Page 110 - Contributed Paper Session (CPS) - Volume 7
P. 110

CPS2042 Anna Christine D. et al.
               Figure 2: Average Yield per Pilot Area in tons/ha















               The direct estimation of total production (kg) and the average  yield (kg/m )
                                                                                         2
               of harvested paddy are  derived from crop-cutting data for the sample of 2.5
               m x 2.5 m (i.e., 6.25 m )  subplots by multiplying  data on harvested  paddy in
                                     2
               each  subplot by the corresponding subplot weights used for the estimation
               of total area.  Table 7 shows relatively high CVs for the direct estimate of the
               total production of rice,  resulting in corresponding wide confidence intervals.
               The main source of this discrepancy is due to  the variability  in the estimate
               of total area planted in rice paddy.
               Table 6: Direct estimates of total production of rice paddy in kg

                              Production,               95% confidence interval     No. of
                Domain/Stratum             SE     CV                        DEFF
                                 kg                                              observations
                                                         Lower     Upper
                Savannakhet    480,210,025   70,689,811   0.147   339,357,502   621,062,547   27.78    136
                  IRRI+GlobCover   232,891,720   28,133,072   0.121   176,835,350   288,948,089    1.27    105
                  IRRI         8,612,733    2,300,195   0.267    4,029,494    13,195,972    0.16    16
                  GlobCover    133,816,763   40,578,604   0.303    52,962,128   214,671,397    4.27    13
                  Other      104,888,810   50,533,806   0.482    4,198,002   205,579,617   11.01    2
                Ang Thong     106,137,971   13,871,631   0.131    77,915,925   134,360,017   15.63    103
                  IRRI+GlobCover    98,753,252   13,644,497   0.138    70,993,313   126,513,191   10.31    81
                  IRRI          772,357    319,637   0.414    122,050    1,422,663    0.14    8
                  GlobCover    6,612,362    2,479,451   0.375    1,567,882    11,656,842    0.78    14
                Thai Binh    260,670,296   27,265,098   0.105   206,150,363   315,190,230   72.01    253
                  IRRI+GlobCover   244,447,928   26,764,357   0.109   190,929,289   297,966,567   29.19    219
                  IRRI           43,599    2,245   0.051    39,111    48,087    0        8
                  GlobCover    16,178,769    5,201,422   0.321    5,777,883    26,579,656    1.57    26
               Direct estimates of the total area planted in rice from the sample plots have a
               relatively high CVs because of the variability in the sizes of the sample plots.
               To produce mesh level estimates, which may improve the precision of the
               estimates for total area planted in rice, additional information to delineate rice
               area  planted  within  each  mesh  were  recorded  on  the  paper  maps,  and
               subsequently digitized using the ALIS  methodology.
                                                   10


               10  The methodology provides for strict digitization guidelines to identify crop area planted.
                                                                   97 | I S I   W S C   2 0 1 9
   105   106   107   108   109   110   111   112   113   114   115