钱凤魁, 项子璇, 王贺兴, 顾汉龙. 基于最小数据集与LESA体系的县域耕地质量评价[J]. 农业工程学报, 2023, 39(8): 239-248. DOI: 10.11975/j.issn.1002-6819.202210249
    引用本文: 钱凤魁, 项子璇, 王贺兴, 顾汉龙. 基于最小数据集与LESA体系的县域耕地质量评价[J]. 农业工程学报, 2023, 39(8): 239-248. DOI: 10.11975/j.issn.1002-6819.202210249
    QIAN Fengkui, XIANG Zixuan, WANG Hexing, GU Hanlong. Evaluating cultivated land quality in county territory using the minimum data set, land evaluation and site assessment (LESA)[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2023, 39(8): 239-248. DOI: 10.11975/j.issn.1002-6819.202210249
    Citation: QIAN Fengkui, XIANG Zixuan, WANG Hexing, GU Hanlong. Evaluating cultivated land quality in county territory using the minimum data set, land evaluation and site assessment (LESA)[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2023, 39(8): 239-248. DOI: 10.11975/j.issn.1002-6819.202210249

    基于最小数据集与LESA体系的县域耕地质量评价

    Evaluating cultivated land quality in county territory using the minimum data set, land evaluation and site assessment (LESA)

    • 摘要: 建立科学合理的县域耕地质量评价体系对快速准确摸清耕地质量家底和建立耕地保护分区具有重要指导意义。该研究基于主成分分析法建立最小数据集精选指标,构建土地评价与立地条件分析(land evaluation and site assessment,LESA)体系,开展耕地质量综合评价,分析耕地质量区域分布特征及差异性特征并划定耕地保护分区。结果表明:1)自然质量指标最小数据集由砂粒、有机质、全钾、有效磷、pH值、综合污染指数、耕层质地、容重、阳离子交换量组成,立地环境指标最小数据集包括排水条件、连片度、生态兼容性、河流距离、路网密度、灌溉能力、农田林网化率、耕地利用类型。2)采用LESA评价模型计算耕地自然质量指数及立地环境指数,确定LESA体系为FLESA=0.5FLE+0.5FSA(FSA、FLE、FLESA分别为耕地立地环境条件、自然质量条件和综合分值),采样点综合评分为51.517~81.838。综合比选各插值误差检验结果后采用普通克里金法进行耕地质量结果空间插值,评价单元耕地质量综合评分为52.148~79.624。3)铁岭县耕地资源可划分为5个等级区:1级区划分为永久基本农田核心保护区,占比20.52%;2级区划分为耕地自然地力条件重点治理区,占比36.79%;3级区和4级区耕地土壤和立地条件均存在多样性的限制因素,可划分为耕地综合整治区,占比36.33%;5级区划分为耕地生态自然保育区,占比6.36%。4)经计算基于最小数据集与LESA相结合的评价结果有效系数为0.615,相对偏差系数为0.009,说明该体系耕地质量评价结果准确,可信度较高。该研究成果简化了县域耕地质量评价指标体系,量化了自然质量与立地条件协同关系,为开展耕地质量提升和保护利用提供了理论和方法依据。

       

      Abstract: Abstract: County-cultivated land can be evaluated to rapidly and accurately determine the quality background for the protection zone. In this study, the minimum data set was established to streamline the selection index using principal component analysis (PCA). A land evaluation and site assessment (LESA) system was then constructed to comprehensively evaluate the cultivated land quality. The regional distribution of cultivated land quality was finally obtained to divide the cultivated land protection zone. The results showed that: 1) The minimum data set of the natural quality index was composed of sand, organic matter, total potassium, available phosphorus, pH, soil comprehensive pollution index, topsoil texture, bulk density, and cation exchange capacity. The minimum data set of the site condition index was the drainage conditions, consecutive degree, ecological compatibility, river distance, road network density, irrigation capacity, farmland forest network rate, and cultivated land utilization type. 2) The LESA model was used to calculate the natural quality score of cultivated land and the site environment. The coupled cooperative model was used to determine the LESA system as FLESA=0.5FLE+0.5FSA. The comprehensive evaluation score of the sample point ranged from 51.517 to 81.838. The interpolation error test was carried out to combine each space interpolation. The ordinary Kriging method was also used for the spatial interpolation of cultivated land quality. The comprehensive score of cultivated land quality was 52.148 to 79.624 in the evaluation unit. 3) The overall distribution trend of cultivated land quality was "excellent in the center, but inferior in the east and west". The cultivated land resources in Tieling County were divided into five grades: The cultivated land area of Grade 1 was 22 294.396 hm2 (accounting for 20.52%), indicating the permanent protection area of basic cultivated land. The Grade 2 area was 39 974.407 hm2 (accounting for 36.79%), including the key control area in the natural fertility conditions of cultivated land. The Grade 3 and 4 areas were 25 649.334, and 13 837.926 hm2, respectively. Less diversity was found under the soil and site conditions of cultivated land in Grade 3 and 4 areas. The comprehensive improvement area of cultivated land then accounted for 36.33%. The Grade 5 area was 6 913.914 hm2 (accounting for 6.36%), indicating the ecological and natural conservation area of cultivated land. 4) The minimum data set of the indicator screening filter rate was 50%. The redundancy between the indicators was removed to significantly simplify the indicator for the key characteristics. A comparison was performed on the Nash effective coefficient and relative deviation coefficient. The evaluation value of cultivated land quality using MDS and LESA was closer to the benchmark, indicating a small deviation. The index system was simplified to quantify the synergistic relationship between natural quality and site conditions during quality evaluation. The finding can provide a strong reference to improve the quality, protection, and utilization of the county-cultivated land.

       

    /

    返回文章
    返回