农业生态环境-遥感和智慧农业Agro-ecosystem & Environment—Romote sensing & Smart agriculture

默认 最新文章 浏览次数
Please wait a minute...
选择: 显示/隐藏图片
1. Soil temperature estimation at different depths, using remotely-sensed data
HUANG Ran, HUANG Jian-xi, ZHANG Chao, MA Hong-yuan, ZHUO Wen, CHEN Ying-yi, ZHU De-hai, Qingling WU, Lamin R. MANSARAY
Journal of Integrative Agriculture    2020, 19 (1): 277-290.   DOI: 10.1016/S2095-3119(19)62657-2
摘要110)      PDF    收藏
Soil temperatures at different depths down the soil profile are important agro-meteorological indicators which are necessary for ecological modeling and precision agricultural activities.  In this paper, using time series of soil temperature (ST) measured at different depths (0, 5, 10, 20, and 40 cm) at agro-meteorological stations in northern China as reference data, ST was estimated from land surface temperature (LST) and normalized difference vegetation index (NDVI) derived from AQUA/TERRA MODIS data, and solar declination (Ds) in univariate and multivariate linear regression models.  Results showed that when daytime LST is used as predictor, the coefficient of determination (R2) values decrease from the 0 cm layer to the 40 cm layer.  Additionally, with the use of nighttime LST as predictor, the R2 values were relatively higher at 5, 10 and 15 cm depths than those at 0, 20 and 40 cm depths.  It is further observed that the multiple linear regression models for soil temperature estimation outperform the univariate linear regression models based on the root mean squared errors (RMSEs) and R2.  These results have demonstrated the potential of MODIS data in tandem with the Ds parameter for soil temperature estimation at the upper layers of the soil profile where plant roots grow in.  To the best of our knowledge, this is the first attempt at the synergistic use of
参考文献 | 相关文章 | 多维度评价
2. Mapping the fallowed area of paddy fields on Sanjiang Plain of Northeast China to assist water security assessments
LUO Chong, LIU Huan-jun, FU Qiang, GUAN Hai-xiang, YE Qiang, ZHANG Xin-le, KONG Fan-chang
Journal of Integrative Agriculture    2020, 19 (7): 1885-1896.   DOI: 10.1016/S2095-3119(19)62871-6
摘要147)      PDF    收藏
Rice growth requires a large amount of water, and planting rice will increase the contradiction between supply and demand of water resources.  Paddy field fallowing is important for the sustainable development of an agricultural region, but it remains a great challenge to accurately and quickly monitor the extent and area of fallowed paddy fields.  Paddy fields have unique physical features associated with paddy rice during the flooding and transplanting phases.  By comparing the differences in phenology before and after paddy field fallowing, we proposed a phenology-based fallowed paddy field mapping algorithm.  We used the Google Earth Engine (GEE) cloud computing platform and Landsat 8 images to extract the fallowed paddy field area on Sanjiang Plain of China in 2018.  The results indicated that the Landsat8, GEE, and phenology-based fallowed paddy field mapping algorithm can effectively support the mapping of fallowed paddy fields on Sanjiang Plain of China.  Based on remote sensing monitoring, the total fallowed paddy field area of Sanjiang Plain is 91 543 ha.  The resultant fallowed paddy field map is of high accuracy, with a producer (user) accuracy of 83% (81%), based on validation using ground-truth samples.  The Landsat-based map also exhibits high consistency with the agricultural statistical data.  We estimated that paddy field fallowing reduced irrigation water by 384–521 million cubic meters on Sanjiang Plain in 2018.  The research results can support subsidization grants for fallowed paddy fields, the evaluation of fallowed paddy field effects and improvement in subsequent fallowed paddy field policy in the future. 
 
参考文献 | 相关文章 | 多维度评价
3. Early-season crop type mapping using 30-m reference time series
HAO Peng-yu, TANG Hua-jun, CHEN Zhong-xin, MENG Qing-yan, KANG Yu-peng
Journal of Integrative Agriculture    2020, 19 (7): 1897-1911.   DOI: 10.1016/S2095-3119(19)62812-1
摘要130)      PDF    收藏
Early-season crop type mapping could provide important information for crop growth monitoring and yield prediction, but the lack of ground-surveyed training samples is the main challenge for crop type identification.  Although reference time series based method (RBM) has been proposed to identify crop types without the use of ground-surveyed training samples, the methods are not suitable for study regions with small field size because the reference time series are mainly generated using data set with low spatial resolution.  As the combination of Landsat data and Sentinel-2 data could increase the temporal resolution of 30-m image time series, we improved the RBM by generating reference normalized difference vegetation index (NDVI)/enhanced vegetation index (EVI) time series at 30-m resolution (30-m RBM) using both Landsat and Sentinel-2 data, then tried to estimate the potential of the reference NDVI/EVI time series for crop identification at early season.  As a test case, we tried to use the 30-m RBM to identify major crop types in Hengshui, China at early season of 2018, the results showed that when the time series of the entire growing season were used for classification, overall classification accuracies of the 30-m RBM were higher than 95%, which were similar to the accuracies acquired using the ground-surveyed training samples.  In addition, cotton, spring maize and summer maize distribution could be accurately generated 8, 6 and 8 weeks before their harvest using the 30-m RBM; but winter wheat can only be accurately identified around the harvest time phase.  Finally, NDVI outperformed EVI for crop type classification as NDVI had better separability for distinguishing crops at the green-up time phases.  Comparing with the previous RBM, advantage of 30-m RBM is that the method could use the samples of the small fields to generate reference time series and process image time series with missing value for early-season crop classification; while, samples collected from multiple years should be further used so that the reference time series could contain more crop growth conditions.
 
参考文献 | 相关文章 | 多维度评价
4. A case-based method of selecting covariates for digital soil mapping
LIANG Peng, QIN Cheng-zhi, ZHU A-xing, HOU Zhi-wei, FAN Nai-qing, WANG Yi-jie
Journal of Integrative Agriculture    2020, 19 (8): 2127-2136.   DOI: 10.1016/S2095-3119(19)62857-1
摘要116)      PDF    收藏
Selecting a proper set of covariates is one of the most important factors that influence the accuracy of digital soil mapping (DSM).  The statistical or machine learning methods for selecting DSM covariates are not available for those situations with limited samples.  To solve the problem, this paper proposed a case-based method which could formalize the covariate selection knowledge contained in practical DSM applications.  The proposed method trained Random Forest (RF) classifiers with DSM cases extracted from the practical DSM applications and then used the trained classifiers to determine whether each one potential covariate should be used in a new DSM application.  In this study, we took topographic covariates as examples of covariates and extracted 191 DSM cases from 56 peer-reviewed journal articles to evaluate the performance of the proposed case-based method by Leave-One-Out cross validation.  Compared with a novices’ commonly-used way of selecting DSM covariates, the proposed case-based method improved more than 30% accuracy according to three quantitative evaluation indices (i.e., recall, precision, and F1-score).  The proposed method could be also applied to selecting the proper set of covariates for other similar geographical modeling domains, such as landslide susceptibility mapping, and species distribution modeling.
 
参考文献 | 相关文章 | 多维度评价
5. Comparison of machine learning algorithms for mapping mango plantations based on Gaofen-1 imagery
LUO Hong-xia, DAI Sheng-pei, LI Mao-fen, LIU En-ping, ZHENG Qian, HU Ying-ying, YI Xiao-ping
Journal of Integrative Agriculture    2020, 19 (11): 2815-2828.   DOI: 10.1016/S2095-3119(20)63208-7
摘要109)      PDF    收藏
Mango is a commercial crop on Hainan Island, China, that is cultivated to develop the tropical rural economy.  The development of accurate and up-to-date maps of the spatial distribution of mango plantations is necessary for agricultural monitoring and decision management by the local government.  Pixel-based and object-oriented image analysis methods for mapping mango plantations were compared using two machine learning algorithms (support vector machine (SVM) and Random Forest (RF)) based on Chinese high-resolution Gaofen-1 (GF-1) imagery in parts of Hainan Island.  To assess the importance of different features on classification accuracy, a combined layer of four original bands, 32 gray-level co-occurrence (GLCM) texture indices, and 10 vegetation indices were used as input features.  Then five different sets of variables (5, 10, 20, and 30 input variables and all 46 variables) were classified with the two machine learning algorithms at object-based level.  Results of the feature optimization suggested that homogeneity and variance were very important variables for distinguishing mango plantations patches.  The object-based classifiers could significantly improve overall accuracy between 2–7% when compared to pixel-based classifiers.  When there were 5 and 10 input variables, SVM showed higher classification accuracy than RF, and when the input variables exceeded 20, RF showed better performances.  After the accuracy achieved saturation points, there were only slightly classification accuracy improvements along with the numbers of feature increases for both of SVM and RF classifiers.  The results indicated that GF-1 imagery can be successfully applied to mango plantation mapping in tropical regions, which would provide a useful framework for accurate tropical agriculture land management. 
参考文献 | 相关文章 | 多维度评价
6. JIA-2021-0227 基于统计数据空间化的农作物空间格局模拟模型
XIA Tian, WU Wen-bin, ZHOU Qing-bo, Peter H. VERBURG, YANG Peng, HU Qiong, YE Li-ming, ZHU Xiao-juan
Journal of Integrative Agriculture    2022, 21 (6): 1786-1789.   DOI: 10.1016/S2095-3119(21)63713-9
摘要232)      PDF    收藏

本研究提出一种统计数据空间化的方法构建多时像农作物种植格局空间数据集来解决数据缺失的问题。该方法采用两层嵌套结构实现土地利用层和农作物层模拟,其中第一层模拟的耕地数据用于控制第二层农作物种植格局空间模拟范围。第二层农作物层采用空间迭代的方法按分配规则进行农作物面积统计数据空间化,最终实现农作物空间格局动态模拟。该模型在中国黑龙江省地区进行2000-2019年农作物空间格局模拟,结果表明模型模拟精度较高,能够实现长时间序列的农作物种植面积统计数据空间化应用,未来该模型能广泛应用于农业土地系统各方面研究及生产应用


参考文献 | 相关文章 | 多维度评价
7. Monthly composites from Sentinel-1 and Sentinel-2 images for regional major crop mapping with Google Earth Engine
LUO Chong, LIU Huan-jun, LU Lü-ping, LIU Zheng-rong, KONG Fan-chang, ZHANG Xin-le
Journal of Integrative Agriculture    2021, 20 (7): 1944-1957.   DOI: 10.1016/S2095-3119(20)63329-9
摘要131)      PDF    收藏

快速、准确地获取大区域、高分辨率的作物类型分布图对农业精准管理与可持续发展具有重要意义。受遥感影像质量和数据处理能力的限制,使用遥感技术进行大尺度的作物分类仍是一项巨大的挑战。本研究的目的是使用Google Earth Engine(GEE)结合Sentinel-1和Sentinel-2影像绘制黑龙江省的作物分布图,首先收集2018年作物生长关键期(5月至9月)覆盖研究区域所有可用的Sentinel-1与Sentinel-2影像,并对影像进行月度合成,然后将月度合成影像的不同反射率波段、植被指数与极化波段作为输入量结合随机森林方法进行作物分类。结果表明使用本研究提出的方法可以准确地获得黑龙江省作物分布图,作物分类总体精度达到89.75%。本研究还发现相比仅使用传统波段(可见光波段和近红外波段)进行作物分类,增加短波红外波段可以显著改善作物分类的准确性,其次是增加红边波段,增加常见植被指数和Sentinel-1数据对作物分类的精度提升不大。本研究还分析了作物分类的时效性,结果表明当7月份的影像可用时,作物分类精度的提升幅度最大,作物分类的总体精度可以达到80%以上。本研究结果为基于遥感的大尺度、高分辨率作物分布图的制作提供支持。


参考文献 | 相关文章 | 多维度评价
8. Winter wheat yield estimation based on assimilated Sentinel-2 images with the CERES-Wheat model
LIU Zheng-chun, WANG Chao, BI Ru-tian, ZHU Hong-fen, HE Peng, JING Yao-dong, YANG Wu-de
Journal of Integrative Agriculture    2021, 20 (7): 1958-1968.   DOI: 10.1016/S2095-3119(20)63483-9
摘要119)      PDF    收藏

为有效验证Sentinel-2影像与CERES-Wheat模型同化进而提高区域作物估产的精度,本文以中国黄土高原东南部三个县(襄汾县、新绛县和闻喜县)为研究区,应用集合卡尔曼滤波算法同化Sentinel-2影像反演的LAI和CERES-Wheat模型模拟的LAI,得到冬小麦生长期逐日的LAI同化值。对比改进的层次分析法、熵值法和归一组合赋权法对不同生育期LAI赋权,并与冬小麦实测单产值进行模型构建,进而对作物进行准确估产。研究结果表明:(1)同化LAI遵循了模拟LAI在冬小麦生育期的生长变化趋势,且在Sentinel-2影像反演LAI的修正下,返青期至抽穗-灌浆期的LAI得到提高,乳熟期的LAI下降减缓,更符合冬小麦LAI的实际生长变化情况;(2)基于实测LAI数据的检验表明,同化LAI比模拟值和反演值的RMSE分别降低了0. 43 m2/m2、0.29 m2/m2,同化过程提高了时间序列LAI的估测精度;(3)归一组合赋权法计算的加权同化LAI与实测单产构建的回归模型决定系数最高R2为0.8627,RMSE最小472.92kg/ha,应用此模型对研究区冬小麦进行估产,县域估测平均单产与统计单产相对误差均小于1%,证明高时空分辨率的Sentinel-2数据融入作物模型能得到更高精度的区域估产结果。


参考文献 | 相关文章 | 多维度评价
9. Simple method for extracting the seasonal signals of photochemical reflectance index and normalized difference vegetation index measured using a spectral reflectance sensor
Jae-Hyun RYU, Dohyeok OH, Jaeil CHO
Journal of Integrative Agriculture    2021, 20 (7): 1969-1986.   DOI: 10.1016/S2095-3119(20)63410-4
摘要69)      PDF    收藏
A spectral reflectance sensor (SRS) fixed on the near-surface ground was developed to support the continuous monitoring of vegetation indices such as the normalized difference vegetation index (NDVI) and photochemical reflectance index (PRI).  NDVI is useful for indicating crop growth/phenology, whereas PRI was developed for observing physiological conditions.  Thus, the seasonal change patterns of NDVI and PRI are two valuable pieces of information in a crop-monitoring system.  However, capturing the seasonal patterns is considered challenging because the vegetation index values estimated by the reflection from vegetation are often governed by meteorological conditions, such as solar irradiance and precipitation.  Further, unlike growth/phenology, the physiological condition has diurnal changes as well as seasonal characteristics.  This study proposed a novel filtering method for extracting the seasonal signals of SRS-based NDVI and PRI in paddy rice, barley, and garlic.  First, the measurement accuracy of SRSs was compared with handheld spectrometers, and the R2 values between the two devices were 0.96 and 0.81 for NDVI and PRI, respectively.  Second, the experimental study of threshold criteria with respect to meteorological variables (i.e., insolation, cloudiness, sunshine duration, and precipitation) was conducted, and sunshine duration was the most useful one for excluding distorted values of the vegetation indices.  After data processing based on sunshine duration, the R2 values between the measured vegetation indices and the extracted seasonal signals of vegetation indices increased by approximately 0.002–0.004 (NDVI) and 0.065–0.298 (PRI) on the three crops, and the seasonal signals of vegetation indices became noticeably improved.  This method will contribute to an agricultural monitoring system by identifying the seasonal changes in crop growth and physiological conditions.
参考文献 | 相关文章 | 多维度评价
10. An entirely new approach based on remote sensing data to calculate the nitrogen nutrition index of winter wheat
ZHAO Yu, WANG Jian-wen, CHEN Li-ping, FU Yuan-yuan, ZHU Hong-chun, FENG Hai-kuan, XU Xin-gang, LI Zhen-hai
Journal of Integrative Agriculture    2021, 20 (9): 2535-2551.   DOI: 10.1016/S2095-3119(20)63379-2
摘要215)      PDF    收藏

氮素营养指数(NNI)是作物氮素诊断的可靠指标。然而,目前还没有适用于多生育时期NNI反演的光谱指数。为克服传统NNI直接反演方法(NNIT1)和通过反演生物量(AGB)和植株氮浓度(PNC)进行NNI间接反演方法(NNIT2)在多生育期应用的局限性,本文构建了一个新的NNI遥感指数(NNIRS)。本文基于连续四年(2012–2013(Exp.1),2013–2014(Exp.2),20142015(Exp.3)和20152016(Exp.4))的冬小麦田间试验,采用交叉验证方法利用氮素相关植被指数和生物量相关植被指数构建了遥感关键氮浓度稀释曲线(Nc_RS)和根据NNI构建原理得到的NNIRS进行综合评价。结果表明:(1)由标准叶面积指数决定指数(sLAIDI)和红边叶绿素指数(CIred edge)构建的NNIRS模型表达式为NNIRS=CIred edge/(a×sLAIDIb),在Exp.1/2/4,Exp.1/2/3,Exp.1/3/4和Exp.2/3/4中参数“a”分别等于2.06,2.10,2.08和2.02,参数“b”分别等于0.66,0.73,0.67和0.62;(2)与NNIT1和NNIT2模型相比,NNIRS模型的精度最高(R2的范围为0.50–0.82,RMSE的范围为0.12–0.14);(3)NNIRS在验证数据集中也达到了较好的精度,RMSE分别为0.09,0.18,0.13和0.10。因此,本文认为NNIRS模型在氮素遥感诊断中具有较大的潜力。


参考文献 | 相关文章 | 多维度评价
11. 中国松嫩平原土地利用/覆盖变化对降水和气温时空变异的影响
CHU Xiao-lei, LU Zhong, WEI Dan, LEI Guo-ping
Journal of Integrative Agriculture    2022, 21 (1): 235-248.   DOI: 10.1016/S2095-3119(20)63495-5
摘要185)      PDF    收藏

土地利用/覆盖变化(LUCC)对区域气候的影响是实现土地利用系统可持续发展和减缓全球气候变化的关键,然而关于LUCC对降水和气温等气候因子变化影响的定量分析研究仍十分有限。本研究采用统计和重心模型模拟相结合的方法,定量分析了1980-2018年间我国东北松嫩平原LUCC对降水和气温的长期影响。结果发现LUCC的时空变化特征如下:该地区水田面积增加最多(15,166.43 km2),旱地次之,由于过度的农业开发利用导致湿地减少最多(19,977.13 km2);1980年以来该地区年平均降水量-9.89 mm/10a的速率呈现下降趋势,年平均气温变化则呈显著上升趋势,上升变化率为0.256℃/10a进一步通过重心模型模拟发现:水田、林地以及湿地重心变化与降水重心的变化呈正相关,建筑用地、旱田与未利用地重心变化与降水重心的变化呈负相关,林地较其它用地类型对年降水量增加的促进作用最为明显。建筑用地是对年平均气温增加起促进作用最大的用地类型,最小的是林地。总之,在区域尺度下土地利用/覆盖变化分析表明湿地减少、建筑用地和农业用地增加导致了松嫩平原持续干旱和气温快速变暖。

参考文献 | 相关文章 | 多维度评价
12. 利用机器学习和环境关联对大面积复杂地区土体深度进行空间预测
LIU Feng, YANG Fei, ZHAO Yu-guo, ZHANG Gan-lin, LI De-cheng
Journal of Integrative Agriculture    2022, 21 (8): 2422-2434.   DOI: 10.1016/S2095-3119(21)63692-4
摘要152)      PDF    收藏

本研究构建了一个可直接估计空间不确定性的集合式机器学习模型,即分位数回归森林,定量土体深度与环境条件之间的关系。将该模型与丰富的环境协同变量结合,预测了位于我国西北地区、面积为14万km2的黑河流域的土体深度空间分布,估算了制图结果的空间不确定性。使用了275个土体深度观测样本和26个环境协同变量数据。结果显示,模型预测精度R2为0.587,RMSE为2.98 cm(平方根尺度),可解释近60%的土体深度变异。土体深度图清晰地展示了土体深度的区域分布模式和局部细节。谷底、平原等低平低洼景观部位土体深度较大,而山坡、山脊、台地等高陡景观部位土体深度较小;绿洲内土体深度明显大于绿洲之外的荒漠地区,冲积平原中部土体深度明显大于边缘地带,而湖泊平原中部土体深度明显小于边缘地带。高的预测不确定性主要出现在可达性差、缺少样本的区域。分析发现,土壤发生过程和地貌过程共同塑造了该流域土体深度的空间模式,但地貌过程起主导作用。这一点可能也适用于世界上其它寒旱地区类似的“高寒山地-平原绿洲-荒漠戈壁”流域。


参考文献 | 相关文章 | 多维度评价