Technology

1.Data resource and preprocessing

The data used in this knowledge application includes meteorological data, basic geographic data, remote sensing data, statistical yearbook data, etc. The meteorological data comes from the daily temperature (average, minimum temperature) and precipitation data which observed by the international exchange weather station provided by the National Oceanic and Atmospheric Administration (NOAA) ; the basic geographic data adopts the 30m ASTER-DEM digital elevation model, and the remote sensing meteorological products are divided into two types: Temperature products include 2m hourly temperature data from ECMWF Re-Analysis 5 (ERA-5) with a spatial resolution of 0.25°×0.25; precipitation products include daily global precipitation measure (GPM) with a spatial resolution of 0.1°×0.1°, spatial 3-hour Multi-Source Weighted-Ensemble Precipitation (MSWEP) with a spatial resolution of 0.25°×0.25°, and the fifth-generation ECMWF atmospheric re-analysis global climate data hourly rainfall data with a spatial resolution of 0.25°×0.25°. Precipitation products include the daily Global Precipitation Measure (GPM) with a spatial resolution of 0.1°×0.1°and the 3-hour Multi-Source Weighted-Ensemble Precipitation (MSWEP) with a spatial resolution of 0.25°×0.25°, and hourly rainfall data from ECMWF Re-Analysis 5 (ERA-5). The data of statistical yearbook includes industrial statistical data provided by economic and social data research platforms of various countries. The data preprocessing part includes the following processes: missing value processing of unit conversion for meteorological data and time series combination. Spatio-temporal condition screening and vector mask extraction for remote sensing products; The process of digital collation for statistical yearbook data.

2.Technical flow

The application of this knowledge is based on different meteorological elements and the characteristics of the research object. It reveals and analyzes the ecological conditions of the Heilongjiang River Basin from the four perspectives of accumulated temperature, precipitation, frost, and drought, and analyzes the applicability and simulation accuracy of each remote sensing product in the basin.

2.1 Analysis of accumulated temperature changes in Heilongjiang watershed based on grid data

Temperature is the primary meteorological factor affecting crop growth, phenological phenomena and agricultural activities. Temperatures of 0 ℃, 5 ℃, 10 ℃ and 15 ℃ are usually defined as the lowest limit for the suitable growth of different types of crops. The accumulated temperature above these threshold temperatures is called accumulated temperature. The accumulated temperature which is an important indicator to measure the heat resources required by crops, and it will also affect their growth process and yield.

In the process of defining the start and end dates of accumulated temperature, the longest period of time when the 5-day moving average is greater than the base temperature is selected. In the first 5 days of this period, the first date when the daily average temperature is greater than the limit temperature is selected. As the starting day when the daily average temperature is stably greater than the limit temperature accumulated temperature; in the last 5 days, the last day when the daily average temperature is greater than the limit temperature accumulated temperature is selected as the end date.
In the process of defining the accumulated temperature zoning, combined with the actual local climate conditions, the regional heat resources are divided according to the accumulated temperature ≥ 10℃. The heat standards of warm temperate zone, middle temperate zone and cold temperate zone are as follows: 3400 ℃≤accumulated temperature＜4500 ℃, 1600 ℃≤accumulated temperature＜3400 ℃, and accumulated temperature＜1600 ℃.

2.2 Analysis of frost disaster changes in Heilongjiang watershed based on grid data

Frost is a short-duration agro-meteorological disaster. It is a phenomenon in which the temperature of plant stems and leaves drops below 0°C due to the decrease of the daily minimum temperature, which causes frostbite of growing plants. The frequency and intensity of frost will have a significant impact on crop yields.

In the process of defining the first and final frost days (the start and end dates of frost), the two points are respectively defined as the first and last day's lowest temperature in the two points ranges from January-July and August-December of a certain year. The date of the threshold temperature. The frost-free period between the two points is the number of days between the final frost day and the first frost day in the same year. If it does not appear in the previous year, the first and last frost days of the previous year will be null.

For frost disaster intensity calculation, Tmin, the lowest daily temperature, was taken as the judging variable according to the Chinese meteorological industry standard "Crop Frost Injury Rating". Frost hazards can be divided into three intensity grades: mild( -3℃≤Tmin≤0℃), moderate (-5℃≤Tmin≤-3℃) and severe(Tmin≤-5℃)in the frost-free period (growing season) between the last frost date and the first frost date.

Based on the above results, the spatial cluster analysis method is used to study the frost disaster zoning within China. The clustering algorithm uses the Kmeans algorithm in Python's sklearn library, and the clustering features use the information of the frost event and frost intensity level.

2.3 Applicability analysis of remotely sensed precipitation products in the Heilongjiang River Basin

Remote sensing precipitation products, as a supplement to ground rain gauge and radar observation of precipitation, adopting an important role in the precipitation simulation process in high-altitude mountainous areas and areas lacking observational material. At present, relevant institutions and researchers have produced a variety of precipitation data, and each precipitation data has a unique method to produce. Therefore, it is necessary to evaluate the reliability of these precipitations before use.

The study selects mean error (ME), root mean squared error (RMSE), and coefficient of determination R2 (R Squared) to reflect the difference between different precipitation products and measured values to evaluate their applicability in the basin and visualize them in time and space.

In terms of rainfall intensity evaluation, the study refers to the relevant classification standards of the China Meteorological Administration on rainfall intensity to compare the probability density function (PDF) of events that occur between each remote sensing product and the measured value under different levels of rainfall intensity conditions. Verifying the ability and reliability indicators of each product in simulating the process of various rainfall events. In the aspect of rainfall intensity assessment, the probability density function of events occurring under different rainfall intensity conditions based on remote sensing products and measured values is studied by referring to the relevant classification standards of rainfall intensity of China Meteorological Administration to verify the ability and reliability indexes of each product in simulating various rainfall events.

2.4 The Temperature Vegetation Dryness Index during the growing season

The Temperature Vegetation Dryness Index (TVDI) is a measurement parameter that estimates the soil surface moisture based on the correlation between the vegetation index and the surface temperature and reflects the degree of drought in the area. Combining the 2 dimentional feature space constructed by ASTER-DEM data, Normalized Difference Vegetation Index, NDVI) and Land Surface Temperature (LST), the above indices are inverted and visualized on time and space.

3.Reference

[1] Bai L, Zhang F, Shang M, et al. Evolution of the multiple accumulated temperature across mainland China in 1961-2018 with the gridded meteorological dataset[J]. Journal of Geo-information Science, 2021,23(8):1446-1460. [2] Bai L, Zhang F, Wen YQ, et al. Evolution of the frost disaster across mainland China in 1961-2018 with the gridded meteorological dataset[J]. Chinese Journal of Agrometeorology,2021,42(9):761-774. [3] Bolgov M V, MD Trubetskova, IA Filippova. Characteristics of extreme precipitation events within the Amur River basin in summer 2013[J]. Geography & Natural Resources, 2017, 38(2):139-146. [4] Dulamsuren C, Hauck M, Leuschner C. Recent drought stress leads to growth reductions in Larix sibirica in the western Khentey, Mongolia[J]. Global Change Biology, 2010, 16(11):3024-3035. [5] Zhou, Y. Z., Wang, J. L., Li, K. Temperature vegetation dryness index 1-km grid dataset in Heilongjiang River Basin (20072018) [J]. Journal of Global Change Data & Discovery, 2021, 5(4): 1-13. 2021.04.01. CSTR:20146.14.2021.04.01.