Performance evaluation of hybrid Wavelet-ANN and Wavelet-ANFIS models for estimating evapotranspiration in arid regions of India

This paper evaluates the ability of wavelet transform in improving the accuracy of artificial neural network (ANN) and adaptive neuro-fuzzy interface systems (ANFIS) models. In this study, the performance of hybrid Wavelet-ANN and Wavelet-ANFIS models for estimating daily evapotranspiration in arid regions was evaluated. Prior to the development of models, gamma test was used to identify the best input combinations that could be used under limited data scenario. Performance of the proposed hybrid models was compared to ANN, ANFIS, and conventionally used Hargreaves equation. The results revealed that use of wavelet transform as data preprocessing technique enhanced the efficiency of ANN and ANFIS models. Wavelet-ANN and Wavelet-ANFIS performed reasonably better than other models. Better handling of wavelet-decomposed input variables enabled Wavelet-ANN models to perform slightly better than the Wavelet-ANFIS models. W-ANN2 (RMSE = 0.632 mm/day and R = 0.96) was found to be the best model for estimating daily evapotranspiration in arid regions. The proposed W-ANN2 model used second-level db3 wavelet-decomposed subseries of temperature and previous day evapotranspiration values as inputs. The study concludes that hybrid Wavelet-ANN and Wavelet-ANFIS models can be effectively used for modeling evapotranspiration.

     

ArcE: A GIS tool for modelling actual evapotranspiration

This paper introduces ArcE, a GIS tool for modelling actual evapotranspiration (E_A) from an undefined number of meteorological stations. From daily data of precipitation and temperature, ArcE uses ArcObjects as the programming language to incorporate equations and hydrological boundary conditions, in order to calculate E_A at monthly and yearly time steps. Because weather data are often missing, ArcE is programmed to use non-global models such as Hargreaves for potential evapotranspiration (E_P) and Budyko for E_A. In arid regions, where results from global and non-global models are expected to deviate, ArcE allows for the segregation of low-divergent areas suitable for interpolating E_A from those that should be excluded for mapping the variable. In the semiarid Almanzora River basin, a heterogeneous region with contrasting climate in SE Spain, divergence in lowlands with a higher aridity index was about 15% with respect to an accurate estimate of E_A from the Penman-Monteith equation. Evaluating E_A is a first step for mapping the non-evaporative fraction of precipitation as the difference in P and E_A.     

Simplifying the revised three-temperature model for remotely estimating regional evapotranspiration and its application to a semi-arid steppe

Remote sensing is a feasible and economical way of mapping the spatial distribution of evapotranspiration (ET) at scales ranging from the regional to the global level. The revised three-temperature (3T-R) model, including land-surface temperature, reference temperature, and air temperature, is a promising approach for ET mapping, but aerodynamic resistance must be included in the model. The objectives of this study are to: (1) propose a simpler remote-sensing algorithm for estimating reference temperatures in the 3T-R model that would not depend on aerodynamic resistance; and (2) test the performance of the simplified 3T-R (3T-S) model. Assuming a pixel with the maximum surface temperature is a site without evaporation or transpiration, a method was proposed to replace the reference temperatures with paired maximum temperatures for soil and canopy, collected from a region with approximately equivalent solar radiation and terrain. A case study was conducted using Landsat Thematic Mapper (TM) images and synchronous ground observations performed in an Ecosystem Observation Station in northern China during the 2009 growing season. The results indicated that: (1) daily ET (ET_{3T_S}) estimated using the 3T-S model was close to the observational data, with a mean absolute error (MAE) of 0.24 mm d^?1 and a mean absolute percentage error (MAPE) of 9.78%; (2) the 3T-S model was much simpler than the 3T-R model with respect to its calculation procedures and data requirements, suggesting it might result in less error propagation, because the MAE and MAPE between ET_{3T_R} (daily ET estimated using the 3T-R model) and the observational data were 0.36 mm d^?1 and 14.71%, respectively. Therefore, it is concluded that the 3T-S model is a simpler method, but also an accurate way of estimating regional ET.     

An R package for modelling actual,potential and reference evapotranspiration

Evapotranspiration (ET) is a vital component of the hydrological cycle and there are a large number of alternative models for representing ET processes. However, implementing ET models in a consistent manner is difficult due to the significant diversity in process representations, assumptions, nomenclature, terminology, units and data requirements. An R package is therefore introduced to estimate actual, potential and reference ET using 17 well-known models. Data input is flexible, and customized data checking and pre-processing methods are provided. Results are presented as summary text and plots. Comparisons of alternative ET estimates can be visualized for multiple models, and alternative input data sets. The ET estimates also can be exported for further analysis, and used as input to rainfall-runoff models.     

Assessing a scheme of spatial-temporal thermal remote-sensing sharpening for estimating regional evapotranspiration

A high temporal frequency of high-resolution thermal data is required in regional evapotranspiration (ET) studies. In this article, a spatial-temporal thermal remote-sensing sharpening scheme, which can be used to perform temporally stable land surface temperature (LST) mapping with high spatial resolution and further facilitate the estimation of ET, is discussed in the context of the Soil Moisture Experiment of 2002. To demonstrate this scheme, relationships between LST and three remote-sensing parameters (normalized difference vegetation index (NDVI), fractional vegetation cover (FVC), and Bowen ratio) were first used in a thermal disaggregation procedure for retrieving LSTs at a 250-m scale. Then, the spatial and temporal adaptive reflectance fusion (STARFM) model was applied to the 250-m LSTs, producing LST data at a fine resolution of 60 m and a fine temporal resolution of 1 day. Two remote-sensing-based energy balance models were then used to retrieve the ET at the Moderate Resolution Imaging Spectroradiometer overpass time respectively using 250- and 60-m LSTs. The results showed that the Bowen ratio-based LSTs were matched field observations better than did the LSTs obtained with the other two approaches (NDVI- and FVC-based) at the 250-m scale, and consequently produced 250-m ET mapping that better matched the observed tower-based values. When combined with the STARFM fusion model, the 250-m Bowen ratio-based LSTs produced more accurate time-series LSTs and ET at the 60-m scale. The Bowen ratio, which is more related to surface energy principles and the soil moisture variation, was effective in disaggregating LSTs and promoting the estimation of ET. Overall, sharpened LSTs using the combination of thermal disaggregation procedure and the STARFM fusion model could substantially improve remote-sensing-based ET estimates. Moreover, the STARFM model that can fuse LST from 250 to ~100 m should be given more attention as long as the thermal disaggregation procedure that can disaggregate LST from 1000 to 250 m, provided that it contributed approximately 10.1% to further improving ET retrieval performance.     

Multilayer perceptron for reference evapotranspiration estimation in a semiarid region

Calculation of reference evapotranspiration (ET_o) is essential in hydrology and agriculture. ET_o plays an important role in planning and management of water resources and irrigation scheduling. The results of many studies strongly support the use of the Penman–Monteith FAO 56 (PMF-56) method as the standard method of estimating ET_o. The basic obstacle to using this method widely is the numerous meteorological variables required. Multilayer perceptron (MLP) networks optimized with different learning algorithms and activation functions were applied for estimating ET_o in a semiarid region in Iran. Four MLP models comprising various combinations of meteorological variables are developed. The MLP model which needs all of the meteorological parameters performed best for ET_o estimation amongst the other MLP models. It was also found that the ConjugateGradient, DeltaBarDelta, DeltaBarDelta and Levenberg–Marquardt were the best algorithms for training the MLP1, MLP2, MLP3 and MLP4 models, respectively.     

Comparison of Meteosat-based rainfall and evapotranspiration mapping in the Sahel region

Abstract

The FAO satellite remote sensing system ARTEMIS has recently been installed. It will support agricultural production and desert locust monitoring in Africa. In relation to these tasks, this study discusses and compares a rainfall and an evapotranspiration mapping technique for the Sahel region. Evapotranspiration maps reveal detailed soil moisture related patterns, resulting from rainfall. The mapping results indicate a potential of this technique for locust habitat monitoring and crop early warning.     

Interannual covariability between actual evapotranspiration and PAL and GIMMS NDVIs of northern Asia

This study examined the covariability between interannual changes in the normalized difference vegetation index (NDVI) and actual evapotranspiration (ET). To reduce possible uncertainty in the NDVI time series, two NDVI datasets derived from Pathfinder AVHRR Land (PAL) data and the Global Inventory Monitoring and Modeling Studies (GIMMS) group were used. Analyses were conducted using data over northern Asia from 1982 to 2000. Interannual changes over 19 years in the PAL-NDVI and GIMMS-NDVI were compared with interannual changes in ET estimated from model-assimilated atmospheric data and gridded precipitation data. For both NDVI datasets, the annual maximum correlation with ET occurred in June, which is the beginning of the vegetation growing season. The PAL and GIMMS datasets showed a significant, positive correlation between interannual changes in the NDVI and ET over most of the vegetated land area in June. These results suggest that interannual changes in vegetation activity predominantly control interannual changes in ET in June. Based on analyses of interannual changes in temperature, precipitation, and the NDVI in June, the study area can be roughly divided into two regions, the warmth-dominated northernmost region and the wetness-dominated southern region, indicating that interannual changes in vegetation and the resultant interannual changes in ET are controlled by warmth and wetness in these two regions, respectively.     

Usefulness and limits of MODIS GPP for estimating wheat yield

Gross primary productivity (GPP) estimates derived from the Moderate Resolution Imaging Spectroradiometer (MODIS) are converted to wheat yield and compared with observed yield for counties, climate districts and entire states for the 2001 and 2002 growing seasons in Montana and North Dakota. Analyses revealed that progressive levels of spatial aggregation generally improved the relations between estimated and observed wheat yield. However, only state level yield estimates were sufficiently accurate (? 5% deviation from observed yield). The statewide yield results were encouraging because they were derived without the use of retrospective empirical analyses, which constitutes a new opportunity for timely wheat yield estimates for large regions. Additionally, this study identifies six practical limits to estimating wheat yield using MODIS GPP. As a result we describe three suggestions for improving wheat yield estimates for scientists willing to re‐compute MODIS‐derived GPP estimates using regionally specific inputs.     

Asymptotic stability and estimating the region of attraction for the swing equations

We state and prove a theorem that asserts the asymptotic stability and provides an estimate of the region of attraction of an equilibrium point of the swing equations. A version of this theorem was originally introduced by Willems; however, his justifications are sketchy and, in our opinion, require additional analysis. All the Liapunov method analyses utilized in the transient stability problem of power systems have been based on the assumed validity of this theorem. The shortcomings of the other proposed proofs in the literature is that they rely on available Liapunov theorems for which an assumption can not be verified. In contrast, the techniques used in the proof here avoid the restriction of the available Liapunov theorems; moreover, they can be extended to apply to a broader class of systems.     

Technical note Analysis of a simplified relation for estimating daily evapotranspiration from satellite thermal IR data

Abstract

Estimation of evapotranspiration from thermal infrared (IR)as been widely studied in recent years and in particular by methods using the simplified relation ETR ? R _n = A ? B(T _s?T _a) proposed by Jackson et al. (1977) and thermal IR data from the NOAA satellite obtained near midday (local time). The representativeness of instantaneous temperature for computing daily fluxes is discussed and a one-layer model is presented for a developed crop. A new theoretical expression of the simplified relation which can be applied to developed crops is presented and the sensitivity of the parameters in this relation is examined.     

Application of recognition procedures for estimating the efficiency region for the LP τ search with averaging

Application of optimization and imitation methods in studying dynamical stochastic systems allows one to solve multicriterion multiparameter optimization problems of a special kind. The quality factors are represented as multidimensional integrals over the domain of alteration of the parameters, which is determined in the process of investigation and approximate evaluation of the integrals. Algorithms for determining the efficiency region, in which the quality factors of the dynamical stochastic system attain their refined values, are considered. Galina Mikhailovna Antonova. Born in 1951. Graduated from Faculty of Radioelectronics of Leningrad Polytechnic Institute in 1974. Received candidate’s degree in 1980 and doctoral degree in 2002. Works at the Institute of Control Problems of RAS and at the Moscow State Institute of Radioengineering, Electronics and Automatics (Technical University). Scientific interests: image recognition, imitational statistic modeling, optimization of stochastic systems.     

A neural network model for estimating option prices

A neural network model that processes financial input data is developed to estimate the market price of options at closing. The network's ability to estimate closing prices is compared to the Black-Scholes model, the most widely used model for the pricing of options. Comparisons reveal that the mean squared error for the neural network is less than that of the Black-Scholes model in about half of the cases examined. The differences and similarities in the two modeling approaches are discussed. The neural network, which uses the same financial data as the Black-Scholes model, requires no distribution assumptions and learns the relationships between the financial input data and the option price from the historical data. The option-valuation equilibrium model of Black-Scholes determines option prices under the assumptions that prices follow a continuous time path and that the instantaneous volatility is nonstochastic.     

Weibull分布在软件最优交付时间估算中的应用

软件最优交付时间在很大程度上取决于软件可靠性估算。Weibull分布较有效地描述了软件测试过程中错误发现的统计特征,是一种有效的软件可靠性分析模型。在分析Weibull分布的基础上,提出了Weibull分布的累计函数拟合模型,解决了错误发现的不连续问题,推导出两种利用Weibull累计函数模型进行软件最优交付时间的估算公式,并结合具体实例验证了模型的有效性。     

A causal model for software cost estimating error

Software cost estimation is an important concern for software managers and other software professionals. The hypothesized model in this research suggests that an organization's use of an estimate influences its estimating practices which influence both the basis of the estimating process and the accuracy of the estimate. The model also suggests that the estimating basis directly influences the accuracy of the estimate. A study of business information systems managers and professionals at 112 different organizations using causal analysis with the Equations Modeling System (EQS) refined the model. The refined model shows that no managerial practice in this study discourages the use of intuition, guessing and personal memory in cost estimating. Although user commitment and accountability appear to foster algorithm-based estimating, such an algorithmic basis does not portend greater accuracy. Only one managerial practice-the use of the estimate in performance evaluations of software managers and professionals-presages greater accuracy. By implication, the research suggests somewhat ironically that the most effective approach to improve estimating accuracy may be to make estimators, developers and managers more accountable for the estimate even though it may be impossible to direct them explicitly on how to produce a more accurate one     

面向测量误差模型的外生变量估计方法

从观察数据中发现因果关系在近年来得到了越来越多学者的关注,其中外生变量在理解因果机制中扮演者重要的角色。然而,现有的因果发现方法大多假设观察变量就是真实发生的因(果)变量,忽略了测量误差带来的影响。为此,提出了一种解决测量模型下的外生变量估计方法。通过引入triad约束,根据此约束来找出与其余所有相关成对变量都满足triad约束的变量,即外生变量。该算法不仅能够解决含有测量误差数据的估计问题,而且对于没有测量误差的数据仍然适用。实验将该算法应用于真实网络产生的数据中,结果表明,无论变量是否含有测量误差,提出方法均优于现有的其他算法。同时,基于移动基站的真实数据实验也验证了算法的有效性。     

Multi-scale approach to estimating aboveground biomass in the Brazilian Amazon using Landsat and LiDAR data

ABSTRACT

Forest degradation from either natural or anthropogenic drivers involves processes that change the capacity of the ecosystem to provide services. In Brazil, estimates of carbon emissions do not currently take into account emissions from forest degradation caused by fire or by selective logging. Here, we present a methodology to estimate aboveground biomass in forest degradedareas, that can be accounted to estimate carbon emissions. We explored a multi-scale and temporal approach involving Airborne Laser Scanning (ALS) and orbital images from Landsat 8 Operational Land Imager (OLI) sensor to estimate the aboveground biomass. Cross-validation results showed that 49% of the variation in biomass could be explained using this approach, with an estimation error 58 Mg ha^?1 (49.08%). Due to the difficulty in measuring biomass in tropical forests, the proposed methodology can be an alternative in future works to estimate aboveground biomass in order to improve the estimates of carbon emissions by the governmental organizations.     

一维时间序列分形维数算法对比分析

分形维数在一维时间序列的分形特性分析中应用非常广泛,其计算方法多种多样,但是相关计算方法的全面对比鲜见文献报道。针对常用的八种一维时间序列分形维数计算方法,以WCF合成时间序列为研究对象,分别就算法的准确性和效率,对数据长度的依赖性进行分析对比。结果表明：准确性较好的三种算法是FA,DFA和Higuchi算法;而运算效率最高的是Sevcik,Katz和Castiglioni算法,但是它们的准确性偏低,而FA和Higuchi算法在计算时间上略微增加,但准确性比较高;在数据长度为4 096点时,各算法的计算值基本稳定,尤其是FA、Higuchi和DFA算法,在数据长度为4 096点时,计算值与理论值比较吻合。由此可以得出结论,Higuchi和DFA算法在计算一维时间序列的分形维数时性能优越,在相关的计算中优先选择。     

Daily evapotranspiration prediction using gradient boost regression model for irrigation planning

The Journal of Supercomputing - Appropriate irrigation planning with reference evapotranspiration (ETo) plays an important role in managing the water resources for agriculture. This paper presents...