Agro-climatic suitability mapping for crop production in the Bolivian Altiplano: A case study for quinoa

Please download to get full document.

View again

of 14
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Similar Documents
Information Report



Views: 44 | Pages: 14

Extension: PDF | Download: 0

Agro-climatic suitability mapping for crop production in the Bolivian Altiplano: A case study for quinoa
  Agro-climatic suitability mapping for crop production inthe Bolivian Altiplano: A case study for quinoa Sam Geerts a, *, Dirk Raes a , Magali Garcia b , Carmen Del Castillo b , Wouter Buytaert a a K.U.Leuven University, Division of Soil and Water Management, Celestijnenlaan 200 E, B-3001 Leuven, Belgium b Universidad Mayor de San Andres, Facultad de Agronomia, La Paz, Bolivia Received 1 March 2006; received in revised form 25 August 2006; accepted 25 August 2006 Abstract An agro-climatic suitability library for crop production was generated by using climatic data sets from 20 to 33 years for 41meteorological stations in the Bolivian Altiplano. Four agro-climatic indicators for the region were obtained by validatedcalculation procedures. Thereference evapotranspiration, thelength oftherainyseason, theseverity ofintra-seasonaldryspellsandthe monthly frost risks were determined for each of the stations. To get a geographical coverage, the point data were subsequentlyentered in a GIS environment and interpolated using ordinary kriging, with or without incorporating anisotropy. The presented casestudy focuses on quinoa ( Chenopodium quinoa  Willd.), an important crop in the region that is cultivated during the short andirregular rainfall season and that is well adapted to the frequent occurrence of drought and frost. The GIS library was used to mark zoneswhere deficitirrigation couldimprovequinoaproduction. With adata query,zoneswere delimited wherethe irrigation canbeuseful to stretch the length of the growing season beyond the limits of the rainy season and/or to mitigate intra-seasonal dry spells.Determined net irrigation requirements were used to assess the vulnerability of the delineated zones. Two regions with a highvulnerability, a severe drought risk and an acceptable frost risk are the eastern region of the Altiplano and the inter-salt depressionregion in the south. Together, they account for around one-third of the Altiplano area. In 1 year out of 2, irrigation in these regionscanstronglyimprovecropproduction.TheuseofirrigationinotherregionsoftheAltiplanowillbelessbeneficialeitherbecausetheirrigation requirements are low (region around Lake Titicaca), or because the frost risk is too high (the dry west, the south-west, andthe region in between Lake Poopo and the Uyuni salt depression). Apart from the presented application, a general view on theclimatic system of the Altiplano could be deduced from the library.The proposed routine in this study yielded a straightforward method to deal with large sets of detailed climatic information andto link them with practical agricultural advice. By redefining query limits and incorporating other data, the GIS library can be usedfor impact assessments of other agricultural practices and for studying the effects of climate change and of the El Nin˜o SouthernOscillation on quinoa production in the delineated zones. # 2006 Elsevier B.V. All rights reserved. Keywords:  Quinoa; Agro-climatology; Suitability mapping; Deficit irrigation; Bolivia; Altiplano 1. Introduction The Bolivian Altiplano is a high plateau of about200,000 km 2 situated for 75% between 3600 and4300 m a.s.l. It ranges from Lake Titicaca in the northto the Uyuni salt depression (Salar de Uyuni) in thesouth and is bounded to the west and east by mountainchains(theAndeanCordillera OccidentalandOriental).Notwithstanding the extreme temperatures, a short andirregular rainfall season and unfavorable soil condi-tions, the Altiplano is avery important agricultural zone and Forest Meteorology 139 (2006) 399–412* Correspondingauthor.Tel.:+3216329754;fax:+321632 9760. E-mail address: (S. Geerts).0168-1923/$ – see front matter # 2006 Elsevier B.V. All rights reserved.doi:10.1016/j.agrformet.2006.08.018  in Bolivia. It is home to over a quarter of the ruralpopulation of the country (Vacher, 1998).The planning and management of sustainablemethods for drought mitigation and production increaserequire detailed agro-climatic information that issummarized in a conceivable way (Hoogenboom,2000; Smith, 2000). In this paper, a GIS-basedsuitability mapping for crop production in the BolivianAltiplano is performed. As a case study it focuses onquinoa production. The pseudo cereal quinoa ( Cheno- podium quinoa Willd.) is a traditional Andean crop withhigh nutritional value that can grow under unfavorablesoil and climatic conditions (Jacobsen and Mujica,2001). Quinoa is produced on 37,000 ha in the BolivianAltiplano (Barrientos and Jacobsen, 2004). It is animportant economic activity in the region.Although quinoa is a suitable crop, the average yieldover the past 10 years was only 0.6 Mg ha  1 (INE,2003).Droughts,lowtemperatures,soilsalinityandlowinput farming are the main reasons forthe relativelylowyields for rainfed quinoa. Given the scarcity of waterresources in the region, full irrigation is not an option.Deficit irrigation however could reduce the problem of droughts during the optimal sowing period and cropsensitive stages. Deficit irrigation (English, 1990;Pereira et al., 2002) aims at obtaining maximum waterproductivity and at stabilizing yields rather than atobtaining maximum yields (Zhang and Oweis, 1999).Garcia (2003) indicated that deficit irrigation wouldindeed be an option to significantly increase the yield of quinoa in the region and to stabilize it at a sustainablelevel of 65% of its maximum yield.Different meteorological indicators that determinethe region’s suitability for crop production are studiedwith a specific focus on quinoa. The following fourindicators were considered: (i) reference evapotran-spiration ( E  0 ), (ii) the length of the rainy season (LRS),(iii) intra-season dry spells and (iv) frost risk. The  E  0 expresses the evaporative demand of the atmosphereindependently of crop type, crop development andmanagement practices (Allen et al., 1998). The LRSstrongly determines the success or failure of rainfedcrops.Because rainfallintheAltiplanoislikelytooccurin delimited small episodes of rain separated by periodsof drought (Garreaud et al., 2003), a study of intra-seasonal dry spells is essential (Fox and Rockstro¨m,2000). Next to drought, frost is one of the major growthlimiting factors in the Altiplano (Carrasco et al., 1997;Hijmans, 1999; Franc¸ois et al., 1999; Jacobsen et al.,2003). The net irrigation water requirement (  I  n ), withwhich the importance of the irrigation introduction canbe assessed, is determined for quinoa in the differentlocations of the Altiplano.In this study, a procedure is elaborated to obtain aGIS library with a compressed summary of theimportant agro-climatic information for the productionof quinoa in the Bolivian Altiplano. Within this library,data queries can be performed. As an example, regionsare marked where deficit irrigation of quinoa might beconsidered to improve crop production, with frost risk as a restrictive factor and irrigation water requirementas a vulnerability index. Knowing the commonvarietiesand production systems of quinoa in the region, thispaper gives guidelines where research on, and (micro-)investment in deficit irrigation for quinoa might bevaluable. 2. Materials and methods 2.1. Derived climatic indicators Climatic data from 41 climatic stations from the 3departments of the Altiplano were used in the analysis.Daily rainfall and maximum and minimum airtemperature were obtained from Servicio Nacional deMeteorologia y Hidrologia (SENAMHI). Additionalmonthly absolute minimum and maximum temperaturedata were provided by Hijmans et al. (2003) for those S. Geerts et al./Agricultural and Forest Meteorology 139 (2006) 399–412 400 Nomenclature AI aridity indexAI is  intra-seasonal aridity index E  0  reference evapotranspirationGIS geographical information systemIDM intra-seasonal drought mitigation  I  n  net irrigation requirementsITCZ inter-tropical convergence zone K  c  crop coefficientLRS length of the rainy seasonMBE mean bias error P  precipitationPE probability of exceedanceRMSE root mean squared errorS.E. g  geographical standard errorSGS stretching of the growing season beyondthe limits of the rainy seasonSV semi-variogramTAW totally available soil water T  dew  dewpoint temperature T  max  maximum temperature T  min  minimum temperature  stations where daily air temperature data were lacking.Data records varied from 20 to 33 years for the periodfrom 1970 to 2003. Homogeneity tests (Buishand,1982) with the software package RAINBOW (Raeset al., 1996) on yearly rainfall sums and on monthlyaverage temperatures did not indicate anytrend over theyears nor showed a significant inhomogeneity in thedata records. Therefore, the complete data sets wereretained for the analysis.Theclimaticindicatorsthatcouldbederivedforeachstation are listed in Table 1. In nearly all stations meanmonthlyminimum and maximumair temperatures wereavailableandreferenceevapotranspiration( E  0 )couldbecalculated. Only in the stations where daily rainfall dataof at least 20 years were available, the length of therainy season, the intra-seasonal aridity index (AI is ),which is an indicator for intra-seasonal dry spells, andthe net irrigation requirement (  I  n ) could be determined. S. Geerts et al./Agricultural and Forest Meteorology 139 (2006) 399–412  401Table 1Geographic information of the 41 climatic stations used in this study, with indication (*) of the climatic indicators that were determinedStation Latitude [dec. deg.] Longitude [dec. deg.] Altitude [m.a.s.l.] LRS/AI is  /   I  n  E  0  FrostDepartment of La PazAyo ayo   17.08   68.00 3856  * * Calacoto   17.28   68.63 3805  * * * Calamarca   16.90   68.13 3954  * * Caquiaviri   17.02   68.60 3800  * * Charan˜a   17.58   69.43 4054  * * * Collana   16.90   68.33 3940  * * Comanche   16.95   68.92 4055  * * * Copacabana   16.13   69.07 3843  * * * Desaguadero   16.57   69.05 3803  * * El Alto   16.52   68.18 4038  * * * El Belen   16.07   68.67 3820  * * * Huarina   16.20   68.63 3825  * * Ichucota   16.17   68.37 4460  * * Isla Del Sol   16.17   69.15 4027  * * La Paz   16.47   68.12 3632  * * Patacamaya   17.25   67.92 3789  * * * Puerto Acosta   15.52   69.25 3835  * * Santiago De Machaca   17.07   69.20 3980  * * Sica Sica   17.37   67.75 3820  * * Tiawanacu   16.55   68.68 3629  * * * Viacha   16.65   68.30 3850  * * * Department of OruroAndamarca   18.77   67.50 3740  * Caracollo   17.63   67.22 3770  * * Chuquin˜a   17.80   67.45 3775  * Eucaliptus   17.60   67.52 3728  * * * Huachacalla   18.77   68.27 3740  * * Orinoca   18.97   67.25 3780  * * Oruro   17.97   67.07 3702  * * * Sajama   18.13   68.98 4220  * * Salinas De G.M.   19.63   67.68 3860  * * * Tacagua   18.88   66.78 3720  * * Department of PotosiCalcha De Lipez   21.02   67.57 3670  * * Chaqui   19.58   65.57 3550  * * Colcha K    20.73   67.67 3700  * Julaca   20.92   67.57 3665  * * * Llica   19.85   68.25 3650  * * * Potosi   19.38   65.75 4060  * * * Puna   19.78   65.50 3420  * * * Rio Mulatos   19.68   66.77 3815  * * * Tomave   20.10   66.57 3920  * Uyuni   20.47   66.83 3669  * * * Yocalla   19.38   65.92 3400  * *  Due to the absence of daily or absolute monthlyminimum air temperature, the frost risk could not bederived for three stations. 2.1.1. Reference evapotranspiration (E  0 ) The mean monthly  E  0  is estimated with the FAOPenman–Monteith equation (Allen et al., 1998). Thecalculationprocedurewasvalidated fortheAltiplanobyGarcia et al. (2003, 2004). Procedures developed byAllen et al. (1998) and evaluated by Garcia et al. (2004)to estimate missing humidity data and radiation datawere used in this study. They consist in assuming thatthe minimum air temperature is a good estimate fordewpointtemperature( T  dew )andthatsolarradiationcanbe estimated by means of the Hargreaves equation(Hargreaves and Samani, 1982) by considering thesquare root of the difference between the maximum( T  max ) and minimum ( T  min ) air temperature. An averagewind speed of 2.9 m s  1 , derived from the historicalrecords offour representative stations in the regions (LaPaz, Oruro, Charan˜a, Potosi), isused in this study.Meanmonthly  E  0  was calculated on basis of the historicalmean monthly  T  min  and  T  max  for 39 stations (Table 1).Problems using  T  min  to estimate  T  dew  in dry areaswere noted by Allen et al. (1998) and Garcia et al. (2003). In dry regions, the air is often not saturated at T  min  and hence  T  dew  is lower than  T  min . The requiredcorrection was derived by comparing measured airhumiditywithestimatedairhumidityusing T  min forfourlocations for which full climatic data sets are available(La Paz, Oruro, Charan˜a, Potosi).To derive 10-daily values of   E  0  from the meanmonthly values, the integration procedure developed byGommes (1983) was used. To assess the quality of thisprocedure, the 10-daily  E  0  calculated with daily datawas compared with the 10-daily  E  0  calculated withmonthly data for 10 stations. 2.1.2. Length of the rainy season An onset criterion was selected with the evaluationprocedure presented by Raes et al. (2004). It consists in:(i)determiningonsetdateswithacertain criterionbasedon rainfall data and (ii) subsequently simulating the soilwater balance for the initial growing period with thederived onsets as sowing dates. The relative transpira-tion during the first 30 growing days, a good indicatorfor the initial crop development, was used to evaluateeach criterion. For the simulations the soil waterbalancemodelBUDGET(Raes,2005;Raesetal.,2006)was used with the crop characteristics presented inTable 3. Geerts (2004) found that the onset of the rainy period is best presented by the date after 1 Octoberwhen accumulated rainfall over 4 days equaled 20 mmor more (adapted from Sivakumar, 1988).The cessation of the rainy period was taken as thedate after which little or no rain ( < 2 mm) occurredduring 15 consecutive days after 15 March (Stern et al.,1982). The LRS is then taken as the period between thedetermined onset and cessation date. The LRS wascalculated for 20 stations (Table 1) for each year usingthe daily rainfall records.After the calculation of the LRS for each year, afrequency analysis was performed on the determinedLRSs for each station with the software packageRAINBOW (Raes et al., 1996) to derive the LRS in anormal (50% probability of exceedance (PE)) and dryyear (80% PE). This was done by comparing the LRSdata tothetheoreticalnormal distributioncurve.Insomecases,thedatawerefirsttransformedusingalogarithmorsquareroot function andthen regressed tothe theoreticalnormal distribution curve in order to optimize the fit.Duetotheextremelylowanderraticrainfallamountsin five meteorological stations in the dry southernAltiplano, the onset criterion could not be fulfilled inmany of the years. For these locations, the LRS wasdetermined by considering the probability of having awet day ( P  >  1 mm). The 20 and 30% probability of havingawet daywas usedtoestimatethe expectedLRSin a normal and dry year. 2.1.3. Intra-seasonal aridity index (AI  is ) A good indicator for drought is the aridity index (AI)(UNESCO, 1979). The AI is calculated as the meanannual precipitation ( P ) over the mean annual  E  0 . Thisindicator was developed for global drought mapping. Inthis study, the aridity index was adapted to mapinformation on the occurrence of intra-seasonal dryspells.Foreach10-dayperiodofthe rainyseason(LRS),the rain sum was divided by the sum of the  E  0 . A 10-dayperiod was selected because it corresponds with theaverage time in which a well-watered root zone isdepletedintheabsenceofrain.Foreachyeartheaverageofthe10-dayvaluesoverthedurationoftherainyseason(LRS) was calculated. For each location, the average of the yearly averages is considered as the intra-seasonalaridity index (AI is ). The AI is  is hence given by:AI is  ¼ P n j ¼ 1 P b  j i ¼ a  j ð P i ;  j = E  0 ; i ;  j Þ  = m  j  n  (1)where n isthenumberofyearsinthemeteorologicaldatarecord,  a  j  the first 10-day period of the LRS for year  j ,  b  j thelast10-dayperiodfortheLRSofyear  j , m  j thenumber S. Geerts et al./Agricultural and Forest Meteorology 139 (2006) 399–412 402  of10-dayperiodsintheLRSforyear  j ( m  j  =  b  j  a  j  + 1),and  P i ,  j  and  E  0, i ,  j , respectively, the total rainfall and thetotal reference evapotranspiration amount for the  i th 10-day period of year  j . A frequency analysis of the yearlyAI is  values is performed per station with RAINBOW tocompose maps representing the geographical situationswith good(20%PE),moderate (50%PE) andpoor (80%PE) 10-day intra-seasonal rainfall coverage of theevapotranspiration demand. 2.1.4. Monthly frost risk  The temperatures below which frost damage in cropsoccurs are often several degrees below zero. Thetemperature thresholds differ between crop types andfor one crop type between growing stages. Threetemperature thresholds were considered:   8,   6 and  4  8 C. By knowing that temperatures at crop canopyheight are generally up to 1  8 C lower than thoserecordedatscreenheight(DuPortal,1993),themonthlyprobability of facing at least one frost event equal to orlower than   7,   5 and   3  8 C was calculated for themonths September–May for 39 meteorological stations(Table 1). The input data used were either dailyminimum temperatures or monthly absolute minimumtemperatures, depending on the availability. 2.2. GIS mapping After the computation of the four meteorologicalindicators, grid layers were generated by performing ageo-statistical analysis. A digital elevation model for theregion was available on a 30 arcsec resolution (gridcell  1 km 2 ) (CIP, 2005). Other basic data layers were available from USGS (2005) (e.g. departmental borders,lakes)orderivedfromsecondarydatabases(e.g.Coipasasalt depression (Salar de Coipasa)). The resolution of allinterpolated layers was of 30 arcsec. The GIS platformsused were Arcview 3.2 and ArcGIS 9.0. The BolivianAltiplano was in this study defined as the region with analtitude higher than 3400 m a.s.l. and lying within theBolivian borders. Regions above 4200 m a.s.l. wereexcluded from the analysis due to the uncertainty on theparameter estimations for these high altitude zones.Bothordinarykrigingwithandwithouttheincorpora-tion of anisotropy were used to interpolate the pointresults in this study. The lag-dependence (positionindependence) of the variances of the point results isrepresentedinthesemi-variogram(SV).Inordertoyielda sound interpolation for each agro-climatic indicator, asuitablekrigingmodel(ordinaryorordinaryanisotropic),a good lag distance and a good distribution for the semi-variogram(e.g.Gaussian,exponential)wereselectedinatrial-and-error procedure. Anisotropic kriging can beused in cases were the studied variable is clearly morestationary in one direction than in the perpendiculardirection. In this case, the kriging model is split up andtwo separate semi-variogram distributions are fitted.Cross-validation is used to evaluate the predictionperformances of the interpolations (Isaaks and Srivas-tava,1989).Themeanbiaserror(MBE)isusedtoverifywhether the estimations are centered on the measure-ment values. The root mean square error (RMSE)assesses how close the predictions are to their truevalue. An assessment of the geographical uncertainty,closelyrelated to the local sampling density, is providedby a pixel-wise mapping of the standard errors of thekriging model. In the final database, each data layercomes with a RMSE and MBE value for the krigingmodel and with an extra map of geographical standarderrors (S.E. g ) on the estimations. With reliableinterpolated grid layers in the GIS library, an unlimitednumber of regional queries can be performed. 2.3. Case study: regional query for the deficit irrigation of quinoa2.3.1. Query details In this study, a query is performed to delineateregions were deficit irrigation of quinoa might stronglyimprove crop production. The performed queries arelisted in Table 2. To identify the regions were irrigationis required to stretch the length of the growing seasonbeyond the limits of the rainy season (SGS), the LRSlayers were overlaid with frost risk constraints in thebeginning and at the end of thegrowing season (Queries1 and 2; Table 2). To delineate regions suitable for themitigation of dry spells, the AI is  layers were overlaidwith frost risk constraints for the critical growth stagesof quinoa (Queries 3 and 4; Table 2).Quinoa is one of the few crops that tolerate frost to ahigh extent. Depending on the duration of the frost, thequinoa variety, the phenological stage, the relativehumidity of the air and the micro-location of the fields(e.g. hill slope with lower frost risk compared tovalleys), the temperature at which frost damage occurscan differ significantly for quinoa. Although more mapswere produced for the library, the frost risk maps of   7  8 C in November,   3  8 C in January and Februaryand  5  8 C in March were used for suitability mapping(temperatures at screen height). Following Jacobsenet al. (2004), the frost damage for the vegetative stagesis estimated with the frost risk map of November  7  8 Cat screen height. The frost risk during the flowering andbud formation stage is estimated with the   3  8 C frost S. Geerts et al./Agricultural and Forest Meteorology 139 (2006) 399–412  403
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!