Monitoring and predicting crop growth
and analysing agricultural ecosystems by remote sensing
Tsuyoshi Akiyama
1
and Y. InoueNational InstituteofAgro-EnvironmenlalSciences, 3—l—l, Kannondai, Tsukuba,Ibaraki305,Japan M. Shibayama
National Grassland ResearchInstitute.Japan Y.Awaya andN.Tanaka
Forestry and Forest Products Research Institute, Japan
LANDSAT/TM data, which are characterized by high spectral/spatial resolutions,are able tocon- tribute topractical agricultural management. In the first part of the paper, the authors review some recent applications of satellite remote sensing in agriculture. Techniques for crop discrimination and mapping have made such rapid progress that we can classify crop types with more than 80%accuracy. The estimation of crop biomass usingsatellitedata,including leaf area,dryand fresh weights, and the predictionofgrain yield, has been attempted using variousspectral vegetation indi- ces. Plant stresses causedby nutrientdeficiencyand water deficit have also been analysed successfully.
Such information may be useful for farm management. Inthe latter half of the paper, we introduce the Arctic ScienceProject,whichwascarried out under the Science andTechnology AgencyofJapan collaborating with Finnish scientists.In thisproject, monitoring of the boreal forest wascarried out usingLANDSATdata. Changes inthe phenology of subarctic ground vegetation, basedonspectral properties, weremeasuredbyaboom-mounted,four-bandspectroradiometer.Theturning pointdates of the seasonal near-infrared (NIR) and red (R) reflectance factors might indicate the end ofgrowth and thebeginning of autumnal tints,respectively.
Key words: arctic area, biomass, environment,phenology, spectroradiometer, vegetation, yield 'Currentaddress: Institutefor Basin EcosystemStudies, Gifu University, Yanagido, Gifu 501-11,
Japan, e-mail:akiyama®green.gifu-u.ac.jp
©Agricultural and Food ScienceinFinland ManuscriptreceivedFebruary 1996
Vol. 5(1996): 367-376.
Table 1.Characteristics of earthobservingsatellites.
Satellite/ Launched/ Wavelength Ground
sensor yr/revolution range nm Band resolution
NOAA/ U.S.A Cl:580-680 Visible I.lkm
AVHRR (1978-) C2;725-1100 Near-1R 1.1 km
1/2 d C3:3550-3930 Mid-IR 1.1 km
C4: 11.5-12.5
n
Thermal 1.1 kmLANDSAT/ U.S.A. MSS4:SOO-600 Visible 80m
MSS (1972-) MSSS:6OO-700 Visible 80m
16(18) d MSS6:7OO-800 Near-IR 80m
MSS7:BOO-1100 Near-IR 80m
LANDSAT/ U.S.A. TM 1:450-520 Visible 30m
TM (1984-) TM2:520-600 Visible 30m
16d TM3:630-690 Visible 30m
TM4: 760- 900 Near-IR 30m
TM5:1550-1750 Mid-IR 30m
TM7:2080-2350 Mid-IR 30m
TM6:10.4-12.5n Thermal 120m
SPOT/ France Bl: 500-590 Visible 20m
HRV (1986-) B2:610-680 Visible 20m
26d 83:740-900 Near-IR 20m
B4:510-730 Panchromatic 10m
IRS-IA/ India Bl: 450-520 Visible 72.5m
LISS-I (1988-) 62:520-590 Visible 72.5m
22d 63:620-680 Visible 72.5m
84:770-860 Near-IR 72.5m
ERS-1/ ESA C: 5.3GHz Microwave 30m
SAR (1991-)
35d
JERS-1/ Japan Bl: 520-600 Visible 18m
OPS (1992-) 62:630-690 Visible 18m
44d 63:760-860 Near-IR 18m
SAR L: 1.275GHz Microwave 18m
Monitoring of crops and environment by earth observation satellites
Characteristics of satellite sensors and
vegetation indices
Satellitesensors
More than two decades have passed since LANDSAT-1 was launched by NASA in 1972.
Atpresent, several kinds of satellitesare orbit- ting and observing the surface of the earth(Ta- ble 1). By the end of thiscentury, stillmore,new
satellites with special missions will be launched.
In the NOAA series,two satellitesaremain- tained in polar orbit, onein a morning orbit and the other in an afternoon orbit. They provide a wide range ofdata, including sea surface tem- perature,cloudcover,data for land surface stud- ies, temperatureand humidity profiles, andozone concentrations.
The LANDSAT and SPOT satellites provide high resolution imagery in the range of visible and infrared bands. They are used extensively for high resolution land surface studies.
The ERS seriesconcentrates on global and regional environmentalissues,makinguseofac-
live microwave techniques, which enablearange of measurements to be made of the land, sea, and ice surface independent of cloudcover.
The aim of JERS-1 is to observe the earth using optical sensors andahigh resolution syn- theticapertureradar. Land surveys and monitor- ing of various resources are the main applica- tions of this satellite.
Spectral vegetation indices
Basedon the fact that green vegetation absorbs red (R) wavelengths but reflects near-infrared (NIR)wavelengths, several spectral vegetation indices(VI) derived from satellite-borne data have been proposed.VI, including ratiovegeta- tion index(RVI, NIR/R), difference vegetation index(DVI, NIR-R),normalized vegetation in- dex(NDVI, (NIR-R) /(NIR+R)), are used the mostby vegetation scientists. In addition,per- pendicular vegetation index(PVI), soil adjusted vegetation index(SAVI),and K values have been proposed to eliminate background soil effects from vegetation.
Crop monitoring from space
In recentdecades, many attempts have been madeatextracting agro-environmental informa- tionatthe regional level,and someuseful tech- nologies have been developed. As a result, the following informationcanbe obtained by space- bornesensors with high spectral/spatial resolu- tions, suchasLANDSAT /TM and SPOT/HRV:
1)crop inventory and planting acreageestimates, 2) leafarea and phytomass estimates and yield prediction, 3) cropstressdetection,and4)agro- environmental survey includingsoil,vegetation, waterand atmosphere.
Crop discrimination andplanting acreage estimation
From 1974onwards,in the Large Area Crop In-
ventoryExperiment(LACIE),jointly undertak- enbyNASA,NOAA andUSDA,satelliteremote sensing technology was applied, on an experi- mental basis, to forecast harvests in the major
wheatproducing areasof the world using LAND- SAT/MSS data. As a result, a 1977 real-time fore- castof the wheat production of the Soviet Un- ion indicated that thesystemcouldoperate and could be applied toother areasand other crops (MacDonald and Hall 1980).This project was followed by AgRISTARS, which focusedon the assessment of crop conditions from space (AgRISTARS 1983).Many attempts atcrop dis- crimination and planting acreage estimation have been made world-wide for a variety of crops using LANDSAT/MSS, LANDSAT /TM and SPOT/HRV data.
Recent studies and classification accuracies aresummarized in Table 2. Most of the studies report accuracies exceeding 80%. Toillustrate, Table3 showscropdiscrimination results using LANDSAT/TM data in the Tokachi district of Japan.Here,the major cropswereclassified with 90% accuracy (Fukuharaetal. 1988).Ingener- al,the accuracy ofcrop discrimination is strongly dependent on the spectral/spatial resolution of the satellitesensor,the timely acquisition of data atsuitable crop stages, and ample field size for detection purposes.
Biomass,
leaf
areaand yield estimation Several spectral vegetation indices (VI) have been proposedto estimate vegetational informa- tion,includingbiomass,yield and leafarea. RVI, which employs the ratio between NIR and R,is often applied for the estimation of aboveground dry weight, fresh weight and leafarea.Estimates ofpasture grass yield at the first cut were madeover Tochigi prefecture in cen- tralJapan, using LANDSAT/MSS data collect- edon May 22, 1979. The analysiswas conduct- ed by attempting firsta land-use classification
to identify pasture and meadows. Secondly, a multiple regression analysis was performed to estimate the yield of the first cutting in each pas- ture plot. To design the regression model, 26 plots with known yield datawere prepared in advance. A high correlationwas observed be- tween the actual forage yieldatthe first cutting and the estimated yield based on LANDSAT/
MSS using the four bands (Akiyamaetal. 1985).
Vol.5(1996): 367-376.
Table2.Results of crop discrimination studies using satellite data.
Country/ Satellite/ Main crops Accuracy Authors
District Sensor
U.S.A./ Landsat/ Com, Soybean 83.9% Batista etal, 1985
Combelt MSS
U.S.A./ Landsat/ Alfalfa, Com, Oats 89.8% Loetal. 1986
Wisconsin MSS
Hungary Landsat/ Com, Alfalfa 90% Csillag 1986
MSS Sunflower etc.
Argentine/ Landsat/ Com,Soybean 80% Badhwar et al. 1987
Buenos Ires MSS6TM Sorghametc.
U.S.A./ Landsat/ 3 Vegetables >90% Williams et al. 1987
NYState TM 4 Crops >75%
Japan/ Landsat/ 7 Crops 90.1% Fukuhara et al. 1988
Tokachi TM
Hungary/ SPOT/XS Wheat, Soya XS:85-88% Biittner et al. 1989
Kiskore Landsat/TM Alfalfa TM: 87-91%
U.K./ SPOT/ 10 Vegetation types 71% Jewell 1989
EastAnglia HRV 4 Vegetation types 88%
Table3.Results of crop classification studies using LANDSAT/TM datain Tokachidistrict,Japan(Fuku- hara etal. 1988).
Performance(%)
Cropname Su Po Ad So Co Wh Pa Fo Others
Sugarbeet (Su) 96.3 0 0 0 0 0 1.7 0 2.0
Potato (Po) 0 98.5 0 0 0 0 0.5 0 1.0
Adzuki bean(Ad) 0 0 2L2 0.5 1.6 0 0 0 0.6
Soybean(So) 0 0 0 100.0 0 0 0 0 0
Com (Co) 0 0 1.2 0 14.4 0 22.5 2.4
Wheat (Wh) 0 0 0 0 3.5 95J. 0 0.2 1.2
Pasture (Pa) 0 0 1.0 0 0 0 88X) 0 11.0
Forest (Fo) 0 0 1.0 0 6.1 0.6 0.5 902 1.6
Underlined numbers indicate theproportionof cropscorrectlyclassified(%).
Recent results of grain yield prediction for rice and wheat arepresented in Table 4. A high accuracy was attained witha multiplere- gression model using2 or 3 TM bands for rice.
Meteorological damage caused by cold wind and floodingare also analysed effectively usingsat- ellite data. For example, rice damage caused by floodingwas highly correlated with the turbidi-
ty of water (Yamagata and Akiyama 1988), which reflected in TM bands 2 and 3.
Cropstressdetection
Plants experience various kinds ofstresses dur- ing growth, includingstresses caused by atmos- phericfactors, by rhizospheric factors, orbiotic stressescaused byavariety of organisms. More- over, in recentyears, it has often beenreported that toxic materials in the environmentresulting from human activities can inflict damage on crops and forests, asin the case of acid rain. If farmers could detect various cropstressesin the
Table4.Results of cropyield prediction usingsatellite data.
Crops Country/ Satellite/ Date of data Model used Correlation, Authors
district sensor acquisition Standard deviation
Rice Thailand/ LANDSAT/ 09/Dec/88 MRM(TM 1,5,7) R2=0.85to0.92 Tennakoon et al. 1992 Saraburi TM 27/Dec/88 Growthmodel
Rice India/ IRS-IA/ 12,13/Oct/ AreaWeighted SD;-2%t0+14% Patel et al. 1991
Orissa LISS-1 88 AverageRatio
Rice Japan/ LANDSAT/ 23/Sep/86 MRM(TM2,3,4,5 R2=0.953 Mubektietal. 1991
Ishikari TM &TM4/TM3)
Coldwind
damaged Japan/ LANDSAT/ 19/Sep/80 MRM(MSSS,6) R=0.91 Miyamaetal. 1983 rice Ishikari MSS
Flood
damaged Japan/ LANDSAT/ 6/Aug/86 MRM(TM2,3) R=0.972 Yamagataand Akiyama 1988 rice Ibaraki TM
Wheat Brazil/ LANDSAT/ 24/Jun/86 Meteo. model R2=0.65 Rudorff and Batista 1991 SaoPaoloTM 27/Jun/87 TM4/TM3
Wheat Japan/ LANDSAT/ 27/Jun/86 MRM(TM2,3) R2=0.66to0.79 Shiga 1993
Ishikari TM 29/May/90
Wheat India/ LANDSAT 22/Feb/86 NDVI, RVI Variance Singhetal. 1992
Sultanpur TM Layersmethod 0.1958
Wheat India/ LANDSAT/
Haryana TM 26/Jan/89 Linearyield-spectral Deviation+l4% Sharmaetal. 1993
IRS/IA/ to index to-18%
LISS-1 18/Feb/89 MRM:Multiple RegressionModel
R;Multiple regression coefficient R 2:Coefficientof determination SD: Standarddeviation
early stages ofan infectionor a nutrient defi- ciency before any symptoms appear, a timely
intervention couldprevent the damage to the yield and savethe quality of the final product.
Water deficit in plants canbe clearly detect- ed by NIR and mid-infrared(MIR) wavelength reflectances, which indicate the thermal chang- escaused by stomatalmovement.Thesameband can be used for the crop stress caused by soil- borne diseases, as it enablesto detect the wilt- ing of the leaves (Torigoe 1992).
Stresses due to meteorological factors do not always show up early. Spikelet sterility is often caused by low temperatures during the flowering stage of rice in the northernpart of Japan. Usually, the fertile ears die offatripen- ing, while the sterile earsremain greenat har-
vesttime. Using LANDSAT /MSS imagery, Mi-
yamaetal.(1983) were able to create a map of the cold damagetoriceon the Ishikari plane of northen Japan.
Agro-environmental monitoring
Soilsurvey
Many scientists have classified soiltypes using satellite data. Informationonorganicmatterand moisture content in the soil is particularly important for farmers. Fukuhara etal. (1980) wereabletodetermine differences in the mois- ture contentsof the soil using LANDSAT/MSS data in the Tokachi district in Japan. In the same area,Hatanakaet al.(1989) classified the organ- icmatter content in the soil, using band 3 re- flectance data from LANDSAT/TM obtained in Vol.5(1996):367-376.
May, when the surface of the upland field was almost bare. The classification resultswereplot- tedon a 1/50,000scale map, and used by offic- ers of the agricultural extension services.
Land evaluation
Itis possible for the developing countries tose- lect the land suitable for agricultural develop- mentusing satellite data. These datacanprovide several thematic maps relating biomass, soil moisture and present landuse. When these the- matic mapsareoverlaidon existing meshedtop- ographicorgeographic maps,we can obtain val- ue-added informationonthe productivity of the land (Akiyamaetal. 1987).
Inrecentyears, the needtoquantify the func- tion of agricultural land has become increasing- ly important in Japan. Attempts have been made
toquantify the multiple functions of such land, for example the impact of the cultivation ofpad- dy riceon soil conservation, landscape mainte- nance or the capacity for water purification.
Remote sensing by a satellite is a prominent tool for such evaluations.
Arctic Science Project with Finland
International Cooperative Study on Observation of Variabilities in the Arctic Atmosphere, Hydrosphere and Biosphere,
and their Interactions
One ofour mostserious concerns today is the fact that global environment change (chiefly deterioration) is taking place atanunprecedented speed, because of ever-growing human activity.
We have come torealize that polar regions areveryimportantareasfor understanding proc- esses of global change; no less so than the tropical and mid-latitude areas, where most of the humanracelive.
The ongoing environmental changesare not
only global inscale, but alsovaried, and involve diverse and complicated processes, calling for an interdisciplinary approach toaddress the is- sue. In thiscontext, in 1990 the Science and Technology Agency of Japan organizeda5-year program "International Cooperative Study on Observation ofVariabilities in the Arctic Atmos- phere, Hydrosphere and Biosphere, and their Interactions" in cooperation with 15 governmen- tal institutes, laboratories and universities.
The Arctic Science Project consists of three sections, namely, 1)Arctic oceanography and glaciology, 2) Atmospheric chemistry, 3)Arctic vegetation and agricultural environment. The Arctic vegetation team enrolled the assistance of the Technical Research Centre of Finland (VTT). Turku University and the Agricultural Research Centre of Finland. Here, we report on two topics investigated by the vegetation moni- toring sub-teams in the Arctic Science Project.
Vegetation change monitoring in the
boreal area
Thispart of the projectwasconducted by scien- tists in the Forestry and Forest Products Research Institute of Japan working together withVTTof Finland. An outline of the collaborative research is given below.
The study concentrated on the monitoring of vegetation change using satellite data. Seasonal and successional changes ofspectra, as a basis for long termmonitoring, were studiedoverthe southern boreal forest zone using LANDSAT/
TM images (Awayaetal. 1995).
Two monitoring methodswere comparedat a test site inKevo, at the northern tip of Fin- land. The first was boundary detection using Laplacian filtering, and the second biomass change detection by subtraction between two normalized difference vegetation indices(NDVI) ofLANDS AT/MSS images acquired in 1972 and 1987. A zonal shift of arctic vegetation wasnot
captured using these methods, but damage to birch forests by the moth caterpillar were clearly detected. This phenomenon appeared to
Vol. 5(1996): 367-376.
be caused by a combination of the mothcater- pillar and by temperatures lower than the aver- age in the latter half of the 1960’5. It suggested that the vegetation changesareoccurring thatare related both ecological processes, i.e. the dam- age and the recovery of thetrees,and the effects of global climate change.
The seasonal spectral patterns of the south- ernboreal forests were clearly identified in the TM3, TM4 and TMS ofLANDSAT in Hokkai- do, Japan. Though the spectra of the young spruce stands seemedto be affected by mixed broadleaved trees in TM3 and TM4, seasonal spectral patterns could be classified into two types, evergreen and deciduous trees, in TMS (Fig. 1, left). Foresttypesclassified accordingto the major tree showed constant intensities in a normalized difference ratio of TMS and TM3 (NDS3) over the growing season(Fig.l, right).
The successional spectral patterns of the spruce stands varied between seasons and channels.
However, the relationship between the digital number and the stand agewas expressed using an exponential function (not shownhere).
Spectral properties of subarctic ground vegetation
This experimentwasconducted by the National
Institute of Agro-Environmental Sciences, and the National Grassland Research Institute of Ja- pancollaborating with VTT of Finland and Turku University (Shibayamaetal. 1995).
Ground truth information forremote sensing, toassess the phenology of boreal plants, is need- ed in global scale climate change research ac- tivities. Aboom-mounted, four-band spectrora-
diometerwas installedto measure seasonal ra- diances in the green (520-600nm), red (630- 690nm), near-infrared(765-900 nm) and mid- infrared (1570-1730nm) spectral bands from boreal shrub canopies overthe growing season in northernmost FinlandatUtsjoki. It measured the spectraof four fixed ground plots, eachone once perhour, from early Juneto mid-Septem- ber between 0600 and 1800 hrs local time. A reflectance panelwasalso measuredafew min- utesbefore and aftermeasurement of each plot.
The plant phenology on each plot was also ob- served weekly during the experiment. Seasonal reflectance factors for each plot in each band werecalculated basedon a new calibration meth- od. It involves a correction for the degradation of the reference panel. A hand-held spectroradi-
ometerwasalso usedtomeasurethe plantcano- pies and the reference panel in the earlyautumn.
The turning point dates of the seasonal near-in- frared (765-900nm) and red (630-690 nm) re- flectance factors might indicate the end of Fig. 1 Seasonal changesof spectra for different the tree speciesestimatedusing TM data(Awayaetal.
1995). Thespectral changes,fromearly spring to lateautumn,weredrawn using eight TM imagestaken between 1985and 1993.The numbersinthe figure captionrepresent the year ofplanting of the trees.
Fig. 2.Seasonal patterns ofdailyreflectance factors at bands560, 658, 833and 1649nm,and the bi-band ratio of the560 and658nm reflectance factors measured for theplant plots (Plots#l,#2)bythe automated four-bandspectroradiometerat Kevo. The lateral bars show the95%confidence intervals of the estimated intersectionpoints(the turning pointdates of the radiometric variables)(Shibayamaetal. 1995).
Vol.5(1996): 367-376.
growth and the beginning of the autumnaltints, respectively. The ratio of the green(520-600nm) and red band reflectance factors, however, seemedtobemore accurate in predicting these turning points (Fig. 2).
Acknowledgements.We wish to express ourgratitude to Dr.T. Häme, Dr. A. SalliandDr. A.Lohi of the Technical
Research Centreof Finland andDr,M.Alanen and Dr.S.
Neuvonen of the Kevo Subarctic ResearchInstitute, Uni- versityofTurku, for their diligentsupport in the Arctic ScienceProject.The authors arealsodeeply grateful to ProfessorT.Mela for hishelpful suggestions.
References
AgRISTARS Program Management Group1983, Agri- culture and resources inventorysurveys throughaero- space remote sensing. Research Report, Fiscal Year
1982,APJ2-0393.p. 1-45.
Akiyama,!., Miyama,K.,Soemarman, H.&Setlyono, J. 1987.Land evaluation system using Landsat datator agriculturaldevelopment.Application ofPATTERNmeth- odin North Sumatera. Journal of the JapaneseAgricul- tural Systems Society 3(2):74-89.
-,Yasuda,Y., Emori,Y.& Miyama,K. 1985. Grassland diagnosis by remote sensing techniques,2.First cutting yield estimate of pasture using Landsat multispectral data. The Journal of JapaneseSociety of Grassland Sci- ence 31: 97-103.(in Japanese)
Awaya,Y., Tanaka, N., Häme,T.& Lohi,A. 1995.Veg- etationalchangemonitoringinthe boreal areausingre- mote sensing. Changesinthe northern Finland and spec- tral characteristicsofspruce. Proceedingsof International Arctic Science Symposium,Tsukuba, D:30-61.
BadhwarG.D., Gargantini, C.E. &Redondo, F.V. 1987.
Landsat classification of Argentinasummer crops. Re- mote Sensingof Environment7: 265-281.
Batista,G.T.,Hixson,M.M.&Bauer,M.E. 1985. LAND- SAT MSScropclassification performanceas afunction ofscene characteristics.International Journal of Remote Sensing6: 1521-1533.
Buttner, G.& Csillag, F. 1989.Comparative studyof cropand soilmapping using multitemporaland multispec- tral SPOT and Landsat Thematic Mapperdata. Remote Sensingof Environment29: 241-249.
Csillag,F. 1986.Comparisonofsomeclassificationmeth- odson atest-site. International Journal ofRemoteSens- ing7: 1705-1715.
Fukuhara, M.,Amano.T.&Miyaji,N. 1988. Creation of cropping map using LandsatTM data. Proceedings of 1988Annual Meeting of the Japan Society of Photogram- metryand RemoteSensing,J-2:169-172. (in Japanese)
Hayashi, S.Yasuda,Y. Emori. Y.& lisaka,J. 1980.
Soilmoisture analysis for soil mapping. Proceedings of the Machine Processing of Remotely Sensed Data Sym- posium,Purdue University, Indiana, US.A,p. 1-10.
Hatanaka, T.,Shiozaki, H., Fukuhara, M.,Miyaji, N.&
Saito, G. 1989.Estimation oforganicmatter contentsof uplandsoils with LandsatTMdata. Japanese Journal of Soil Scienceand Plant Nutrition 69: 426-431.(in Japa- nese)
Jewell, N. 1989. Anevaluation of multi-date SPOT data for agriculture and land usemapping inthe United King- dom. International Journal of RemoteSensing 10: 939- 951.
Lo,T.H.C., Scarpace,F.L. &Lillesand,T.M. 1986. Use of multitemporal spectral profilesinagriculturalland-cover classification. Photogrammetric Engineering and Remote Sensing52: 535-544.
MacDonald,R.B. & Hall,F.G. 1980. Global crop fore- casting. Science 208: 670-679.
Miyama, K.Sato, H.Yasuda,Y. &Emori,Y. 1983. An applied study on aremote sensing techniques to survey agriculturalland.Survey techniqueforgeographicaldis- tribution of damage to rice from cold weather usingLand- sat MSSdata. Transactions of JSIDRE 105: 27-35. (in Japanese)
Mubekti, Miyama,K. &Ogawa, S.1991. Studyon rice yielddistribution by usingLandsat TM data. TheHokkai- do National Agricultural Experiment Station, Rural De- velopmentResearch 5:101-113.
Patel, N.K., Ravi, N.,Navalgund,R.R., Dash,R.N., Das,
K.C. &Patnaik, S. 1991.Estimation of rice yield using
IRS-1 Adigitaldata in coastaltractof Orissa. International Journal of Remote Sensing 12: 2259-2266.
Rudorff, B.F.T. &Batista, G.T. 1991.Wheat yield esti- mation at the farm level usingTMLandsat and agrome- teorologicaldata. International Journal of RemoteSens- ing12: 2477-2484.
Sharma,T., Sudha, K.S. Ravi, N., Navalgund, R.R., Tomar, K.P. Chakravarty,N.V.K.&Das, D.K. 1993.Pro- cedures for wheat yield prediction using Landsat MSS and IRS-1Adata. Intrernational Journalof Remote Sens- ing14:2509-2518.
Shibayama, M., Salli, A., Hame, T., Iso-livari, L., Morinaga, S.,Inoue, Y. & Akiyama, T. 1995.Spectral properties of subarctic ground vegetation during the growth period. -Preliminaryresults ofa 1994experiment inthe northernmost Finland. Proceedings of the Interna- tional Arctic Science Symposium,Tsukuba,Japan,D: 6- 29.
Shiga,H.1993.Agriculturalinformation systeminHokkai- do: Integrationof arable land informationand applica- tion of remote sensing for land evaluation. Journal of the JapaneseAgricultural Systems Society 9: 32-39. (in Japanese)
Singh, R., Goyal, R.C., Saha, S.K.& Chhikara, R.S.
1992.Use of satellite spectral data incrop yieldestima- tion surveys. International Journal of Remote Sensing
13: 2583-2592.
Tennakoon, 5.8., Murty, V.V.N. &Eiumnoh, A. 1992.
Estimation of croppedareaand grain yield of rice using remote sensing data. International Journal of Remote Sensing13: 427-439.
Torigoe,Y. 1992.Low input sustainable agricultureand information system: Developing integrated pest manage-
mentfor soilborne disease. Journal of theJapanese Ag- ricultural Systems Society8: 149-157.(in Japanese) Williams,V.L.,Philipson,W.R. &Philpot,W.D. 1987.
Identifying vegetablecrops with Landsat Thematic Map- per data. Photogrammetric Engineering and Remote Sensing53: 187-191.
Yamagata,Y.&Akiyama,T. 1988. Flooddamageanal- ysis using multi-temporalLandsat Thematic Mapper data.
International Journal of Remote Sensing9:503-514,
SELOSTUS
Maatalousekosysteemien analysointi ja sadon ennustaminen kaukokartoituksen avulla
Tsuyoshi Akiyama, Y. Inoue, M.Shibayama, Y. Awaya ja N.Tanaka
National InstituteofAgro-EnvironmentalSciences, National Grassland Research Instituteja Forestryand Forest Products ResearchInstitute,Japani
Tämän artikkelin alussa arvioidaan muutamia satel- liittikaukokartoituksen uusimpia sovelluksia maata- loudessaja lopussa esitellään japanilais-suomalainen yhteistyöhanke Arctic Science Project. LANDSAT/
TM-satelliittikuva-aineistoavoidaankäyttää maata- loustuotannon seurannassa, sillä kuvien spektri/spa- tiaalinen tarkkuusonhyvä. Erittely- jakartoitustek- niikat ovatkehittyneet niin paljon, että viljalajit voi- daan erottaa toisistaan 80 % tarkkuudella. Satelliit-
tiaineistonavullaonarvioitu sadon biomassaa,esi-
merkiksi lehtialaa sekä kuiva- ja tuorepainoa, jaen-
nustettu viljasatoa erilaisia laajaspektrisiä kasvilli- suusindeksejä käyttäen. Myösravinteiden javeden puutteen aiheuttamaa stressiäononnistuttuanalysoi-
maan, Arctic ScienceProjectissa seurattiin boreaali- sia metsäalueita LANDSAT-satelliittiaineiston avul- la. Subarktisen pohjakasvillisuuden spektriominai- suuksiin perustuvia fenologisiamuutoksia mitattiin nelikanavaisellaspektroradiometrillä. Käännekohdat vuodenaikaisissa lähes-infrapuna- ja punaheijastu- missa saattavat osoittaa kasvunpäättymistä jaruskan alkua.
