TY - JOUR
T1 - Crop water status estimation using thermography
T2 - multi-year model development using ground-based thermal images
AU - Cohen, Y.
AU - Alchanatis, V.
AU - Sela, E.
AU - Saranga, Y.
AU - Cohen, S.
AU - Meron, M.
AU - Bosak, A.
AU - Tsipris, J.
AU - Ostrovsky, V.
AU - Orolov, V.
AU - Levi, A.
AU - Brikman, R.
N1 - Publisher Copyright:
© 2014, Springer Science+Business Media New York.
PY - 2015/6/1
Y1 - 2015/6/1
N2 - Thermal crop sensing technologies have potential as tools for monitoring and mapping crop water status. To create maps of water status from thermal images, a reliable relationship between direct water status measures like leaf water potential (LWP) and thermal water status measures like temperature and crop water stress index (CWSI) should be established for different crops and for different growth stages. The objective of this study was to define the relationships for cotton between LWP and CWSI derived from high-resolution ground-based thermal images and more specifically to examine whether robust relationships exist between the two measures for different varieties, through a cotton growing season, across seasons and under different geographical areas (different climate and soils). A dataset from three cotton growing seasons and from different geographical areas was built to explore the relationship between CWSI and LWP in cotton. CWSI was calculated based on ground-based thermal images and measured dry (Tair + 5 °C) and wet references (Artificial wet reference surface—AWRS). A linear CWSI–LWP relationship was found with high coefficient of determination (R2 = 0.7). This relationship changed over the cotton growth stages and different CWSI–LWP relationships were established to the flowering, boll-filling and defoliation stages. The boll-filling relationship was found to be insensitive to a range of meteorological conditions. The flowering and the boll-filling models were initially validated using diagonal (oblique) thermal images from dates that were not used for calibration. For CWSI calculation, the average temperature of the lowest decile was used for the wet reference instead of the AWRS. The comparison between predicted and observed values of the validation sets yielded RMSE of 0.18 and 0.15 for the flowering and boll-filling stages, respectively. The successful use of the lowest decile as the wet reference enables a future application of the CWSI–LWP relationship to map LWP at a commercial field scale.
AB - Thermal crop sensing technologies have potential as tools for monitoring and mapping crop water status. To create maps of water status from thermal images, a reliable relationship between direct water status measures like leaf water potential (LWP) and thermal water status measures like temperature and crop water stress index (CWSI) should be established for different crops and for different growth stages. The objective of this study was to define the relationships for cotton between LWP and CWSI derived from high-resolution ground-based thermal images and more specifically to examine whether robust relationships exist between the two measures for different varieties, through a cotton growing season, across seasons and under different geographical areas (different climate and soils). A dataset from three cotton growing seasons and from different geographical areas was built to explore the relationship between CWSI and LWP in cotton. CWSI was calculated based on ground-based thermal images and measured dry (Tair + 5 °C) and wet references (Artificial wet reference surface—AWRS). A linear CWSI–LWP relationship was found with high coefficient of determination (R2 = 0.7). This relationship changed over the cotton growth stages and different CWSI–LWP relationships were established to the flowering, boll-filling and defoliation stages. The boll-filling relationship was found to be insensitive to a range of meteorological conditions. The flowering and the boll-filling models were initially validated using diagonal (oblique) thermal images from dates that were not used for calibration. For CWSI calculation, the average temperature of the lowest decile was used for the wet reference instead of the AWRS. The comparison between predicted and observed values of the validation sets yielded RMSE of 0.18 and 0.15 for the flowering and boll-filling stages, respectively. The successful use of the lowest decile as the wet reference enables a future application of the CWSI–LWP relationship to map LWP at a commercial field scale.
KW - CWSI
KW - LWP
KW - Precision irrigation
UR - http://www.scopus.com/inward/record.url?scp=84939975311&partnerID=8YFLogxK
U2 - 10.1007/s11119-014-9378-1
DO - 10.1007/s11119-014-9378-1
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
AN - SCOPUS:84939975311
SN - 1385-2256
VL - 16
SP - 311
EP - 329
JO - Precision Agriculture
JF - Precision Agriculture
IS - 3
ER -
