Devendra Dahal

Development of a generic auto-calibration package for regional ecological modeling and application in the Central Plains of the United States

article i nfo Process-orientedecologicalmodels are frequentlyusedfor predicting potentialimpactsof global changes suchas climate and land-cover changes, which can be useful for policy making. It is critical but challenging to automat- ically derive optimal parametervaluesat different scales,especially at regional scale, and validate the model per- formance. In this study, we developed an automatic calibration (auto-calibration) function for a well-established biogeochemical model—the General Ensemble Biogeochemical Modeling System (GEMS)-Erosion Deposition Carbon Model (EDCM)—using data assimilation technique: the Shuffled Complex Evolution algorithm and a model-inversion R package—Flexible Modeling Environment (FME). The new functionality can support multi- parameter and multi-objectiveauto-calibration of EDCM atthe both pixel and regional levels. Wealsodeveloped a post-processing procedure for GEMS to provide options to save the pixel-based or aggregated county-land cover specific parameter values for subsequent simulations. In our case study, we successfully applied the up- dated model (EDCM-Auto) for a single crop pixel with a corn-wheat rotation and a large ecological region (Level II)—Central USA Plains. The evaluation results indicate that EDCM-Auto is applicable at multiple scales and is capable to handle land cover changes (e.g., crop rotations). The model also performs well in capturing the spatial pattern of grain yield production for crops and net primary production (NPP) for other ecosystems across the region, which is a good example for implementing calibration and validation of ecological models with readily available survey data (grain yield) and remote sensing data (NPP) at regional and national levels. The developed platform for auto-calibration can be readily expanded to incorporate other model inversion algorithms and potential R packages, and also be applied to other ecological models. Published by Elsevier B.V.

Simulating the water budget of a Prairie Potholes complex from LiDAR and hydrological models in North Dakota, USA

Abstract Hydrological processes of the wetland complex in the Prairie Pothole Region (PPR) are difficult to model, partly due to a lack of wetland morphology data. We used Light Detection And Ranging (LiDAR) data sets to derive wetland features; we then modelled rainfall, snowfall, snowmelt, runoff, evaporation, the “fill-and-spill” mechanism, shallow groundwater loss, and the effect of wet and dry conditions. For large wetlands with a volume greater than thousands of cubic metres (e.g. about 3000 m3), the modelled water volume agreed fairly well with observations; however, it did not succeed for small wetlands (e.g. volume less than 450 m3). Despite the failure for small wetlands, the modelled water area of the wetland complex coincided well with interpretation of aerial photographs, showing a linear regression with R2 of around 0.80 and a mean average error of around 0.55 km2. The next step is to improve the water budget modelling for small wetlands. Editor Z.W. Kundzewicz; Associate editor X. Chen Citation Huang, S.L., Young, C., Abdul-Aziz, O.I., Dahal, D., Feng, M., and Liu, S.G., 2013. Simulating the water budget of a Prairie Potholes complex from LiDAR and hydrological models in North Dakota, USA. Hydrological Sciences Journal, 58 (7), 1434–1444.

Projection of corn production and stover-harvesting impacts on soil organic carbon dynamics in the U.S. Temperate Prairies

Terrestrial carbon sequestration potential is widely considered as a realistic option for mitigating greenhouse gas emissions. However, this potential may be threatened by global changes including climate, land use, and management changes such as increased corn stover harvesting for rising production of cellulosic biofuel. Therefore, it is critical to investigate the dynamics of soil organic carbon (SOC) at regional or global scale. This study simulated the corn production and spatiotemporal changes of SOC in the U.S. Temperate Prairies, which covers over one-third of the U.S. corn acreage, using a biogeochemical model with multiple climate and land-use change projections. The corn production (either grain yield or stover biomass) could reach 88.7–104.7 TgC as of 2050, 70–101% increase when compared to the base year of 2010. A removal of 50% stover at the regional scale could be a reasonable cap in view of maintaining SOC content and soil fertility especially in the beginning years. The projected SOC dynamics indicated that the average carbon sequestration potential across the entire region may vary from 12.7 to 19.6 g C/m2/yr (i.e., 6.6–10.2 g TgC/yr). This study not only helps understand SOC dynamics but also provides decision support for sustainable biofuel development.

The carbon cycle and hurricanes in the United States between 1900 and 2011

Hurricanes cause severe impacts on forest ecosystems in the United States. These events can substantially alter the carbon biogeochemical cycle at local to regional scales. We selected all tropical storms and more severe events that made U.S. landfall between 1900 and 2011 and used hurricane best track database, a meteorological model (HURRECON), National Land Cover Database (NLCD), U. S. Department of Agirculture Forest Service biomass dataset, and pre- and post-MODIS data to quantify individual event and annual biomass mortality. Our estimates show an average of 18.2 TgC/yr of live biomass mortality for 1900–2011 in the US with strong spatial and inter-annual variability. Results show Hurricane Camille in 1969 caused the highest aboveground biomass mortality with 59.5 TgC. Similarly 1954 had the highest annual mortality with 68.4 TgC attributed to landfalling hurricanes. The results presented are deemed useful to further investigate historical events, and the methods outlined are potentially beneficial to quantify biomass loss in future events.

Fusing MODIS with Landsat 8 data to downscale weekly normalized difference vegetation index estimates for central Great Basin rangelands, USA

Data fused from distinct but complementary satellite sensors mitigate tradeoffs that researchers make when selecting between spatial and temporal resolutions of remotely sensed data. We integrated data from the Moderate Resolution Imaging Spectroradiometer (MODIS) sensor aboard the Terra satellite and the Operational Land Imager sensor aboard the Landsat 8 satellite into four regression-tree models and applied those data to a mapping application. This application produced downscaled maps that utilize the 30-m spatial resolution of Landsat in conjunction with daily acquisitions of MODIS normalized difference vegetation index (NDVI) that are composited and temporally smoothed. We produced four weekly, atmospherically corrected, and nearly cloud-free, downscaled 30-m synthetic MODIS NDVI predictions (maps) built from these models. Model results were strong with R2 values ranging from 0.74 to 0.85. The correlation coefficients (r ≥ 0.89) were strong for all predictions when compared to corresponding original MODIS NDVI data. Downscaled products incorporated into independently developed sagebrush ecosystem models yielded mixed results. The visual quality of the downscaled 30-m synthetic MODIS NDVI predictions were remarkable when compared to the original 250-m MODIS NDVI. These 30-m maps improve knowledge of dynamic rangeland seasonal processes in the central Great Basin, United States, and provide land managers improved resource maps.

Grassland and Cropland Net Ecosystem Production of the U.S. Great Plains: Regression Tree Model Development and Comparative Analysis

This paper presents the methodology and results of two ecological-based net ecosystem production (NEP) regression tree models capable of up scaling measurements made at various flux tower sites throughout the U.S. Great Plains. Separate grassland and cropland NEP regression tree models were trained using various remote sensing data and other biogeophysical data, along with 15 flux towers contributing to the grassland model and 15 flux towers for the cropland model. The models yielded weekly mean daily grassland and cropland NEP maps of the U.S. Great Plains at 250 m resolution for 2000–2008. The grassland and cropland NEP maps were spatially summarized and statistically compared. The results of this study indicate that grassland and cropland ecosystems generally performed as weak net carbon (C) sinks, absorbing more C from the atmosphere than they released from 2000 to 2008. Grasslands demonstrated higher carbon sink potential (139 g C·m−2·year−1) than non-irrigated croplands. A closer look into the weekly time series reveals the C fluctuation through time and space for each land cover type.

Evaluation of the Initial Thematic Output from a Continuous Change-Detection Algorithm for Use in Automated Operational Land-Change Mapping by the U.S. Geological Survey

The U.S. Geological Survey (USGS) has begun the development of operational, 30-m resolution annual thematic land cover data to meet the needs of a variety of land cover data users. The Continuous Change Detection and Classification (CCDC) algorithm is being evaluated as the likely methodology following early trials. Data for training and testing of CCDC thematic maps have been provided by the USGS Land Cover Trends (LC Trends) project, which offers sample-based, manually classified thematic land cover data at 2755 probabilistically located sample blocks across the conterminous United States. These samples represent a high quality, well distributed source of data to train the Random Forest classifier invoked by CCDC. We evaluated the suitability of LC Trends data to train the classifier by assessing the agreement of annual land cover maps output from CCDC with output from the LC Trends project within 14 Landsat path/row locations across the conterminous United States. We used a small subset of circa 2000 data from the LC Trends project to train the classifier, reserving the remaining Trends data from 2000, and incorporating LC Trends data from 1992, to evaluate measures of agreement across time, space, and thematic classes, and to characterize disagreement. Overall agreement ranged from 75% to 98% across the path/rows, and results were largely consistent across time. Land cover types that were well represented in the training data tended to have higher rates of agreement between LC Trends and CCDC outputs. Characteristics of disagreement are being used to improve the use of LC Trends data as a continued source of training information for operational production of annual land cover maps.

Spatially explicit surface energy budget and partitioning with remote sensing and flux measurements in a boreal region of Interior Alaska

Extrapolating energy fluxes between the ground surface and the atmospheric boundary layer from point-based measurements to spatially explicit landscape estimation is critical to understand and quantify the energy balance components and exchanges in the hydrosphere, atmosphere, and biosphere. This information is difficult to quantify and are often lacking. Using a Landsat image (acquired on 5 August 2004), the flux measurements from three eddy covariance flux towers (a 1987 burn, a 1999 burn, and an unburned control site) and a customized satellite-based surface energy balance model of Mapping Evapotranspiration at High Resolution with Internalized Calibration (METRIC), we estimated net radiation, sensible heat flux (H), latent heat flux (LE), and soil heat flux (G) for the boreal Yukon River Basin of Interior Alaska. The model requires user selection of two extreme conditions present within the image area to calibrate and anchor the sensible flux output. One is the “hot” condition which refers to a bare soil condition with specified residual evaporation rates. Another one is the “cold” condition which refers to a fully transpiring vegetation such as full-cover agricultural crops. We selected one bare field as the “hot” condition while we explored three different scenarios for the “cold” pixel because of the absence of larger expanses of agricultural fields within the image area. For this application over boreal forest, selecting agricultural fields whose evapotranspiration was assumed to be 1.05 times the alfalfa-based reference evapotranspiration as the “cold” pixel could result in large errors. Selecting an unburned flux tower site as the “cold” pixel could achieve acceptable results, but uncertainties remain about the energy balance closure of the flux towers. We found that METRIC performs reasonably well in partitioning energy fluxes in a boreal landscape.

Rapid Crop Cover Mapping for the Conterminous United States

Timely crop cover maps with sufficient resolution are important components to various environmental planning and research applications. Through the modification and use of a previously developed crop classification model (CCM), which was originally developed to generate historical annual crop cover maps, we hypothesized that such crop cover maps could be generated rapidly during the growing season. Through a process of incrementally removing weekly and monthly independent variables from the CCM and implementing a ‘two model mapping’ approach, we found it viable to generate conterminous United States-wide rapid crop cover maps at a resolution of 250 m for the current year by the month of September. In this approach, we divided the CCM model into one ‘crop type model’ to handle the classification of nine specific crops and a second, binary model to classify the presence or absence of ‘other’ crops. Under the two model mapping approach, the training errors were 0.8% and 1.5% for the crop type and binary model, respectively, while test errors were 5.5% and 6.4%, respectively. With spatial mapping accuracies for annual maps reaching upwards of 70%, this approach demonstrated a strong potential for generating rapid crop cover maps by the 1st of September.

Estimating carbon and showing impacts of drought using satellite data in regression-tree models

ABSTRACT Integrating spatially explicit biogeophysical and remotely sensed data into regression-tree models enables the spatial extrapolation of training data over large geographic spaces, allowing a better understanding of broad-scale ecosystem processes. The current study presents annual gross primary production (GPP) and annual ecosystem respiration (RE) for 2000–2013 in several short-statured vegetation types using carbon flux data from towers that are located strategically across the conterminous United States (CONUS). We calculate carbon fluxes (annual net ecosystem production [NEP]) for each year in our study period, which includes 2012 when drought and higher-than-normal temperatures influence vegetation productivity in large parts of the study area. We present and analyse carbon flux dynamics in the CONUS to better understand how drought affects GPP, RE, and NEP. Model accuracy metrics show strong correlation coefficients (r) (r ≥ 94%) between training and estimated data for both GPP and RE. Overall, average annual GPP, RE, and NEP are relatively constant throughout the study period except during 2012 when almost 60% less carbon is sequestered than normal. These results allow us to conclude that this modelling method effectively estimates carbon dynamics through time and allows the exploration of impacts of meteorological anomalies and vegetation types on carbon dynamics.