application. For the year 2000, approximately 125 Tg N yr−1 is applied as manure and 62 Tg N yr−1 is applied as synthetic fertilizer. We estimate the resulting global NH3 emissions are 21 Tg N yr−1 from manure (17 % of manure production) and 12 Tg N yr−1 from fertilizer (19 % of fertilizer application); reactive nitrogen runoff during rain events is calculated as 11 Tg N yr−1 from manure and 5 Tg N yr−1 from fertilizer. The remaining nitrogen from manure (93 Tg N yr−1) and synthetic fertilizer (45 Tg N yr−1) is captured by the canopy or transferred to the soil nitrogen pools. The parameterization was implemented in the CLM from 1850 to 2000 using a transient simulation which predicted that, even though absolute values of all nitrogen pathways are increasing with increased manure and synthetic fertilizer application, partitioning of nitrogen to NH3 emissions from manure is increasing on a percentage basis, from 14 % of nitrogen applied in 1850 (3 Tg NH3 yr−1) to 17 % of nitrogen applied in 2000 (21 Tg NH3 yr−1). Under current manure and synthetic fertilizer application rates we find a global sensitivity of an additional 1 Tg NH3 (approximately 3 % of manure and fertilizer) emitted per year per °C of warming. While the model confirms earlier estimates of nitrogen fluxes made in a range of studies, its key purpose is to provide a theoretical framework that can be employed within a biogeochemical model, that can explicitly respond to climate and that can evolve and improve with further observation.
Connections between wildfires and modes of variability in climate are sought as a means for predicting fire activity on interannual to multi-decadal timescales. Several fire drivers, such as temperature and local drought index, have been shown to vary on these timescales, and analysis of tree-ring data suggests covariance between fires and climate oscillation indices in some regions. However, the shortness of the satellite record of global fire events limits investigations on larger spatial scales. Here we explore the interplay between climate variability and wildfire emissions with the preindustrial long control numerical experiments and historical ensembles of CESM1 and the NOAA/GFDL ESM2Mb. We find that interannual variability in fires is underpredicted in both Earth System models (ESMs) compared to present day fire emission inventories. Modeled fire emissions respond to the El Niño/southern oscillation (ENSO) and Pacific decadal oscillation (PDO) with increases in southeast Asia and boreal North America emissions, and decreases in southern North America and Sahel emissions, during the ENSO warm phase in both ESMs, and the PDO warm phase in CESM1. Additionally, CESM1 produces decreases in boreal northern hemisphere fire emissions for the warm phase of the Atlantic Meridional Oscillation. Through analysis of the long control simulations, we show that the 20th century trends in both ESMs are statistically significant, meaning that the signal of anthropogenic activity on fire emissions over this time period is detectable above the annual to decadal timescale noise. However, the trends simulated by the two ESMs are of opposite sign (CESM1 decreasing, ESM2Mb increasing), highlighting the need for improved understanding, proxy observations, and modeling to resolve this discrepancy.
Clark, Spencer K., Daniel S Ward, and Natalie M. Mahowald, November 2015: The sensitivity of global climate to the episodicity of fire aerosol emissions. Journal of Geophysical Research: Atmospheres, 120(22), DOI:10.1002/2015JD024068. Abstract
Here we explore the sensitivity of the global radiative forcing and climate response to the episodicity of fire emissions. We compare the standard approach used in present day and future climate modeling studies, in which emissions are not episodic but smoothly interpolated between monthly mean values and that contrast to the response when fires are represented using a range of approximations of episodicity. The range includes cases with episodicity levels matching observed fire day and fire event counts, as well as cases with extreme episodicity. We compare the different emissions schemes in a set of Community Atmosphere Model (CAM5) simulations forced with reanalysis meteorology and a set of simulations with online dynamics designed to calculate aerosol indirect effect radiative forcings. We find that using climatologically observed fire frequency improves model estimates of cloud properties over the standard scheme, particularly in boreal regions, when both are compared to a simulation with meteorologically synchronized emissions. Using these emissions schemes leads to a range in global indirect effect radiative forcing of fire aerosols between −1.1 and −1.3 W m−2. In cases with extreme episodicity, we see increased transport of aerosols vertically, leading to longer lifetimes and less negative indirect effect radiative forcings. In general, the range in climate impacts that results from the different realistic fire emissions schemes is smaller than the uncertainty in climate impacts due to other aspects of modeling fire emissions.
