GISdevelopment.net ---> AARS ---> ACRS 1990 ---> Poster Session

Influence of the cultural practices to the reflecting spectra in the wheat fields

Zhang Hongmin, wang Jiashen, Zhang Xioping, Lu Zhiguang and Cao Daicheng
Agricultural Remote Sensing Center of Beijing Agricultural
University Beijing, China


Abstract
This article deals with the influence of the main cultural practices (including the different sowing period, the different planting density and the different fertilizer level,ect.) to the reflecting spectra in the wheat fields through the field experiment analysis. The results show that the difference of fertility has the greatest influence on the wheat infrared reflection and the different ridge types make the spectra-reflecting rate of relay intercropping in the wheat fields lower than the flat furrow ploughing wheat fields.

Introduction
To monitor and estimate the yield of the crops by the way of remote sensing information, the main basis is the characteristic of the reflecting spectra of the crops in the fields. But in fact, the characteristic of the crop spectra in the fields is a kind of comprehensive reflecting result of a combined spectrum. it depends not only on the conditions of crops themselves, but on the influence of the background of the soil condition. The crops themselves may show various growth situations because of the various cultural practices so as to obtain the spectra characteristic under different conditions. To monitor and estimate the yield of crops in a large area by the way of remote sensing method, that is to utilize these different spectra reflections to understand the different growth conditions of the crops and judge the possible causes of these conditions and its influence to the yield.

At first, to understand the influence of some main cultural practices for the wheat spectra in the fields through the field remote sensing experiments must be the important components of the research to monitor and estimate the wheat yield by remote sensing. We carried

out the spectra observation of the different crop conditions due to the various cultural practices, planting density, fertilizer level and showing early or late by the way of the different productive plots in the fields and special field tests. Now the preliminary results were analyzed as follows and for the reference of remote sensing image interpretation.

Research Arrangement
The whole research program was carried out in two experimental areas from 1987 to 1988.
  1. Tongxian Country Experimental Area:

    Together with the Tongxian Agricultural Extension Station, two kinds of plots were selected for the observation. The first kind was the wheat field with two types of wheat culture and different ridge width; one with 2.5m broad ridge width and the other with 1.5m narrow ridge width. The second kind was the wheat field with different fertilizer level. Its experimental treatment was with 225kg/ha,112.5kg/has and without ammonia sulphuric as the basic fertilizer and with different dressing fertilizer during the wheat reviving stage and elongation stage. All the wheat variety was the Noida 146 and the soil texture was loam. There were two different sowing variety was the Nonda 146 and the soil texture was loan. There were two different sowing periods: one early sowing on 22-24 Sep., the other late sowing on Oct.4.

  2. Malianwa experimental Area:

    This area was near Beijing Agricultural University. Its soil and wheat variety wheat variety were similar to the of Tongxian Country experimental area. In order to gain various experimental combination data without too many treatments, the orthogonal rotary regression design was adopted there were two factors: Sowing density and fertilizer level, with five ranks. The greatest density was 4,080,000 seedlings per hectare and the smallest 1,200,000 seedlings per hectare. The different range between the ranks was 1018,500 seedlings. The greatest basic fertilizer was 300 kg/ha N and 150 kg/ha P. the smallest was without basic fertilizer. The different range between the ranks was 1018,500 seedlings per hectare and the smallest 1,200,000 seedlings per hectare. The different range between the ranks was 1018,500 seedlings. The greatest basic fertilizer was 300kg/ha N and 150kg/ha. P. The smallest was without basic fertilizer. The different range between the ranks was 106.05kg N and 57.7kg P.

  3. Observation Method:

    With the four spectrum ranges, i.e. 0.5-0.6, 0.6-0.7, 0.7-0.8, 0.8-1.1 of American 100 AX-M spectror and the four spectrum rangs, i.e. 0,45-0.52, 0.52-0.6, 0.63-0.69, 0.76-0.90 of the Chinese RS-II-type spector, the spectra were determined during the main wheat growth stages. During fine weather without wind at 10:00-14:00 noon, the spectrum surveying was carried out to make the main ray axis of lens vertical to the wheat canopy and over the canopy 1.2m. Every treatment was reassured for 9 times, to record and count with auto-microcomputer.

    During the spectrum observation, the wheat samples were carried out to take the plant dry-fresh weight, LAI and the chlorophyII content,etc.
Result analysis
  1. Fertility, density and snowing period influenced the wheat reflectance spectrum: it was well known that the amount fertilizer the height or low density of planting and the sowing period early or late might influence the growth and development of winter wheat this influence showed the various agronomical characters of the winter wheat population such as LAI dry weight fresh weight chlorophyll content etc.

    According to research known the difference of the wheat water content between different treatments was rather small but that LAI chlorophyll dry was rather great although in general the tendency of these three cultural factors of winter wheat population growth function was the similar but the function course handsome differences the level of fertility influenced the wheat root system development tillering ability and loaf size but planting density might regulate the relationship between the individual and population growth the density increase might raise might the total biomass.

    The different cultural practices made the difference of the above agronomic character of the wheat population and that certainly reflected the reflectance spectrum over the canoy. From the spectral curves of the treatments in fig 1 there was surely the difference of the spectrum of the canopy and this difference was showed at the different observation during the wheat chlorenchyma growth stage. The common spectrum different characters made by different cultural practices were as follows.

    The reflectance of the I and II wave bands of the winter wheat with fertility high density and appropriate sowing period were lower than those of low level and late sowing period but those of III and IV wave bands were higher than those of corresponding low level treatment or late sowing period the different points were two spectral reflectance differences of the two levels among the three treatments in the infrared band wave were greater especial attention point was that during the spectral comparison of several cultural practice treatments the margin made by different fertility level in the near infrared band wave showed the greatest. As the reflectance on 3 May the reflectance of the winter wheat with high fertility was higher than that with high ferity but this high density was only r\times in 1988 because of the freeze injury of the wheat seedlings during the over wintering stage the above difference was not obvious in the fertility experimental fields but there appeared some similar tendencies among some individual data of the observational days the visible spectral reflectance of the high fertility field was lower and the near infrared spectral reflectance was higher this shown in table 1.


    Fig. 1 Comparison of wheat spectral reflectance under the different treatment
    I=0.5-0.6 , II=0.6-0.7 , III=0.7-0.8 , Iv=0.8-1.1 ; A Fertilizer treatment,
    B Density, C Sowing time treatment.

    Table 1 Comparison of the spectral reflectance of the corresponding four TM wavebands with the application of fertilizer or without the application of fertilizer
    Waveband\
    Reflectance%\
    Treatment    		 0.45-0.52 0.52-0.60 0.63-0.69 0.76-0.90
    with fertilizer
    225 kg /ha     		3.0 4.3 4.2 26.9
    Without fertilizer 3.5 4.4 4.4 25.2

    The characteristics of the above some cultural practices to the to the reflect ant effect of the wheat spectrum have certain refer significance to interret the difference of the crop condition and its reason to the application of the remote sensing information.

  2. The influence of the difference Ridge planting to the mixed spectrum of the wheat field.

    In order to observe the influence of different ridge planting to the wheat field spectrum two types of ridge planting were selected for the spectral observation one with ridge width 2.5 m and bank width 0.3m the other with ridge width 1.5m and bank width 0.45m if we looked the field within the ridge and the bank as two kinds of ground matters then the acreage proportion of the bank bare soil of these two ridge planting within certain acreage for example within one pixel range of TM image was different it was the image that the proportion of the wheat covered field and bare soil within pixel was different then there the proportion the mixed spectrum do that the luminance of the pixel might change then that will influenced the interpretation to the wheat field.

    in order to seek the size and characteristics of above influence during the observations the reflectance spectrum of the wheat field the canopy and bank bare soil was measured respectively .According to the proportion between them and the respective spectrum their combined might be extracted from following formula.

    R = A1R1+A2R2+..........+AnRn / A1+A2+................+An

    Where R is the combined spectral reflectance Ai(i=1,2,.....n) is the acreage of the ground matters within S Ai of certain acreage are the spectral reflectance of the ground matter in Ai.

    During the observation there were only two ground matters the wheat covered field within the ridge and bank bare soil comparison of the spectrum in the different ridge planting forms was extracted with in certain as shown in Table 2 and table 3.

    From the above tables the general tendency was that mixed spectral reflectance of I and II wave bands was higher than spectral reflectance of the wheat covered within the ridge but that of III and Iv wave bands turned out contrary to the above reflectance and its mixed spectral reflectance was lower than that of the wheat covered field within the ridge among which that of IV wave band was the greatest the narrower the wheat ridge the greater its difference according to the observation data 20 Apr the difference of the wide ridge was and the difference of the narrow ridge was 7%,

    The above difference mainly was because of the different reflectance between the bank bare soil and wheat covered field on 28 April the difference of the average value between bank spectrum and wheat field spectrum was obvious as shown in table 4.

    The change of the mixed spectrum varied with the growth of wheat during the seedling stage the wheat-covered rate within the ridge was low so the spectral difference was field and bund was the smallest.

    Table2. Comparison of the wheat covered field within the ridge and the wheat field mixed spectral reflectance(ridge widht 2.5m, conversion acreage 100m2
    Waveband I II II IV
    Observation
    date
    \Type

    Reflectance(%)
    		 Wheat Covered
    field    		 Mixed Wheat Covered
    field    		 Mixed Wheat covered
    field    		 Mixed Wheat covered
    field    		 Mixed Proportion of the
    ridge and bund
    bare soil    		 The difference
    of IV waveband
    between the wheat
    covered field and
    mixed spectrum    		 LAI
    14 Mar.
    9 Apr.
    28 Apr.
    12 May.
    27 May.    		   9.3
    4.9
    3.4
    4.2
    3.4    		 9.6
    5.9
    3.5
    4.1
    3.2    		 12.9
    5.1
    2.6
    3.5
    3.0    		 13.2
    6.7
    2.8
    3.5
    3.0    		 19.4
    26.4
    32.2
    38.1
    27.6    		 19.4
    25.4
    31.2
    36.9
    26.4    		 25.0
    35.6
    46.7
    56.4
    41.9    		 24.3
    33.7
    44.8
    54.7
    40.1    		 16.80
    14.52
    8.32
    7.2
    11.6    		 0.7
    1.9
    1.9
    1.7
    1.8    		 0.15
    1.77
    5.52
    5.12
    4.97


    Table3Comparison of the wheat coered field within the ridge and the wheat field mixed spectral reflectance(ridge width 1.5m,conversion acreage 100m
    Waveband I II II IV
    Observation
    date
    \Type

    Reflectance(%)
    		 Wheat
    field    		 Mixed Wheat
    field    		 Mixed Wheat
    field    		 Mixed Wheat
    field    		 Mixed Proportion of the
    ridge and bund
    bare soil    		 The difference
    of IV waveband
    between the wheat
    covered field and
    mixed spectrum    		 LAI
    14 Mar.
    28 Apr.
    12 May.
    27 May.    		   8.3
    4.4
    4.5
    3.7    		 9.0
    4.6
    4.4
    6.0    		 11.1
    3.3
    3.6
    3.1    		 12.4
    4.0
    3.7
    5.4    		 18.8
    37.7
    37.5
    39.6    		 18.5
    33.6
    34.2
    38.1    		 23.9
    53.9
    55.3
    43.6    		 22.7
    46.9
    48.5
    38.3    		 31.9
    21.2
    21.0
    26.6    		 1.0
    7.0
    6.8
    5.3    		 0.3
    6.64
    5.65
    4.91

    Special attention must be paid to the spectral difference that was caused by the different ridges during the interoperation of the satellite image in the interpellation area.

  3. The analyzing results have the practical significance for the interpretation of the wheat remote sensing the comparison between the wheat bund spectrum and the spectrum within the ridge on 28 April.

    Table 4 The comparison between the wheat bund spectrum and the spectrum within the ridge on 28 April (ridge widht 1.5m,width 0.45m)
    Waveband I II III IV
    Reflectnce(%)\type        
    Bank(average vlaue) 4.1 4.6 20.7 23.5
    Wheat field
    (average value)    		 3.4 2.6 32.2 46.7

    Difference has the greatest significance to the infrared reflectance of the winter wheat about twice as that of the density and sowing period measures when the wheat field imagine spot with close colors but rather great difference of saturation degree is shown on the satellite image or there are different infrared reflectance's on these two wheat fields In general it is caused but the fertility factor but when image is in the duration shortly after the emergence of wheat seedlings the colors difference may be considered as caused by the sowing period shown in the following figure from it we see the wheat fields with pale red are shown rather lately.

    The different ridges make the spectral reflectance of the near infrared wave band of the interpolating wheat field lower than that of the flat plaguing wheat field this difference is not obvious during the seedlings stage when the wheat acreage is extracted and the growth conditions are analyzed on various ridge planting areas with the satellite image it is wheat growth vigour is analyzed with the picture with different time image when the notes because the inter planting field may make it reduce .


    Figure 2. image (Tongxian) of 31 October 1986