Overestimation Of Leaf Area Index Research Articles

Data-constrained modeling of terrestrial gross primary production over the Tibetan Plateau for 2003–2019

The gross primary productivity (GPP) of terrestrial ecosystem is the largest carbon flux between atmosphere and land surface. However, accurately simulating ecosystem GPP remains a great challenge for most land surface models (LSMs) due to the biased leaf area index (LAI) simulated by the models. In this study, we use remotely sensed LAI dataset for the period 2003–2019 to drive the Farquhar model and simulate the spatial-temporal changes of GPP for the alpine ecosystem over Tibetan Plateau. The annual GPP over the Tibetan Plateau is estimated to be 540.8 ± 27.3 Tg C yr−1, which is consistent with the benchmark datasets derived from flux tower measurements and remote sensing estimations. The GPP for the past two decades has been increasing at a rate of 2.6 Tg C yr−2. The canopy greening featured by increasing LAI contributed to 27.2 % of the increasing trend in GPP, while the direct impacts from warming and wetting climate contributed to 65.2 % of GPP changes. The contribution from the direct impact of elevated atmosphere CO2 concentration is only 6.7 %. In this study we show that constraining Farquhar model with remotely sensed LAI datasets could provide reliable simulation of GPP. Given the great challenge in modeling LAI and the considerable overestimation of LAI in current LSMs, our results highlight the importance of improving LAI modeling in LSMs and the necessity of constraining LAI when tuning the parameters for photosynthesis module.

A chlorophyll-constrained semi-empirical model for estimating leaf area index using a red-edge vegetation index

The parameter of leaf area index (LAI) provides valuable information about plant growth. Most remote sensing methods for estimating LAI are based on the vegetation indices (VIs) in the visible and near-infrared (VNIR) spectral regions. Recently, these VIs calculated with one or more bands in the red-edge region become popular in estimating LAI. Since the reflectance in the red-edge spectral region is sensitive to both LAI and leaf chlorophyll content (LCC), most red-edge-based VIs (such as normalized difference red edge index, NDRE) should be used to estimate the canopy chlorophyll content (the product of LAI and LCC) in theory. When applying NDRE to estimating LAI alone, their relationship is affected by the variation in LCC. To mitigate the LCC effect on LAI estimation using NDRE, this study proposed a chlorophyll-constrained semi-empirical model (CSE). The coefficients of the CSE model were determined from the simulated dataset based on radiative transfer models, and hence were not dependent on measured data. The performance of the CSE model on LAI estimation was evaluated using simulated and measured datasets, including a crop dataset collected in China and a multi-ecosystem dataset obtained from the National Ecological Observatory Network (NEON). The results showed that the root mean square error (RMSE) obtained by the CSE model ranged from 0.58 to 0.72 m2/m2 for the simulated and measured datasets. In particular, the CSE model also exhibited wide generality across multiple vegetation types for the NEON dataset, with the RMSE in the range of 0.33–0.86 m2/m2. In comparison, without considering the LCC effect, the model would lead to clear underestimation or overestimation of LAI (RMSE = 1.26–2.11 m2/m2). These findings improved our understanding of the LCC effects on universal LAI predictive models and would facilitate the accurate mapping of LAI over large regions for monitoring plant growth status.

Responses of runoff processes to vegetation dynamics during 1981–2010 in the Yarlung Zangbo River basin

Study regionThe Yarlung Zangbo River basin (YRB), southwest China. Study focusThe vegetation conditions play a significant role in regulating river runoff under climate change. However, the effect of vegetation growth induced by climate change to runoff variation remains controversial as relevant studies are mainly focused on the influences of climate change to runoff variation and vegetation growth, respectively. In this study, the response of runoff generation to change of Leaf Area Index (LAI) inputs during 1981–2010 in YRB were explored by using the Variable Infiltration Capacity (VIC) model. New hydrological insightsThe VIC model with dynamic LAI inputs performs better in the YRB, especially in the sub-basins with apparent peak clipping effect by LAI overestimation. The impact mechanism of how runoff variation is driven by LAI change in the meadow, tundra and cultivated vegetation cover is evapotranspiration-controlled by energy redistribution. However, the LAI changes in coniferous and broad-leaved forest cover affect the runoff generation mainly by redistributing energy-water to alter both evapotranspiration and surface soil moisture, so do the shrub and grassland during the growing season (Apr.∼Sep.). The results highlight implanting grassland would be efficient to reduce the flash flood runoff in a short term and balance groundwater runoff in a long run. The broad-leaved forest and coniferous forest with a longer growth cycle also play a main role in adjusting soil moisture.

An assessment study of three indirect methods for estimating leaf area density and leaf area index of individual trees

Field measurements of leaf area density (LAD) and leaf area index (LAI) for individual trees have increased in importance and relevance with the advent of high spatial resolution remote sensing for the urban landscape. However, indirect field measurements of LAD/LAI for widely dispersed individual trees have not been comprehensively evaluated. The present study compares the accuracy of three indirect LAD/LAI estimation methods, including single-return terrestrial laser scanning (TLS), LAI-2200, and digital hemispherical photography (DHP) on urban trees. To this end, field measurements were inter-compared and physically modeled by discrete anisotropic radiation transfer (DART). The inter-comparisons of field data revealed substantial inconsistencies between DHP and the other two approaches. For the physical modeling, reference LAI was obtained from realistic 3-D tree objects in DART, and the LAD/LAI was derived from simulated TLS, LAI-2200, and DHP acquisitions and was evaluated against the references. The physical modeling results showed that TLS could reasonably estimate LAD/LAI for LAD < 3 and LAI < 6 using a small (0.3 m) voxel size. However, the measurements became saturated for dense foliage (LAD > 3 and LAI > 6). The LAD/LAI accuracy was sharply reduced as voxel size increased. In addition, the trunks caused overestimation for both of LAD and LAI, while branches caused LAD underestimation and LAI overestimation. Further research is needed to compensate for the effects of occlusions and clumping in the estimates of LAD/LAI for dense-foliated trees using TLS. LAI-2200 grossly underestimated LAD for all cases, while it accurately estimated LAI for LAI < 5 and became gradually saturated for LAI > 5. The estimation accuracy of LAI-2200 declined markedly with increasing uncertainty in crown shape. The 90° view cap had higher accuracy than the 180° or 270° view caps using four or all five LAI-2200 rings. Additional sensors or algorithms for crown shape measurement should be developed for LAI-2200 to reduce its reliance on other data sources. DHP is not recommended for individual trees as the LAI estimations were biased from the reference values, and improvements in reliability will depend on new algorithms to account for differences in path length. These results serve as a benchmark for evaluating the accuracy of in situ LAD/LAI measurement techniques and for optimizing the configurations required for each indirect measurement method applied to individual trees.

Calibration and Validation of the SWAT Model for Predicting Daily ET over Irrigated Crops in the Texas High Plains Using Lysimetric Data

<abstract> <b><sc>Abstract.</sc></b> The Soil Water Assessment Tool (SWAT) is a widely used watershed model for simulating stream flow, overland flow, and sediment, pesticide, and bacterial loading in response to management practices. All SWAT processes are directly dependent on accurate representation of hydrology. Evapotranspiration (ET) is commonly the most significant portion of the hydrologic cycle, especially for irrigated lands in semiarid environments when ET demands are met or exceeded. However, no studies using long-term data to evaluate the SWAT model‘s capacity to estimate daily, monthly, and seasonal ET have been performed. In this study, daily and monthly ET values were simulated using ArcSWAT 2012 for an irrigated lysimeter field at the USDA-ARS Conservation and Production Research Laboratory (CPRL) at Bushland, Texas, and compared to measured ET values from 2001-2010. Crops grown during the study period included cotton, soybean, grain and silage sorghum, sunflower, and corn silage. A one-year warm-up (2000) and equal division of the remaining years were used for the calibration (2001-2005) and validation (2006-2010) periods. SWAT achieved a Nash-Sutcliffe efficiency (NSE) of 0.67 for daily ET during the calibration period, resulting in a “good” performance rating. An NSE of 0.78 resulted in a “very good” rating for the validation period. NSE values for simulated average monthly ET were improved, at 0.77 and 0.85 for the calibration and validation periods, respectively. Analysis of simulated versus measured ET values both during and outside of the growing season revealed better agreement during the former than the latter. The SWAT model generally underestimated ET at both daily and monthly levels but overestimated annual cotton ET due to overestimation of leaf area index during the senescing stage. Overall, SWAT was able to simulate daily and average monthly ET reasonably well for major summer crops grown in the semiarid Texas High Plains. These results should reinforce confidence in the SWAT model‘s capacity to accurately simulate ET in fully irrigated watersheds. However, limitations in accuracy appear to exist for certain crops, such as cotton and sunflower, and particularly under limited irrigation conditions. These deficiencies may be related to issues with the embedded Leaf Area Index (LAI) crop growth model and default plant parameter values in the crop database in SWAT.

Impact of climate, vegetation, soil and crop management variables on multi-year ISBA-A-gs simulations of evapotranspiration over a Mediterranean crop site

Abstract. Generic land surface models are generally driven by large-scale data sets to describe the climate, the soil properties, the vegetation dynamic and the cropland management (irrigation). This paper investigates the uncertainties in these drivers and their impacts on the evapotranspiration (ET) simulated from the Interactions between Soil, Biosphere, and Atmosphere (ISBA-A-gs) land surface model over a 12-year Mediterranean crop succession. We evaluate the forcing data sets used in the standard implementation of ISBA over France where the model is driven by the SAFRAN (Système d'Analyse Fournissant des Renseignements Adaptés à la Nivologie) high spatial resolution atmospheric reanalysis, the leaf area index (LAI) time courses derived from the ECOCLIMAP-II land surface parameter database and the soil texture derived from the French soil database. For climate, we focus on the radiations and rainfall variables and we test additional data sets which include the ERA-Interim (ERA-I) low spatial resolution reanalysis, the Global Precipitation Climatology Centre data set (GPCC) and the MeteoSat Second Generation (MSG) satellite estimate of downwelling shortwave radiations. The evaluation of the drivers indicates very low bias in daily downwelling shortwave radiation for ERA-I (2.5 W m−2) compared to the negative biases found for SAFRAN (−10 W m−2) and the MSG satellite (−12 W m−2). Both SAFRAN and ERA-I underestimate downwelling longwave radiations by −12 and −16 W m−2, respectively. The SAFRAN and ERA-I/GPCC rainfall are slightly biased at daily and longer timescales (1 and 0.5 % of the mean rainfall measurement). The SAFRAN rainfall is more precise than the ERA-I/GPCC estimate which shows larger inter-annual variability in yearly rainfall error (up to 100 mm). The ECOCLIMAP-II LAI climatology does not properly resolve Mediterranean crop phenology and underestimates the bare soil period which leads to an overall overestimation of LAI over the crop succession. The simulation of irrigation by the model provides an accurate irrigation amount over the crop cycle but the timing of irrigation occurrences is frequently unrealistic. Errors in the soil hydrodynamic parameters and the lack of irrigation in the simulation have the largest influence on ET compared to uncertainties in the large-scale climate reanalysis and the LAI climatology. Among climate variables, the errors in yearly ET are mainly related to the errors in yearly rainfall. The underestimation of the available water capacity and the soil hydraulic diffusivity induce a large underestimation of ET over 12 years. The underestimation of radiations by the reanalyses and the absence of irrigation in the simulation lead to the underestimation of ET while the overall overestimation of LAI by the ECOCLIMAP-II climatology induces an overestimation of ET over 12 years. This work shows that the key challenges to monitor the water balance of cropland at regional scale concern the representation of the spatial distribution of the soil hydrodynamic parameters, the variability of the irrigation practices, the seasonal and inter-annual dynamics of vegetation and the spatiotemporal heterogeneity of rainfall.

Comparison of Five Slope Correction Methods for Leaf Area Index Estimation From Hemispherical Photography

We compare five slope correction methods developed by Walter et al., Montes et al., Schleppi et al., España et al., and Gonsamo et al. (referred to as WAL, MON, SCH, ESP, and GON, respectively) using artificial fisheye pictures simulated by graphics software and a lookup table (LUT) retrieval method. The LUT is built by simulating the directional gap fraction as a function of leaf area index (LAI) and average leaf inclination angle (ALIA) using the Poisson law. LAI and ALIA estimates correspond to the case of the LUT that provides the lowest root-mean-square error between the observed gap fractions after slope correction and the simulated ones. Three LAI values (1.5, 3.5, and 5.5), four ALIA values (26.8°, 45°, 57.5°, and 63.2°), and three slope angles (0°, 20°, and 50°) constituted 36 samples of random scenes. ESP is recommended because its results are accurate and independent on the leaf angle distribution (LAD), while GON only performs well for spherical LAD. The three other methods present less good performances with underestimation or overestimation of LAI and/or ALIA depending on the LAD, and the recommended order for them is MON, SCH, and WAL.

Validation of Oryza2000 model under combined nitrogen and water limited situations

Oryza2000 is an eco-physiological crop model to simulate growth and development of rice in situations of potential production, water limitations, and nitrogen limitations. In the present investigation, unlike independent situation, the applicability under conditions of both water and N limitations has been investigated using data generated from field experiments over 5 years on long and short duration rice varieties. At higher N level model performs better for TDM and yield performance long duration varieties compared to short (Rasi) and medium duration (Ajaya) varieties. On the other, overestimation of leaf area index (LAI) at all nitrogen levels was consistent indicating a need in this area for further improvement of the model. In this context, model performance was also assessed by statistical analysis (R2, D-index and NOF). D-index is ~0.8–0.9 for two varieties at 3 nitrogen application levels for Ajaya and BPT varieties. The results indicated applicability of model Oryza2000 can be extended to rice genotypes varying in their growth periods and situations like fertilizer, water limited conditions.

Effects of crop residue cover resulting from tillage practices on LAI estimation of wheat canopies using remote sensing

Much research has been conducted seeking to reduce the noise interference from soil in remote sensing of vegetation, and thus more accurately predict plant biophysical parameters by developing different kinds of vegetation indices. However, few studies have examined the effect of residue cover on the estimation of leaf area index (LAI) using remote sensing techniques. Using the WinSail model along with ground measurements, this study sought to quantify the differences in spectral reflectance and commonly used ratio-based (NDVI and RVI), soil-adjusted (TSAVI and SAVI2) and hyperspectral (dRE and REIP) vegetation indices between wheat canopies with bare soil and rice residue as backgrounds and to identify the vegetation indices that represented the best combination of low sensitivity to residue cover and high sensitivity to variation in LAI. We found large variations in residue cover in wheat fields, with average cover of 6.70% and 45.9% for intensively tilled and untilled fields, respectively. Changing the background from soil to residue resulted in substantial changes in both reflectance and vegetation indices of canopies when LAI varied between 0.1 and 1.0, with average absolute values of relative percent differences (|RPD|) ranging from 4.21% to 19.03% for the six vegetation indices used in this study. We found that changes in the background from bare soil to rice residue would result in underestimation of LAI by NDVI, RVI, TSAVI, and SAVI2, and overestimation of LAI by dRE and REIP. More green leaves in the wheat canopy were covered by residue in untilled fields than in intensively tilled fields, with an average 8.57% obscuring rate when green cover was between 1.0% and 85%; this would lead to the underestimation of LAI by remote sensing techniques in untilled fields. Ultimately, we determined that dRE was the best choice of the six indices for predicting LAI because of its significant relationship with LAI and because it was least sensitive to residue effects yet remained sensitive to variation in LAI, underestimating LAI in untilled fields by only 3.70%. The better performance of dRE was attributed to the opposite additive influences of the residue obscuring effect and brightness differences between soil and residue.

Analyzing the Ecosystem Carbon Dynamics of Four European Coniferous Forests Using a Biogeochemistry Model

This paper provides the first steps toward a regional-scale analysis of carbon (C) budgets. We explore the ability of the ecosystem model BIOME-BGC to estimate the daily and annual C dynamics of four European coniferous forests and shifts in these dynamics in response to changing environmental conditions. We estimate uncertainties in the model results that arise from incomplete knowledge of site management history (for example, successional stage of forest). These uncertainties are especially relevant in regional-scale simulations, because this type of information is difficult to obtain. Although the model predicted daily C and water fluxes reasonably well at all sites, it seemed to have a better predictive capacity for the photosynthesis-related processes than for respiration. Leaf area index (LAI) was modeled accurately at two sites but overestimated at two others (as a result of poor long-term climate drivers and uncertainties in model parameterization). The overestimation of LAI (and consequently gross photosynthetic production (GPP)), in combination with reasonable estimates of the daily net ecosystem productivity (NEP) of those forests, also illustrates the problem with modeled respiration. The model results suggest that all four European forests have been net sinks of C at the rate of 100 ‐300 gC/m 2 /y and that this C sequestration capacity would be 30%‐70% lower without increasing nitrogen (N) deposition and carbon dioxide (CO 2 ) concentrations. The magnitude of the forest responses was dependent not only on the rate of changes in environmental factors, but also on sitespecific conditions such as climate and soil depth. We estimated that the modeled C exchange at the study sites was reduced by 50%‐100% when model simulations were performed for climax forests rather than regrowing forests. The estimates of water fluxes were less sensitive to different initializations of state variables or environmental change scenarios than C fluxes.

Determining Cotton Water Use in a Semiarid Climate with the GOSSYM Cotton Simulation Model

AbstractSimulation models are useful tools that provide information about potential changes to production systems before committing time and resources. GOSSYM, a physiologically based cotton (Gossypium hirsutum L.) simulation model, was developed as an experimental tool but has been modified and adapted as a crop management tool. For GOSSYM to be used successfully in crop management, especially as an irrigation scheduling predictor, its ability to accurately calculate and partition total water use between transpiration (T) and soil water evaporation (E) must be evaluated. A field study with cotton was conducted at Lubbock, TX, in 1994 in an Olton soil (fine, mixed, thermic Aridic Paleustoll) to compare GOSSYM's calculations of E, T, and evapotranspiration (ET = E + T) with values measured by neutron attenuation, microlysimetry, and stem flow gauges. During a 12‐d period when E and T were measured separately, GOSSYM underestimated cumulative E by 18%, while cumulative T was in close agreement with measured values. Underestimation of E was due to overestimation of leaf area index (LAI), thus reducing simulated incident irradiance at the soil surface. A comparison of calculated and measured values of daily T over 60 d showed that calculated daily T's were 40% greater than measured values during the initial 10 d, 50% less during the last 10 d, and in close agreement in between. Differences between calculated and measured daily T‐values were attributed to GOSSYM underestimating potential ET and overestimating LAI. However, ET over 102 d was calculated within 10% at the end of the measuring period. These data suggest that GOSSYM can be used to assess water use in cotton, and as a tool for scheduling irrigations in a semiarid region, provided that current algorithms used to calculate potential ET are modified to include air humidity.

Application of SORKAM to Simulate Shattercane Growth Using Forage Sorghum

AbstractShattercane [Sorghum bicolor (L.) Moench] is a serious weed problem. Few methods are available to assess the impact of shattercane competition on crop growth and yield. This study evaluated model sensitivity to variations in light, water, and plant density and assessed the applicability of the SORKAM model for simulating forage sorghum (used to represent shattercane) growth and development. Sensitivity analysis of SORKAM to photosynthetically active radiation (50, 75, and 100% of actual); water (50, 75, and 100% of actual precipitation, and no water stress); and plant densities of 3, 6, 12, 24, 48, and 96 plants m−2 was performed. ‘Rox Orange’ forage sorghum (used to represent shattercane) was grown at three densities, under irrigated and dryland environments over a 3‐yr period. Experiments were conducted on Eudora (coarse‐silty, mixed, mesic Fluventic Hapludoll) and Kahola silt loam (fine‐silty, mixed, mesic Cumulic Hapludoll) soils, in Manhattan, KS. The sensitivity analysis showed that SORKAM responded to a broad range of light, water, and plant density environments, with significant light by water interaction for biomass, and light by density interaction for biomass, leaf area index (LAI), and tillering. The SORKAM model simulated growth of Rox Orange from inputs of weather, soil, and plant data were compared with measured values. There was good agreement between simulated and measured LAI through 60 d after planting (r2 = 0.82, intercept not significantly different from 0, and slope = 0.83). Accurate estimation of LAI during this period of vegetative growth is important, because LAI determines the competition between a crop and shattercane for light and water. Agreement of measured and simulated biomass was good for low plant densities (slope and intercept not significantly different from 1 and 0 with high r2 = 0.78), with progressively greater underestimation at higher densities. Tiller number was generally underestimated, with no apparent correlation to model under‐ or overestimation of LAI or biomass. The sensitivity analysis and comparison of measured and simulated LAI indicates that the SORKAM model can be used to assess Rox Orange (shattercane) competition.