黑河生态水文遥感试验：黑河流域中游覆盖度和生物量观测数据集（2013）

本数据包括大满超级站、湿地、沙漠、荒漠和戈壁五个站点植被一个生长周期内的覆盖度数据集以及大满超级站玉米和湿地芦苇两种植被一个生长周期内的生物量数据集。观测时间自2013年5月19日开始，9月15日结束。
1覆盖度观测
1.1观测时间
1.1.1超级站：观测时间段2013年5月20日-9月15日， 7月31日以前每5天观测一次，7月31后每10天观测一次，共做了18次观测，具体观测时间如下；
超级站：2013-5-20、2013-5-25、2013-5-30、2013-6-5、2013-6-10、2013-6-16、2013-6-22、2013-6-27、2013-7-2、2013-7-7、2013-7-12、2013-7-17、2013-7-27、2013-8-3、2013-8-13、2013-8-25、2013-9-5、2013-9-15
1.1.2其它四个站：观测时间段2013年5月20日-9月15日，每10天观测一次，共做了12次观测，具体观测时间如下；
其它四个站：2013-5-20、2013-6-5、2013-6-16、2013-6-27、2013-7-7、2013-7-17、2013-7-27、2013-8-3、2013-8-13、2013-8-25、2013-9-5、2013-9-15
1.2观测方法
1.2.1测量仪器与原理：
采用数码相机拍照的方法测量，将数码相机置于简易支撑杆前端的仪器平台，保持拍摄的竖直向下，远程控制相机测量数据。观测架可以用来改变相机的拍摄高度，面向不同类型植被实现有针对性的测量。
1.2.2样方的设计
超级站：共取3块样地，每块样地样方大小10×10米，每样地每次测量时沿两条对角线依次拍照，共取9-10张照片；
湿地站：共取2块样地，每块样地样方大小10×10米，每样地每次测量拍9-10张照片；
其它3个站：选取1块样地，每块样地样方大小10×10米，每样地每次测量拍9-10张照片；
1.2.3拍摄方法
针对超级站玉米和湿地站芦苇，直接采用观测架观测，保证观测架上的相机距离植被冠层的高度远大于植被冠幅，在方形样方内沿着对角线采样，然后做算术平均。在视场角度不大(<30°)的情况下，视场内包括大于2个整周期的垄行，相片的边长与垄行平行；其它三个站点由于植被比较低矮，直接用相机垂直向下拍照（未使用支架）。
1.2.4 覆盖度计算
覆盖度计算由北京师范大学完成，采用一种自动分类方法，具体见 “建议参考文献”第1条文献。通过RGB颜色空间转换到更容易区分绿色植被的Lab空间，对绿度分量a的直方图进行聚类，分离出绿色植被和非绿色背景2组分，获得单张相片的植被覆盖度。该方法的优点在于其算法简单、易于实现而且自动化程度和精度较高。今后还需要更多的快速、自动、准确的分类方法，最大限度发挥数码相机方法的优势。
2生物量观测
2.1观测时间
2.1.1玉米：观测时间段2013年5月20日-9月15日， 7月31日以前每5天观测一次，7月31后每10天观测一次，共做了18次观测，具体观测时间如下；
玉米：2013-5-20、2013-5-25、2013-5-30、2013-6-5、2013-6-10、2013-6-16、2013-6-22、2013-6-27、2013-7-2、2013-7-7、2013-7-12、2013-7-17、2013-7-27、2013-8-3、2013-8-13、2013-8-25、2013-9-5、2013-9-15
2.1.2芦苇：观测时间段2013年5月20日-9月15日，每10天观测一次，共做了12次观测，具体观测时间如下；
芦苇：2013-5-20、2013-6-5、2013-6-16、2013-6-27、2013-7-7、2013-7-17、2013-7-27、2013-8-3、2013-8-13、2013-8-25、2013-9-5、2013-9-15
2.2观测方法
玉米：选取3块样地，每块样地每次观测选取代表样地平均水平的三株玉米分别称每株玉米的鲜重（地上生物量+地下生物量）和相应的干重（85℃恒温烘干），根据种植的株距和行距计算单位面积玉米的生物量；
芦苇：设置2个0.5mÍ0.5m的样方，齐地刈割，分别称取芦苇的鲜重（茎叶）和干重（85℃恒温烘干）。
2.3观测仪器
天平（精度0.01g）、烘箱。
3数据的存储
所有观测数据先手薄记录后整理到Excel表中存储，同时整理了玉米种植结构数据，包括种植的株距、行距，种植时间、灌水时间、除父本时间以及收割时间等相关信息。

This dataset includes vegetation coverage datasets over one growing season from five stations: Daman Super Station, Wetland, Desert, Barren Land, and Gobi, as well as biomass datasets over one growing season for two vegetation types: maize from Daman Super Station and reed from Wetland. The observation period runs from May 19 to September 15, 2013.

1 Vegetation Coverage Observation
1.1 Observation Time
1.1.1 Daman Super Station: The observation period is May 20 to September 15, 2013. Observations were conducted every 5 days before July 31, and every 10 days after July 31, totaling 18 times. The specific observation dates are as follows:
Daman Super Station: 2013-5-20, 2013-5-25, 2013-5-30, 2013-6-5, 2013-6-10, 2013-6-16, 2013-6-22, 2013-6-27, 2013-7-2, 2013-7-7, 2013-7-12, 2013-7-17, 2013-7-27, 2013-8-3, 2013-8-13, 2013-8-25, 2013-9-5, 2013-9-15
1.1.2 The other four stations: The observation period is May 20 to September 15, 2013. Observations were conducted every 10 days, totaling 12 times. The specific observation dates are as follows:
The other four stations: 2013-5-20, 2013-6-5, 2013-6-16, 2013-6-27, 2013-7-7, 2013-7-17, 2013-7-27, 2013-8-3, 2013-8-13, 2013-8-25, 2013-9-5, 2013-9-15
1.2 Observation Methods
1.2.1 Measuring Instruments and Principles:
Vegetation coverage was measured using digital photography. The digital camera was mounted on an instrument platform at the front of a simple support pole, keeping the shooting direction vertically downward. The camera was controlled remotely for data collection. The observation rack could adjust the shooting height of the camera to conduct targeted measurements for different vegetation types.
1.2.2 Quadrat Design
For Daman Super Station: A total of 3 sample plots were set up, each with a quadrat size of 10×10 meters. During each measurement, photos were taken sequentially along the two diagonals of each sample plot, with 9 to 10 photos collected per plot.
For Wetland Station: A total of 2 sample plots were set up, each with a quadrat size of 10×10 meters. 9 to 10 photos were taken per plot during each measurement.
For the other three stations: 1 sample plot was selected, with a quadrat size of 10×10 meters. 9 to 10 photos were taken per plot during each measurement.
1.2.3 Shooting Methods
For maize at Daman Super Station and reed at Wetland Station: The observation rack was used directly. The height of the camera on the rack was set much higher than the vegetation canopy width. Sampling was conducted along the diagonals within the square quadrat, and the results were averaged arithmetically. When the field angle was small (<30°), the field of view included more than 2 complete ridge row cycles, and the side length of the photo was parallel to the ridge rows. For the other three stations, since the vegetation was relatively short, photos were taken vertically downward directly without using the support rack.
1.2.4 Coverage Calculation
The coverage calculation was completed by Beijing Normal University using an automatic classification method; please refer to the first literature in the "Recommended References" for details. The RGB color space was converted to the Lab color space, which better distinguishes green vegetation. Clustering was performed on the histogram of the greenness component a to separate green vegetation and non-green background, thereby obtaining the vegetation coverage of a single photo. The advantages of this method are simple algorithm, easy implementation, high automation level and high accuracy. In the future, more rapid, automatic and accurate classification methods are needed to maximize the advantages of the digital photography method.

2 Biomass Observation
2.1 Observation Time
2.1.1 Maize: The observation period is May 20 to September 15, 2013. Observations were conducted every 5 days before July 31, and every 10 days after July 31, totaling 18 times. The specific observation dates are as follows:
Maize: 2013-5-20, 2013-5-25, 2013-5-30, 2013-6-5, 2013-6-10, 2013-6-16, 2013-6-22, 2013-6-27, 2013-7-2, 2013-7-7, 2013-7-12, 2013-7-17, 2013-7-27, 2013-8-3, 2013-8-13, 2013-8-25, 2013-9-5, 2013-9-15
2.1.2 Reed: The observation period is May 20 to September 15, 2013. Observations were conducted every 10 days, totaling 12 times. The specific observation dates are as follows:
Reed: 2013-5-20, 2013-6-5, 2013-6-16, 2013-6-27, 2013-7-7, 2013-7-17, 2013-7-27, 2013-8-3, 2013-8-13, 2013-8-25, 2013-9-5, 2013-9-15
2.2 Observation Methods
For Maize: 3 sample plots were selected. During each observation, 3 maize plants representative of the average level of the plot were chosen. The fresh weight (above-ground biomass + below-ground biomass) and corresponding dry weight (dried at a constant temperature of 85℃) of each plant were measured. The biomass per unit area of maize was calculated based on the plant spacing and row spacing of planting.
For Reed: 2 quadrats of 0.5m × 0.5m were set up. The reeds were cut at ground level, and the fresh weight (stems and leaves) and dry weight (dried at a constant temperature of 85℃) were measured respectively.
2.3 Observation Instruments
Balance (accuracy: 0.01g) and oven.

3 Data Storage
All observation data were first recorded in handwritten notebooks and then organized and stored in Excel spreadsheets. In addition, maize planting structure data were organized, including relevant information such as plant spacing, row spacing, planting time, irrigation time, time of removing the male parent and harvesting time.