Gee: explore the change of water area in the North Canal basin over the past 30 years [year by year]

---

Preface

Use GEE Analyze the change characteristics of water area in the basin , Take the North Canal basin as an example

For the same area analysis, see ：
GEE： Reclassification of land use in the North Canal basin

The analysis area is as follows ：

---

One 、 Analysis steps

Main analysis steps ：

Two 、python Code

1、 Load watershed range

python The code is as follows

#  Add display underlay , Select Gaode map 
Map = geemap.Map()
Map.add_basemap('HYBRID')

#  Import watershed range 
basin_shp = '../world_basins/byh.shp'
basin_bj = geemap.shp_to_ee(basin_shp)
roi = basin_bj.geometry()
#  Add display 
Map.addLayer(ee.Image().paint(roi, 0, 2), {
    'palette':'red'}, 'roi')
Map.centerObject(basin_bj)
Map

give the result as follows ：

2、 Image capture and synthesis

• geemap.landsat_timeseries()

python The code is as follows ：

#  Import image data , Annual synthesis 
images = geemap.landsat_timeseries(roi=roi,
                                    start_year=1984,
                                    end_year=2020,
                                    start_date='01-01',
                                    end_date='12-31')
#  Check the image band 
first = images.first()
first.bandNames().getInfo()
# ['Blue', 'Green', 'Red', 'NIR', 'SWIR1', 'SWIR2', 'pixel_qa', 'NBR']
#  Add display 
Map.addLayer(first, {
    'bands':['SWIR2', 'SWIR1', 'Green'], 'min':0, 'max':3000}, 'first image')

give the result as follows ：

3、NDWI Calculation and water surface extraction

• normalizedDifference()

• ee.Image.pixelArea()
  - ee.Image.reduceRegion()
  

python The code is as follows ：

def ndwi_cac(image):
    ndwi_image = image.normalizedDifference(['Green', 'SWIR1']).rename('ndwi')
    return ndwi_image.gt(0).selfMask()
    
#  Get water surface 
ndwi_water = images.map(ndwi_cac)
Map.addLayer(ndwi_water.first(), {
    'palette':'blue'}, 'first water')

#  Calculate the water surface area , The unit is according to km2 meter 
def cal_area(image):
    pixel_area = image.multiply(ee.Image.pixelArea()).divide(1e6)
    img_area = pixel_area.reduceRegion(**{
    
        'reducer':ee.Reducer.sum(),
        'geometry':roi,
        'scale':30,
        'maxPixels':1e13
    })
    return image.set({
    'water_area':img_area})

water_areas = ndwi_water.map(cal_area)
water_stats = water_areas.aggregate_array('water_area').getInfo()
water_stats

give the result as follows ：
[{‘ndwi’: 35.52171285398048},
{‘ndwi’: 49.82769552462259},
{‘ndwi’: 95.52709095289183},
{‘ndwi’: 88.0812142082916},
{‘ndwi’: 93.93069886911051},
{‘ndwi’: 80.60827039904903},
{‘ndwi’: 136.78296227087307},
{‘ndwi’: 108.33794441999756},
{‘ndwi’: 171.22477916253243},
{‘ndwi’: 127.52933278774321},
{‘ndwi’: 86.3748384304267},
{‘ndwi’: 112.22821871601487},
{‘ndwi’: 101.4823960050295},
{‘ndwi’: 89.49672596974511},
{‘ndwi’: 295.2877416742835},
{‘ndwi’: 80.7449547841241},
{‘ndwi’: 120.11593294853824},
{‘ndwi’: 134.62668476349447},
{‘ndwi’: 149.84151247550966},
{‘ndwi’: 236.29430841558016},
{‘ndwi’: 158.22021181601244},
{‘ndwi’: 70.20089856972656},
{‘ndwi’: 97.93744088662496},
{‘ndwi’: 108.04200576387466},
{‘ndwi’: 107.22519927537127},
{‘ndwi’: 63.4901824073373},
{‘ndwi’: 85.06409974042677},
{‘ndwi’: 74.55227930805393},
{‘ndwi’: 221.75472506976308},
{‘ndwi’: 85.00424085790755},
{‘ndwi’: 60.99660440626263},
{‘ndwi’: 68.4423834487381},
{‘ndwi’: 67.66773335986407},
{‘ndwi’: 84.48483915371168},
{‘ndwi’: 76.99267855924732},
{‘ndwi’: 58.5695615691552},
{‘ndwi’: 87.93523549212745}]

4、 Statistical analysis

python The code is as follows ：

#  Time series change histogram shows 
import matplotlib.pyplot as plt

x = list(range(1984, 2021))
y = [item.get('ndwi') for item in water_stats]

plt.bar(x, y, align='center', alpha=0.5)
plt.ylabel('Area hm2')
plt.title('surface water dynamics in BYH')
plt.show()

---

3、 ... and 、 comparative analysis

It is proposed to adopt JRC The data directly analyzes the changes of water surface , Its python The code is as follows ：
【 To be analyzed ...】

give the result as follows ：
Comparison and analysis of the difference between the two calculation results ：

1. It can be seen that ,

---

Four 、 Summary and discussion

1. The characteristics of long-time series water surface changes in the North Canal basin are analyzed
2. Problems to be explored , Water surface changes in different zones can be considered , Spatial and temporal changes of water surface in different administrative regions or upstream and downstream areas of the basin

---

Reference resources ：

• https://geemap.org/notebooks/tn_surface_water/
• https://developers.google.com/earth-engine/apidocs/ee-image-pixelarea?hl=en